Why don't large companies use Ada?

Why don't large companies use Ada?

[Jamie Jamison]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1000 bytes --]

	I am taking an introductory computer science class at the UW and 
we are learning the Ada language. So far it's pretty nice. The language 
seems to lend itself to easy readability, the fact that the language is 
case insensitive is also a nice factor and the language is incredibly 
powerful. We're in the second quarter of this class and we're already 
writing ADT's and generic functions. Ada seems to be pretty neat, strong 
typing, bounds checking and generic functions are all nice language 
features, and from what I've seen Ada is a lot better than C. So why is 
it that the real world, such�as it is, programs in C. Sure, C compilers 
are cheaper, but it seems to me that the labor that you put into writing 
and maintaining the programs is the real cost, and I know that, at least 
for me, writing something in C takes a lot longer than writing something 
in Ada. So why hasn't Ada caught on? Why aren't people developing 
applications for it?


Jamie Jamison
niteowl@u.washington.edu

Re: Why don't large companies use Ada?

[R. William Beckwith]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 3499 bytes --]

Jamie Jamison (niteowl@u.washington.edu) wrote:
: 	I am taking an introductory computer science class at the UW and 
: we are learning the Ada language. So far it's pretty nice.  The language
: seems to lend itself to easy readability, the fact that the language is 
: case insensitive is also a nice factor and the language is incredibly 
: powerful. We're in the second quarter of this class and we're already 
: writing ADT's and generic functions. Ada seems to be pretty neat, strong 
: typing, bounds checking and generic functions are all nice language 
: features, and from what I've seen Ada is a lot better than C.

Yes, Ada is a wonderful language.

: So why is 
: it that the real world, such�as it is, programs in C. Sure, C compilers 
: are cheaper, but it seems to me that the labor that you put into writing 
: and maintaining the programs is the real cost, and I know that, at least 
: for me, writing something in C takes a lot longer than writing something 
: in Ada. So why hasn't Ada caught on? Why aren't people developing 
: applications for it?

You've directed your question to comp.lang.ada.  We love the language.
I'm cross-posting this to several groups for a broader range of opinions.

Most large, commercial customers use Ada in the embedded and safety
critical world.  (See the safety-critical survey posting.)  However,
I think you're addressing the usage of Ada for MIS.

Most who read this group recognize certain structural factors that have
impeded the use of Ada in MIS but generally are wondering when large,
commercial companies with get the idea that Ada is great for MIS systems.

IMHO, the structural factors include (in order):

    1.	operating systems and windowing software were written in C

	 MS-Windows and X/Xt/Motif are written in C.  Thus, it was
	 easier to write software in C the environment.  In addition,
	 Ada 83's built in multi-threading wreaked havoc on the
	 non-reentrant O/S and windowing software.
	 
	 The O/S's and windowing systems are finally catching up
	 to Ada's multi-threading capabilities. In addition,
	 Ada 9X's improved non-Ada interfacing features and the
	 advent of Fresco for Ada 9X Win NT and X11R6 windowing
	 software and CORBA for Ada 9X should remove these problems.

    2.	high price of Ada compilers compared to C/C++/BASIC products

	 I agree with your assessment of the real cost of software,
	 but most companies must realize their return on investment
	 in the first year or two to get project funding.  Thus,
	 they won't pay four times the price of a C++ compiler for
	 an Ada compiler.
	 
	 Fortunately, Ada compiler vendors are pricing their new
	 Ada 9X products competitively with C++ products.  In
	 addition, the free GNAT compiler allows one to learn the
	 language without any up front cost.

    3.	lack of flexibility in the first version of Ada (Ada 83)

	 The lack of program pointers, polymorphism, and inheritance
	 precluded the introduction of extensible tools in Ada 83.
	 Ada 83 was designed for ultimate reliability and
	 maintainability only.
	 
	 Ada 9X adds incredible flexibility.  Thus, Ada 9X addresses
	 these requirements and much, much more. 

None of these reasons exist anymore.  Let's see how large
corporations react in the next couple of years.

... Bill

-- 
e-mail: Bill.Beckwith@ois.com       |    Team Ada
Objective Interface Systems, Inc.   | dist, full O-O
1895 Preston White Drive, Suite 250 | multithreading
Reston, VA  22091-5448  U.S.A.      |    built in

Re: Why don't large companies use Ada?

[Jeff Reinholz]

Spending a couple years with ada environments in the defense industry
I can add my two cents worth as to how ada go a lousy reputation. One becuase
of large object code and ada compiler companies inability to stick to
a standard. Secondly because there are a vast number of C programmers and
a comparitivly small handful of ada programmers to maintain code. Third that
same readablility of code creates a over developed syntax that can be 
cumbersome in many applications. Forth because of object oriented 
requirements that ada does not match up with(some say C++ is more object 
oriented than ada) Similarities with Modulus-2(just kidding). .) 
I don't think it will ever really gain in popularity, more than it has. 
Developers are pushing toward a system with a more pure object oriented
environment than ada can offer. I.E. Smalltalk. Which is where I would
put money the industry will move. Not necessarly Smalltalk but some
dirivative. I am certain to catch hell for this from some ada devote's,
but for what ada was supposed to do, no one can deny it got off to a
bad start.

Re: Why don't large companies use Ada?

[R. William Beckwith]

: Forth because of object oriented 
: requirements that ada does not match up with(some say C++ is more object 
: oriented than ada) Similarities with Modulus-2(just kidding). .) 
: I don't think it will ever really gain in popularity, more than it has. 
: Developers are pushing toward a system with a more pure object oriented
: environment than ada can offer. I.E. Smalltalk. Which is where I would
: put money the industry will move. Not necessarly Smalltalk but some
: dirivative. I am certain to catch hell for this from some ada devote's,
: but for what ada was supposed to do, no one can deny it got off to a
: bad start.

Jeff Reinholz (reinholz@SG0D12.sig01) wrote:
: Spending a couple years with ada environments in the defense industry
: I can add my two cents worth as to how ada go a lousy reputation. One becuase
: of large object code and ada compiler companies inability to stick to
: a standard.

In my experience the Ada 83 compilers are much closer to the Ada standard
than C compilers are to the ANSI C standard.  Smalltalk and C++ have no
standard yet.

: Secondly because there are a vast number of C programmers and
: a comparitivly small handful of ada programmers to maintain code.

The defense cutbacks have improved the availability of good Ada programmers
in the U.S.  ;-)

I'll bet if three equally bright C programmers start learning
respectively Ada 9X, C++, and Smalltalk-XX, that the one learning
Ada 9X will produce working (non-leaking, robust, etc.) software
sooner than the C++ programmer and about par with the Smalltalk
programmer.  My hunch is that the Smalltalk programmer will build
smaller (working) systems faster and the Ada 9X programmer will
build larger (working) systems faster.  (I have no evidence
whatsoever to support this claim.  Just my intuition and past
discussions with large Ada, C++, and Smalltalk shops.  :-)

: Third that
: same readablility of code creates a over developed syntax that can be 
: cumbersome in many applications.

Examples?  I've never heard this one.  Most find Ada's syntax elegant
and clean.  Smalltalk is a least consistent.  C++ occasionally reminds
me of my APL days (not always a bad thing).

: Forth because of object oriented 
: requirements that ada does not match up with(some say C++ is more object 
: oriented than ada) Similarities with Modulus-2(just kidding). .) 

I guess you quit using Ada before Ada 9X was out.  There is a GNU version
of Ada 9X in cs.nyu.edu::pub/gnat.  It is not a complete compiler yet
but it is helpful when learning the new language.

: I don't think it will ever really gain in popularity, more than it has. 
: Developers are pushing toward a system with a more pure object oriented
: environment than ada can offer. I.E. Smalltalk. Which is where I would
: put money the industry will move. Not necessarly Smalltalk but some
: dirivative.

Ada 9X _is_ a pure, strongly-typed, object-oriented language.
IMHO, Ada 9x is not a pure as Smalltalk but more pure than C++.
BTW, you can use very weak typing in your class hierarchy with
Ada 9X (if you are willing to accept the Ada equivalent of
`message not understood').

Interesting to note, ParcPlace is working on adding strong typing
features to VisualWorks.

: I am certain to catch hell for this from some ada devote's,
: but for what ada was supposed to do, no one can deny it got off to a
: bad start.

No hell.  You have a right to your opinion.

I might suggest that Ada did quite well considering the popularity of
Ada outside of the U.S. and the good success of Ada within the U.S.
Dept of Defense.

... Bill

-- 
e-mail: Bill.Beckwith@ois.com       |    Team Ada
Objective Interface Systems, Inc.   | dist, full O-O
1895 Preston White Drive, Suite 250 | multithreading
Reston, VA  22091-5448  U.S.A.      |    built in

Re: Why don't large companies use Ada?

[Christopher K. Krebs]

Jeff Reinholz (reinholz@SG0D12.sig01) wrote:
: Spending a couple years with ada environments in the defense industry
: I can add my two cents worth as to how ada go a lousy reputation. One becuase
: of large object code and ada compiler companies inability to stick to
: a standard. Secondly because there are a vast number of C programmers and
: a comparitivly small handful of ada programmers to maintain code. Third that
: same readablility of code creates a over developed syntax that can be 
: cumbersome in many applications. Forth because of object oriented 
: requirements that ada does not match up with(some say C++ is more object 
: oriented than ada) Similarities with Modulus-2(just kidding). .) 
: I don't think it will ever really gain in popularity, more than it has. 
: Developers are pushing toward a system with a more pure object oriented
: environment than ada can offer. I.E. Smalltalk. Which is where I would
: put money the industry will move. Not necessarly Smalltalk but some
: dirivative. I am certain to catch hell for this from some ada devote's,
: but for what ada was supposed to do, no one can deny it got off to a
: bad start.
My big bitch with Ada is that it has historically bad interfaces to 
commercial libraries for such things as Unix, Motif, Sybase etc.  We
won't even talk about how bad Windows bindings are.  So, for developement
using these types of products, Ada stinks and the tools to support developemtn
with these products (if they exist) are extremely expensive.

For Whats its worth
Chris Krebs

Re: Why don't large companies use Ada?

[Jean D. Ichbiah]

ckrebs@s-cwis.unomaha.edu (Christopher K. Krebs) writes:

>My big bitch with Ada is that it has historically bad interfaces to 
>commercial libraries for such things as Unix, Motif, Sybase etc.  We
>won't even talk about how bad Windows bindings are.  So, for developement
>using these types of products, Ada[...] are extremely expensive.

You have a point although you should realize that this has more to
do with implementations than with the language.

Unfortunately, Ada vendors have had to  divert too many ressources 
to satisfy rigid validation requirements that had little to do with what
you need for the kind of applications you are mentioning: the real ones. 
More recently, the language revision  suffered  from the influence  of
pressure  groups (such  as Artwg) and this led to an overweight definition
which will again tax severely vendors without - in my opinion -
corresponding user benefits.

On a more positive side, the object-oriented features of 9X offer some
hope: see the article by Tom Quiggle - Re: SGI inheriting C++ classes -
later in this conference.

This may be the best  chance for Ada to survive: One one hand, there
is no way the Ada vendors could do the kind of investments that has been
done  in class frameworks developed around C++, Borland Pascal, or the 
forthcoming Delphi.  My guess is that we are  talking of several hundred  of  
man-years and it is clearly out of question to duplicate it. (I do share 
your opinion about current Windows bindings and the fact that Ada
development  for Windows is consequently overly expensive.)

But this approach of inheriting these frameworks at moderate cost,
offers a chance to put Ada back into the mainstream  applications.

Jean D. Ichbiah

Re: Why don't large companies use Ada?

[R. William Beckwith]

Jean D. Ichbiah (ichbiah@jdi.tiac.net) wrote:

: On a more positive side, the object-oriented features of 9X offer some
: hope: see the article by Tom Quiggle - Re: SGI inheriting C++ classes -
: later in this conference.

: This may be the best  chance for Ada to survive: One one hand, there
: is no way the Ada vendors could do the kind of investments that has been
: done  in class frameworks developed around C++, Borland Pascal, or the 
: forthcoming Delphi.  My guess is that we are  talking of several hundred  of  
: man-years and it is clearly out of question to duplicate it. (I do share 
: your opinion about current Windows bindings and the fact that Ada
: development  for Windows is consequently overly expensive.)

: But this approach of inheriting these frameworks at moderate cost,
: offers a chance to put Ada back into the mainstream  applications.

We at OIS have debated on whether to interface to existing C++
libraries or rewrite them several times.

Our windows version of Screen Machine was built above a C++ class
library.  This caused considerable grief in terms of interfacing
complexity (we hand-built the interface code to do what the GNAT
C++ interface does), memory management (earlier versions of the
C++ libraries leaked memory), and design clarity.

More recently we decided to base our future Ada GUI products on
the X11R6/Fresco system.  Thus, we faced the decision to reuse
or rewrite again.  This time we chose the path of rewriting the
C++ in Ada 9X.

We are _very_ happy with our decision.  Since Fresco is based on
CORBA the Fresco interfaces are defined in IDL.  Via a nice clean
mapping from IDL to Ada 9X, we could build cleaner, simpler Ada 9X
implementations of the Fresco graphical classes.

C++ has many arbitrary complexities that would have precluded a
straightforward Ada interface.  It just doesn't seem right to
make an Ada programmer pass a null object pointer as an argument
to a program that really should have an `out' parameter (one small
example).  Rewriting the C++ code in Ada 9X has allowed us to
remove a considerable amount of unnecessary C++ code and
complexity.

We felt that the GUI toolkit is so pervasive in an application
that the toolkit needs an especially clean API.  We think the
new Fresco Ada 9X interface achieves this goal because of the
rewrite.

It would be nice to see other domains dispense with bindings
and build directly in Ada 9X.  The code can sometimes be
smaller than the binding code would be.  (BTW, if C++
programmers want to reuse Ada 9X Fresco widgets then they just
generate the C++ client interfaces with a C++ IDL compiler.)

The Smalltalk community has chosen to write their graphics toolkit
in Smalltalk.  They don't use Motif widgets.  I think the
results are a very tight integration between the user interface
graphics and the native language.  They have defined their own
framework.

There are certainly many domains where the cost of rewriting can't
be justified to achieve a superior API.  It's nice to know that
the C++ code can be reused.  However, if you are going to reuse
the code frequently I suggest applying the power of Ada 9X
directly to the problem via a rewrite.  Otherwise, the C++ code
you reuse may be your biggest detriment.

... Bill

Re: Why don't large companies use Ada?

[Robert Dewar]

Following Bill's comments on rewriting C++ interfaces (we really *should*
change the title of this thread, it is long ago obsolete, and was never
reasonable, since it presumes a false fact :-)

Indeed, rewrites may be appropriate for important standard interfaces, but
realistically, it is inconceivable to rewrite all interfaces. Tom Quiggle
got the Puck program operating in four days with the reuse approach. 
Rewriting Inventor in Ada 9X would be, I would guestimate, a several
person-year project.

One other approach here is to put the energy in at a slightly different
level. Use a tool to generate a thin binding, and then build a thick
binding with the desired abstractions above it. This has the advantage of
presenting a more appropriate interface without rewriting the C++ class
definitions, and may be the appropriate approach in many situations.

Re: Why don't large companies use Ada?

[Robert Dewar]

Note also that Bill made his decision to rewrite in Ada 9X long before we
completed or even designed the GNAT C++ interface. It is not possible to
duplicate this interface manually, because among other things, it depends
on GNAT using exactly the same dispatch table structure as C++, which is
something we have only recently achieved.

In addition, neither the C++ interface facilities in GNAT, nor the SGI
binding tool that takes advantage of them is complete, so Tom's experience
is only a hint of what will eventually be achieved.

Nevertheless, I agree with Bill that in some situations it definitely WILL
make sense to rewrite interfaces in Ada 9X.

Indeed, I am much less concerned about bindings to standard interfaces like
CORBA, X, Motif etc than I am about situations like SGI's Inventor. It is
the latter more specialized cases where huge interfaces exist in C and
C++ and it is vital to be able to reuse them.

Re: Why don't large companies use Ada?

[Cyrille Comar]

>>>>> "Robert" == Robert Dewar <dewar@cs.nyu.edu> writes:
In article <3ao0jj$jhg@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:


    Robert> Note also that Bill made his decision to rewrite in Ada 9X long
    Robert> before we completed or even designed the GNAT C++ interface. It is
    Robert> not possible to duplicate this interface manually, because among
    Robert> other things, it depends on GNAT using exactly the same dispatch
    Robert> table structure as C++, which is something we have only recently
    Robert> achieved.

Robert, 
 Let me give some more technical details on this subject, GNAT doesn't use the
same dispatch table format than c++ for all tagged types but only for tagged
types matching c++ classes or deriving from them. It doesn't matter to use a
completely different format for unrelated tagged types and this is what GNAT
does by default.  This is not possible to adopt the c++ dispatch table format
unconditionnally for many reasons, the main one being that there not ONE format
but one for each different c++ compiler.  There is one case where it is useful
to adopt the c++ Dispatch Table format for regular tagged types, this is when
you want to derive a c++ class from an Ada tagged type: this is possible and
easy to change GNAT in order to match the format of a particular c++ compiler,
it is not even necessary to change the compiler itself but only one file in the
run-time thank's to the Dispatch Table abstraction that is defined in the
private part of Ada.Tags

--
------------------------------------------------------------------------
Cyrille Comar,                                  E-mail: comar@cs.nyu.edu
Gnat Project                                    US phone: (212) 998-3489

Re: Why don't large companies use Ada?

[R. William Beckwith]

Robert Dewar (dewar@cs.nyu.edu) wrote:
: Note also that Bill made his decision to rewrite in Ada 9X long before we
: completed or even designed the GNAT C++ interface. It is not possible to
: duplicate this interface manually, because among other things, it depends
: on GNAT using exactly the same dispatch table structure as C++, which is
: something we have only recently achieved.

We were going to use a code generator to create the code to reconcile
the different dispatching tables.  It wouldn't have been any more work
to create C++ bindings that the GNAT approach.  However, both approaches
inherit the C++ memory management problems.

: Indeed, I am much less concerned about bindings to standard interfaces like
: CORBA, X, Motif etc than I am about situations like SGI's Inventor. It is
: the latter more specialized cases where huge interfaces exist in C and
: C++ and it is vital to be able to reuse them.

Agreed.  Large bodies of highly specialized code that will be used by the
few rather than the many should not be rewritten.  Small and large bodies
of code that everyone will use (GUIs, DBMS clients, CORBA interfaces, etc.)
are the interfaces we want perfected. (in Ada 9X ;-)

... Bill

Re: Why don't large companies use Ada?

[Jean D. Ichbiah]

In article <CzK17K.Ino@ois.com> beckwb@ois.com (R. William Beckwith) writes:
>From: beckwb@ois.com (R. William Beckwith)

>There are certainly many domains where the cost of rewriting can't
>be justified to achieve a superior API.  It's nice to know that
>the C++ code can be reused.  However, if you are going to reuse
>the code frequently I suggest applying the power of Ada 9X
>directly to the problem via a rewrite.  Otherwise, the C++ code
>you reuse may be your biggest detriment.

Can you seriously talk of rewriting frameworks of classes as encompassing
as Borland's OWL, Microsoft Foundation classes, or Zinc? Does not make much
sense to me and goes against the most important lesson we have learnt in
the past twenty years:  Reuse, Reuse, Reuse, ...

Not only the overwhelming difficulty of rewriting the code.  But you will 
have to test it, certify it, and ... maintain it.  And if you have not been
discouraged by this mountain of effort, you will have to produce user
documentation of comparable quality - in paper an on-line - to that 
existing for these frameworks.

See you in a few hundred man-years.

Jean D. Ichbiah

Re: Why don't large companies use Ada?

[David Weller]

In article <ichbiah.55.2ECFDDC1@jdi.tiac.net>,
Jean D. Ichbiah <ichbiah@jdi.tiac.net> wrote:
>In article <CzK17K.Ino@ois.com> beckwb@ois.com (R. William Beckwith) writes:
>>From: beckwb@ois.com (R. William Beckwith)
>
>>There are certainly many domains where the cost of rewriting can't
>>be justified to achieve a superior API.  It's nice to know that
>>the C++ code can be reused.  However, if you are going to reuse
>>the code frequently I suggest applying the power of Ada 9X
>>directly to the problem via a rewrite.  Otherwise, the C++ code
>>you reuse may be your biggest detriment.
>
>Can you seriously talk of rewriting frameworks of classes as encompassing
>as Borland's OWL, Microsoft Foundation classes, or Zinc? Does not make much
>sense to me and goes against the most important lesson we have learnt in
>the past twenty years:  Reuse, Reuse, Reuse, ...
>
>Not only the overwhelming difficulty of rewriting the code.  But you will 
>have to test it, certify it, and ... maintain it.  And if you have not been
>discouraged by this mountain of effort, you will have to produce user
>documentation of comparable quality - in paper an on-line - to that 
>existing for these frameworks.
>
>See you in a few hundred man-years.
>

Jean, that's not fair to Bill.  It has been clear from his anecdotes
that he _is_ concerned with costs across the "life cycle".  You may
be pounding the pulpit about reuse, but the reality is that there's
also a cost to reuse software.  It's only when we wisely compare the
respective costs (emphasis on "wisely", and I think Bill very much
qualifies) that we can make reasonable decisions.  Admittedly the
heuristics for such comparisons are somewhat thin right now, but many
experienced developers (those that have not fossilized into a
particular school of thought) can make a pretty good estimate.


-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
      Just another dull, stodgy, non-creative developer who uses Ada.

Re: Why don't large companies use Ada?

[Jean D. Ichbiah]

 dweller@Starbase.NeoSoft.COM (David Weller) writes:

>>Can you seriously talk of rewriting frameworks of classes as encompassing
>>as Borland's OWL, Microsoft Foundation classes, or Zinc? Does not make much
>>sense to me and goes against the most important lesson we have learnt in
>>the past twenty years:  Reuse, Reuse, Reuse, ...

>Jean, that's not fair to Bill.  It has been clear from his anecdotes
>that he _is_ concerned with costs across the "life cycle".  You may
>be pounding the pulpit about reuse, but the reality is that there's
>also a cost to reuse software.  It's only when we wisely compare the
>respective costs (emphasis on "wisely", and I think Bill very much
>qualifies) that we can make reasonable decisions.  Admittedly the
>heuristics for such comparisons are somewhat thin right now, but many
>experienced developers (those that have not fossilized into a
>particular school of thought) can make a pretty good estimate.

I reiterate my disagreement.  For the above, I cannot see how
it could ever make sense to rewrite.  This completely underestimates
the fact that a significant body of literature, documentation, books,
trainers, examples,..., and other  forms of knowhow go with them 
in each case, which you will never be in a situation to replicate.



Jean D. Ichbiah

Re: Why don't large companies use Ada?

[R. William Beckwith]

Jean D. Ichbiah (ichbiah@jdi.tiac.net) wrote:

: Can you seriously talk of rewriting frameworks of classes as encompassing
: as Borland's OWL, Microsoft Foundation classes, or Zinc? Does not make much
: sense to me and goes against the most important lesson we have learnt in
: the past twenty years:  Reuse, Reuse, Reuse, ...

Reuse is for projects not for products.

Projects typically have low reuse.  COTS software companies have
very high reuse if you count their customers.  If you are writing
software that will be used in only a few projects then reuse is
critical.  However, COTS software gets used in hundreds or thousands
of projects.  Development cost is insignificant to the potential
benefit to the customer base.

Thus, most software companies partially or completely rewrite their
products every three to five years.  Look at the most visible examples:

    * Oracle did it with version 7.
    * Sybase did it with System 10.
    * Rational did it with Apex.
    * Borland did it with version 2.0 of OWL.
    * Microsoft did it with MFC before they even released the first
      version! (see OOPSLA proceedings a couple of years ago for MFC
      design team article)
    * We at OIS are on our second re-write of Screen Machine.

Certain bodies of code are so frequently reused by others that it
is imperative to make fundamental, architectural improvements to
the code to increase flexibility, reliability, and performance.

The products you mention, OWL, MFC, and Zinc, are fairly thin object
oriented covers above a non-object oriented API.  They are much better
interfaces that the underlying API's.  However, they are not even
close to the depth and power of a fully implemented class framework.

The real frameworks of the future don't use the underlying toolkits
(Taligent, OpenStep, Fresco).  They provide a much broader, more
flexible architecture that is implemented in that O-O language.

Again, I must mention the Smalltalk communities `lack of reuse'.
They don't reuse Motif or (most of) MS-Windows.  (I see everything
through GUI glasses ;-)

: Not only the overwhelming difficulty of rewriting the code.  But you will 
: have to test it, certify it, and ... maintain it.  And if you have not been
: discouraged by this mountain of effort, you will have to produce user
: documentation of comparable quality - in paper an on-line - to that 
: existing for these frameworks.

So I guess you believe that David Weller should be writing bindings
to the C++ Booch components instead of re-writing them?  Uggh.
Makes me shudder at the thought.

We have a better language than the developers of the C++ frameworks did.
We will end up with better frameworks by using the languages power, not
by binding to other frameworks.

: See you in a few hundred man-years.

I'd like our customers to think that it takes us `a few hundred man-years'
to write our products.  But in practice, the development, testing, and 
maintenance of bindings closely rivals the re-implementation of a
framework using more native approaches.  And the bindings can't even
compare to the functionality of the pure approach.

: Jean D. Ichbiah

... Bill Beckwith

Re: Why don't large companies use Ada?

[Robert Dewar]

First, I don't think that there is as much disagreement between Bill and
Jean as appears on the surface. Bill is not saying that everything should
be reengineered, and Jean is not saying that nothing should be reengineered.

I do think Bill goes much too far when he says that reuse is for projects
not products. Let's take the SGI Inventor example. If SGI is to succeed in
promoting Ada 9X, they have to provide usable interfaces to their primary
capital advantage, which is a huge library of impressive graphics stuff.

If the Ada enthusiasts inside SGI told management that this entails a lot
of reengineering, it is simply a nail in Ada's coffin in terms of being
accepted as a feasible alternative. SGI needs to be able to provide
productized versions of these interfaces, which are heavily object based,
and require the capbility of extending existing classes, with absolutely
minimal effort. Reuse is *the* key to seeing Ada 9X as a feasible 
technology in this environment.

I am not saying that everything can or should be done this way, but the
important point is that in the past we have always had the necessary
tools for doing the kind of reengineering that Bill is talking about, and
there are indeed examples of positive results from this (one reimplementation
of Motif for Ada turned out to be much more reliable and efficient than the
original), but missing was a reasonable technology for reuse.

What is exciting to me is that the combination of

  New features in Ada 9X promoting easier interfacing
  New features in Ada 9X to interface to (notably the OO stuff)
  Technologies for more direct binding to C++ (as implemented in GNAT,
     and hopefully similiar or superior approaches will appear in other
     Ada 9X compilers)
  AUtomatic binding generating tools of the kind demonstrated by SGI

seem to point in the direction of a practical reuse technology for Ada 9X
which was sorely lacking in the Ada 83 case. 

This does not mean that I think Bill is going in the wrong direction in
his efforts, on the contrary, I think he is on track to providing some
very powerful, and very comfortable-to-use interfacing to some important
technology. What it *does* mean is that we don't have to depend on such
re-engineering efforts for cases where at least at first it will be
impractical to follow anything other than a reuse path.

Now if Ada 9X becomes so successful that Microsoft, Borland, ... do all
their programming in Ada 9X, then presumably all these wonderful interfaces
will indeed be reengineered in Ada 9X, but till then ... :-) :-)

Re: Why don't large companies use Ada?

[Jean D. Ichbiah]

In article <CzMHDw.L0t@ois.com> beckwb@ois.com (R. William Beckwith) writes:

>Reuse is for projects not for products.

This distinction escapes me.  We were  initially discussing the fact
that to make Ada usable in application development you need access 
to framework of classes.   I maintain that rewriting does not make
sense - in an economic point of view.  The only sensible, timely,
and economic solution is  to interface existing classes and so to
reuse them.

> [...] most software companies partially or completely rewrite their
>products every three to five years.  Look at the most visible examples:

[...]
>    * Borland did it with version 2.0 of OWL.
>    * Microsoft did it with MFC before they even released the first
>      version! (see OOPSLA proceedings a couple of years ago for MFC
>      design team article)

But you are confusing two different matters. In each of the above cases
rewriting produces a version that makes the previous one Obsolete.
On the other hand, if you were to rewrite MFC or OWL just to be able to 
use it in Ada, I presume that your ambition would not be to
obsolete these.

>The products you mention, OWL, MFC, and Zinc, are fairly thin object
>oriented covers above a non-object oriented API.  They are much better
>interfaces that the underlying API's.  However, they are not even
>close to the depth and power of a fully implemented class framework.

Fine that you want something  more ambitious, but you must be aware
that none of the available Ada solution is even at the level of OWL, MFC,
or Zinc: they are at the level of the API - unusable unless you  are very rich
and prepared to spend many more man-years.

 Jean D. Ichbiah

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <ichbiah.54.2ECE3971@jdi.tiac.net>,
Jean D. Ichbiah <ichbiah@jdi.tiac.net> wrote:

>You have a point although you should realize that this has more to
>do with implementations than with the language.

Certainly.

>Unfortunately, Ada vendors have had to  divert too many ressources 
>to satisfy rigid validation requirements that had little to do with what
>you need for the kind of applications you are mentioning: the real ones. 

I cannot agree with this standard oversimplification. It was _not_
necessary for all the vendors to invest in dozens of essentially
duplicative ports, just to compete on every single possible platform.

Code generators are not free, nor were additional validation fees.
I would have expected more segmentation of the market, instead of
the blind competition that we saw. There was NO need for _every_
vendor to do a VAX/VMS, AND a Sun-3, AND an HP, and,..., and,...

In the early 80's, we were all enthusiastic about the possibility
of real competition to build the best compilers for a standard
language. But the process got away from us, I think, and we ended
up with a kind of Sorceror's Apprentice that produced, at one point
around 88-89 (I think it was) TEN compilers for VAX -> Vax and
another TEN compilers for Sun3 -> Sun3. The market - especially
given the poor attitude (IMHO) of the vendors toward a _commercial_
market at that time - could never have supported THAT much 
duplication of effort. Hence the consolidations and takeovers.

>More recently, the language revision  suffered  from the influence  of
>pressure  groups (such  as Artwg) and this led to an overweight definition
>which will again tax severely vendors without - in my opinion -
>corresponding user benefits.

Possibly, though much of the stuff that ARTWG would've demanded
should be in the annexes, right? Or do protected types add THAT
much more weight to the tasking model?

I hear conflicting stories on this issue. On the one hand, the vendors have
argued that their biggest customers are the real-time folks, which is
why they had to neglect other stuff. So why are they screaming so
loud about "demands from pressure groups"? I thought the name of the
game was "the customer is always right."

>On a more positive side, the object-oriented features of 9X offer some
>hope: see the article by Tom Quiggle - Re: SGI inheriting C++ classes -
>later in this conference.

Yes, I agree. Unfortunately, the traditional Ada vendors have not
done much with this sort of thing, unless they have some really good
stuff in their "skunk works". That SGI has done so much, so fast,
for (apparently) such a low cost (over and above GNAT), tells me that
with a bit of imagination, Alsys and Rational could also do it.

>This may be the best  chance for Ada to survive: One one hand, there
>is no way the Ada vendors could do the kind of investments that has been
>done  in class frameworks developed around C++, Borland Pascal, or the 
>forthcoming Delphi.  My guess is that we are  talking of several hundred  of  
>man-years and it is clearly out of question to duplicate it. (I do share 
>your opinion about current Windows bindings and the fact that Ada
>development  for Windows is consequently overly expensive.)

I have said a number of times in this conference that I think the
vendors could do quite well by concentrating on "value-added"
things like bindings, runtimes, and tools, instead of continuing
indefinitely with a head-to-head competition for compilers on
every platform. If they were to "standardize" on GNAT as the core
compiler technology, and everyone contributed something to GNAT's
completion and stabilization on all the usual platforms, some of
the bucks now going into over-competition might be better invested
in proprietary added value on top of the free baseline compiler.

Undoubtedly someone will jump in and tell me I need some business
courses to understand why this idea is not feasible...:-)

>But this approach of inheriting these frameworks at moderate cost,
>offers a chance to put Ada back into the mainstream  applications.

Exactly. But are the traditional Ada vendors working on this?
Surely the quality of the Ada Windows add-ons leaves much to be
desired. They need a paradigm shift and a more commercial mentality.
And if they think GNAT has handed them a lemon, they need the
imagination to figure out how to make lemonade from it.

Nice to see you on the net, Jean; hope to hear from you soon.
I'd really appreciate hearing from you on what GNAT's role should
be in all this.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Non illegitimi carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[David Emery]

>My big bitch with Ada is that it has historically bad interfaces to 
>commercial libraries for such things as Unix, Motif, Sybase etc.  

I've been doing Ada bindings to C for over a decade now.  When I
started, there were some implementation problems.  But, even now, the
biggest problem is lack of understanding on how to do language
bindings.  

In short, there are two approaches:
	"thin" (quick and dirty :-)
	"thick" (do it right :-)
In order to do a good job of defining an Ada binding (or a binding
from language X to language Y, for all languages X and Y, including C
and C++), you have to understand the semantics of the interface.  Then
you have to present these semantics in the appropriate style using Ada
(or language X).  Once you understand the semantics, the presentation
is easy.  

The difficulty comes in understanding the basic semantics.  This is a
place where skill, experience, and the quality of the interface all
come into play.  If you're given a bad/incomplete/inconsistent/poorly
designed C interface, it's more difficult to do a good Ada binding.

Since most of the effort comes in understanding the C (or whatever)
interface, language bindings are still mostly mental effort.  This is
why 'automatic generation' of language bindings aren't a good idea.
Instead of getting something that has been thought through, you just
get C (or whatever) with Ada syntax.  You're better off sticking with
C.  

So the bindings 'problem' is a classic engineering problem, and one
that does not have any easy solutions.  On the other hand, given a
reasonably designed C interface, I can produce a good ("thick") Ada
binding in 2-3 weeks.  Our experience is that a good Ada binding
produces substantial payback with even a small number of users.  In
particular, a good Ada binding can prevent or detect errors before
they propogate into the underlying implemenation, where it is much
harder to detect and debug errors.  

It's unfortunate that the Ada community, and the vendor community,
haven't worked together to solve this problem, which, after all, is
not that difficult.  Too many people have taken the approach of "I'll
do it when I get paid to", which has not helped matters.

				dave
--
--The preceeding opinions do not necessarily reflect the opinions of
--The MITRE Corporation or its sponsors. 
-- "A good plan violently executed -NOW- is better than a perfect plan
--  next week"                                      George Patton
-- "Any damn fool can write a plan.  It's the execution that gets you
--  all screwed up"                              James Hollingsworth
-------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Robert Temple]

Well, Much of the Software being put on the Boeing 777 is Ada.  Although 
many of the advatages of Ada are not being used, such as strict type 
checking and exceptions.

-Bobo

==============================================================
GO BULLS!
Robert Temple   Home (602)867-7280   Work (602)436-5274

Re: Why don't large companies use Ada?

[Doug Robertson]

My question is - why do Universities teach Ada - sometimes as
the main language on courses ?
-- 
Doug

STOP!!! (WAS: Why don't large companies use Ada?)

[Q Vincent Yin]

In <784995260snz@rippleco.demon.co.uk> Doug@rippleco.demon.co.uk (Doug Robertson) writes:

>My question is - why do Universities teach Ada - sometimes as
>the main language on courses ?
>-- 
>Doug

Same thing for Pascal.  But the point is: I see that the article is
cross posted to: 
comp.lang.ada,comp.lang.c,comp.programming,comp.lang.c++,comp.object,comp.databases.sybase,comp.databases.oracle,comp.client-server

Could you guys stop cross-posting to comp.databases.sybase.  We Sybase
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
sa/programmers don't care about your opinion of Ada!!!!!!!
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

-- 

Q Vincent Yin                   |       Repeat
umyin@mctrf.mb.ca               |          delete(next_bug);
                                |       Until 0 = 1;

Re: STOP!!! (WAS: Why don't large companies use Ada?)

[Dean Souleles]

In article <3ae1li$jsc@canopus.cc.umanitoba.ca>,
Q Vincent Yin <umyin@raphael.mctrf.mb.ca> wrote:
>In <784995260snz@rippleco.demon.co.uk> Doug@rippleco.demon.co.uk (Doug Robertson) writes:
>
>>My question is - why do Universities teach Ada - sometimes as
>>the main language on courses ?
>>-- 
>>Doug
>
>Same thing for Pascal.  But the point is: I see that the article is
>cross posted to: 
>comp.lang.ada,comp.lang.c,comp.programming,comp.lang.c++,comp.object,comp.databases.sybase,comp.databases.oracle,comp.client-server
>
>Could you guys stop cross-posting to comp.databases.sybase.  We Sybase
>^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>sa/programmers don't care about your opinion of Ada!!!!!!!
>^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>

Except for those of us who write our Sybase access code in Ada.

-- 
Dean Souleles                  dsoulele@uccs.jpl.nasa.gov
Jet Propulsion Laboratory 
4800 Oak Grove Drive M/S 514-100
Pasadena, CA 91109

Re: STOP!!! (WAS: Why don't large companies use Ada?)

[Norman H. Cohen]

In article <3ae1li$jsc@canopus.cc.umanitoba.ca>, umyin@raphael.mctrf.mb.ca
(Q Vincent Yin) writes: 

|> Could you guys stop cross-posting to comp.databases.sybase.  We Sybase
|> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
|> sa/programmers don't care about your opinion of Ada!!!!!!!
|> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

This reasonable plea by several readers of other newsgroups has been so
widely ignored that I can only conclude that many people don't know how
to ascertain and control the set of newsgroups to which their posts are
targeted.  With xrn, at least, it is quite easy.  When I click on the
Followup button, an editing session pops up in which the first few lines
of the file contain header information that is read by your news server
when you save the file and exit the editor.  For my current posting, the
second line of the file orginally read: 

 Newsgroups: comp.lang.ada,comp.lang.c,comp.programming,comp.lang.c++,comp.object,comp.databases.oracle,comp.client-server

I edited this line to read

 Newsgroups: comp.lang.ada

so now this posting will go only to comp.lang.ada instead of to all the
newsgroups originally listed.   That's all there is to it.
This act of common courtesy takes only a second of your time.

I realize that this information is well-known to many of you, but a
handful of gratuitous cross-postings can create a lot of resentment.

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[Robert Dewar]

Why do universities teach Ada?                        

presumably because they feel it is a good vehicle for teaching basic
programming and software engineering concepts. Note that this has nothing
to do with whether it is or is not used. For years, Pascal has been taught
in universities because it is an effective teaching tool, not because it
is widely used in real programming projects.

The argument that language X should be taught because X is used widely is
what kept US universities teaching FOrtran when the rest of the world had
moved on (see the survey in the British Journal of Computing, around 1971).
It is a badly flawed argument which does students tremendous disservice
and one that in the last 20 years has largely been rejected (how many
major universities have used COBOL to teach the beginning course, because
of the quite correct observation that it was -- and probably *still* is, the
most widely used programming language)?

It is a shame to see people use this decrepit line of reasoning to justify
teaching C in beginning courses. I guess part of the reason for this is
that we are now seeing the graduate students who learned C hacking in
universities under the illusion that it is programming in positions as
university professors eager to teach C under the illusion that they are
teaching programming :-)

Re: Why don't large companies use Ada?

[mat]

In article <3ae5n0$o0e@gnat.cs.nyu.edu>, dewar@cs.nyu.edu (Robert Dewar) writes:
> Why do universities teach Ada?                        
 
> It is a shame to see people use this decrepit line of reasoning to justify
> teaching C in beginning courses.  ...

As I recall, we differed on this point once before, many years ago, you
and I.  I've changed my opinion somewhat.

It's a shame that we teach people a programming language before we teach
them how to think about problems.

I remember my undergrad engineering education.  I spent two years on basic
math and science and then on basic electrical theory (circuit theory) before
I was taught any electronics.  And _then_ I was allowed to work in the lab,
closely supervised for the first few weeks to make sure I didn't destroy
anything.

Maybe we should see how this experience can be projected into software.
It's not the academics that took the first step towards making software
an engineering discipline, and heaven knows it wasn't the engineers.  It
was the business programmers and people working for them (Constantine,
Jackson, et. al.)

BTW, the most valuable single course I ever took was one that Jack Schwartz
of NYU gave on proving programs.  The most harmful single course I ever
took was fundamental algorithms--because the `fundamental' things include
bags of ad-hoc tricks, especially for searching and sorting.

But I've said that before.

The best thing I can say about Ada for the purpose espoused is that it
is infinitely better than Pascal, lacking as it does the idiot mistakes
that were written into Pascal.  The worst thing (for teaching the beginning
programmer) is that it requires its own bag of tricks to deal with some of
the peculiarities of declaring variables.

And the thing I least like about it is its excessive wordiness ;^)

I do prefer C++.
-- 
 (This man's opinions are his own.)
 From mole-end				Mark Terribile
 mat@mole-end.matawan.nj.us, Somewhere in Matawan, NJ
	(Training and consulting in C, C++, UNIX, etc.)

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <784995260snz@rippleco.demon.co.uk>,
Doug Robertson <Doug@rippleco.demon.co.uk> wrote:
>My question is - why do Universities teach Ada - sometimes as
>the main language on courses ?

Well, I could go on all night on this, but won't. Suffice to say that
most of us teaching Ada as the main (or first) language do so because 
Ada has all the pedagogical features (strong typing, abstract data types,
runtime checking) that we like in Turbo Pascal, but it's portable.
An Ada program that avoids _machine-specific_ libraries (like PC 
graphics) will compile and execute using any compiler on any platform.

That not only makes life easier for teachers and students - they can
write code on their home PC's, then compile for submission on the
university's Unix or VMS box - but it also sets an excellent example
for students on the value of strong and well-enforced standards.
IMHO, good language standards are a sign of maturity in the industry,
an end to "feature wars" such as still rage in, say, Pascal.

Further, Ada has enough richness to be a useful language for upper-
level courses such as compilers, operating systems, and (obviously)
project-oriented SE courses. At GW as elsewhere, most of us teach our
undergraduates several languages, usually including C and C++, but
we believe that as imperative languages go, Ada makes the best 
first language. Whether it is the most popular in industry is of
little relevance in selecting it as a _first_ language.

I hope this doesn't start yet another thread on first languages;
we've had about one per month in this group and are getting bored.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Non illegitimi carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <784995260snz@rippleco.demon.co.uk>,
Doug Robertson <Doug@rippleco.demon.co.uk> wrote:
>My question is - why do Universities teach Ada - sometimes as
>the main language on courses ?

This gives me an excuse to post an update to my list of universities in
which Ada is introduced in first-year courses.

--- cut here ---

Ada as a Foundation Programming Language

Michael B. Feldman
Chair, ACM-SIGAda Education Working Group
Department of Electrical Engineering and Computer Science
The George Washington University
Washington, DC 20052
(202) 994-5919 (voice) - (202) 994-0227 (fax)
mfeldman@seas.gwu.edu

November 1994

This report lists the colleges and universities known -- or at least credibly 
believed -- to have adopted Ada as a "foundation language," at CS1, CS2, 
or CS7 level in their computer science (or similar) curricula. I choose
to keep track of precisely these courses because they are taken by 
students in either first or second year, and thus early enough to serve
as a foundation upon which to build a large portion of the undergraduate 
software curriculum.

This data is, for the most part, verifiable, based on first-hand reports 
from the teachers of those courses. In a few cases, publishers' textbook 
adoption data was used. I would like to keep this list as complete and up to 
date as possible, so please let me know of any additions or corrections.

Group 1: Colleges and Universities Introducing Ada as the First Language
         Taught in an Undergraduate Computing Curriculum.

Allan Hancock College, California
Arizona State University
Auburn University
Australian Defence Force Academy, Canberra
Basque University, San Sebastian, Spain
Birmingham Southern College, Alabama
California State University, Long Beach
California Polytechnic State University, San Luis Obispo
Chalmers University of Technology, Sweden
Clemson University, South Carolina
Conservatoire National des Arts et Metiers, Paris,  France
Cranfield Institute of Technology, United Kingdom
Cypress College, California
Ecole Nationale Superieure des Telecommunications de Bretagne, France
Ecole Superieure en Sciences Informatiques, Nice, France
Edith Cowan University, Perth, Australia
Embry-Riddle Aeronautical University, Florida
Florida Institute of Technology
Fayetteville State University, North Carolina
Flinders University, Australia
The George Washington University, Washington, DC
Indiana-Purdue University, Ft. Wayne
Institut National des Sciences Appliquees, Toulouse, France
Jerusalem College of Technology, Israel
LeMoyne College, New York
Lenoir-Rhyne College, North Carolina
Loyola Marymount University, California
Marion County Technical Center, West Virginia
Marshall University, West Virginia
Montana State University
Muskingum College, Ohio
National Defense Management College at Taiwan, Republic of China
Northeast Missouri State University
Northern Arizona University
Norwich University, Vermont
Otterbein College, Ohio
Portsmouth Polytechnic, United Kingdom
Royal Melbourne Institute of Technology, Australia
Royal Military Academy, Belgium
Sacred Heart College (Connecticut)
Saint Mary College, Kansas
Sam Houston State University, Texas
San Diego Mesa College, California
Santa Clara University
Seattle University
South Bank University, United Kingdom
Southampton Institute of Higher Education, United Kingdom
Southern Arkansas University
St. Cloud State Univ., Minnesota
State University of New York at Plattsburgh
Stockton College of New Jersey
Swinburne University of Technology, Australia
Technical College of Berlin, Germany
Technical University of Madrid (School of Telecommunications), Spain
United States Air Force Academy
University of Adelaide, Australia
University of Aston, United Kingdom
University of Bradford, United Kingdom
University of Canberra, Australia
University of Dayton, Ohio
University of Glamorgan, Wales, United Kingdom
University of Lancaster, United Kingdom
University of Liverpool, United Kingdom
University of Maryland (University College)
University of Muenster, Germany
University of New Orleans
University of North Dakota
University of Paisley, United Kingdom
University of Rome at Tor Vergata, Italy
University of Salzburg, Austria
University of Sofia, Bulgaria
University of South Dakota
University of South Florida
University of Stafford, United Kingdom
University of Wales, Aberystwyth, United Kingdom
University of York, United Kingdom
Valparaiso University, Indiana
West Virginia University

Group 2: Colleges and Universities Not Introducing Ada as the First Language
         but Rather in a CS2 or Data Structures Course (or equivalent)

Briar Cliff College, Iowa
California Polytechnic State University, Pomona
California State University, Fullerton
College of West Virginia, Beckley
Daniel Webster College, New Hampshire
Davis and Elkins College, West Virginia
Ecole d'Ingenieurs de l'Etat de Vaud, Switzerland
Florida International University
Gallaudet University, Washington, DC
Georgia State University
Indiana University, New Albany
Lenoir Rhyne College, North Carolina
Mesa State College, Colorado
Monterey Peninsula College, California
Mount Mercy College, Iowa
Murray State University, Kentucky
National University, California
Nicholls State University, Louisiana
Northern Arizona University
Northern Kentucky University
Oglethorpe University, Georgia
Ohio University, Athens
Pennsylvania State University, Harrisburg
Rose Hulman Institute of Technology, Indiana
Southwest Baptist College, Missouri
Shippensburg University, Pennsylvania
State University of New York at Fredonia
Swiss Federal Institute of Technology, Lausanne
University of Alaska, Fairbanks
University of Brighton, United Kingdom
University of Geneva, Switzerland
University of Missouri, Columbia
University of Richmond, Virginia
University of Scranton, Pennsylvania
University of Seville, Spain
University of Texas, Austin
University of Texas, Permian Basin
Western New England College, Massachusetts

Re: Why don't large companies use Ada?

[Robert C. Lokerson]

In Article <3aehe6$cp4@felix.seas.gwu.edu> "mfeldman@seas.gwu.edu (Michael Feldman)" says:
> In article <784995260snz@rippleco.demon.co.uk>,
> Doug Robertson <Doug@rippleco.demon.co.uk> wrote:
> >My question is - why do Universities teach Ada - sometimes as
> >the main language on courses ?
> 
> This gives me an excuse to post an update to my list of universities in
> which Ada is introduced in first-year courses.
> 
 
[Stuff Deleted]
 
> 
> Group 1: Colleges and Universities Introducing Ada as the First Language
>          Taught in an Undergraduate Computing Curriculum.
> 
 
[Stuff Deleted]
 
> Lenoir-Rhyne College, North Carolina
 
[Stuff Deleted]
 
> 
> Group 2: Colleges and Universities Not Introducing Ada as the First Language
>          but Rather in a CS2 or Data Structures Course (or equivalent)
> 
 
[Stuff Deleted]
 
> Lenoir Rhyne College, North Carolina
 
[Stuff Deleted]
 
But how does Lenoir-Rhyne/Lenoir Rhyne introduce students to set-thoery?
 
     ;-)
 


  R.C. LOKERSON (BOB)      MAIL ADDRESS:      VOICE : 407.729.2257
  HARRIS MDSO/MIS/DCC/ATG     HARRIS E.S.S.   SMTP  : RLOKER01@HARRIS.COM
  BLDG 15/RM 838              P.O. BOX 37     CCMAIL: RLOKER01
  PALM BAY, FLORIDA           MELBOURNE FL    FCC   : WA3PKX
                              32902           HOME  : 407.725.8024

  (I WROTE IT, NOT MR. HARRIS)
 
When rayguns are outlawed, only outlaws will have rayguns-Natl Raygun Assc

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <19941117145014.RLOKER01@rlokerson9.ess.harris.com>,
Robert C. Lokerson <RLOKER01@HARRIS.COM> wrote:

>[Stuff Deleted]

>> Lenoir-Rhyne College, North Carolina

>[Stuff Deleted]

>> Group 2: Colleges and Universities Not Introducing Ada as the First Language
>>          but Rather in a CS2 or Data Structures Course (or equivalent)

>[Stuff Deleted]

>> Lenoir Rhyne College, North Carolina

>[Stuff Deleted]
> 
>But how does Lenoir-Rhyne/Lenoir Rhyne introduce students to set-thoery?

Oops! I forgot to delete this when they moved from Group 2 to Group 1.
Whaddya want for the price? :-)

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Non illegitimi carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[James Hopper]

In article <784995260snz@rippleco.demon.co.uk> Doug Robertson,
Doug@rippleco.demon.co.uk writes:
>My question is - why do Universities teach Ada - sometimes as
>the main language on courses ?
>--
>Doug


Well i am sure people like Mike Feldman will answer this from the 
point of view of the educational community, i want to take a shot
at it from an industry perspective. 

one case in point is a couple of years ago i lost my best Ada
engineer to NCR (They are headqaurted here in dayton) to do
C/C++ commercial development.  I asked her what NCR thought of
the fact that she was an Ada engineer with NO C/C++ experience
at all.  She said that they told her they considered her Ada
experience a big plus as they associated Ada with Software
Engineering.  They were more interested in hiring people who
understood mature development processes than in hiring coders
who could hack out C.  This feedback has been passed back
a lot from my contacts in the commercial world. 

So the answer to your question from an industrial perspective
is because it teaches you good Software Enginnering practices
that are far harder to teach than simple language syntax!

best jim

Jim Hopper
Chief Tech Advisor
Systems and Software Technology Division
SAIC

Re: Why don't large companies use Ada?

[Carlos Perez]

James Hopper (hopperj@dayton.saic.com) wrote:
: They were more interested in hiring people who
: understood mature development processes than in hiring coders
: who could hack out C.  This feedback has been passed back
: a lot from my contacts in the commercial world. 

My experience has been mixed.  Most good software development organization
understand that having Ada experience is a positive but many other
organizations think that Ada is some specialized language like Jovial.
Also Ada is associated with the "US military" and so "can't be appropriate"
for commercial use.  I hope someday that Ada will be regarded like 
TCP/IP and recognized as a military invention useful to the commercial world
in a broad way.

I ended up having to put the following paragraph in my resume to "educate"
recruiters on Ada: 

  (Ada is a modern software engineering language supporting object-oriented
  principles such as information hiding, type inheritance, strong typing,
  templates, exception handling, and process concurrency) 

I didn't like using words like "templates" but my experience has been
that most managers are unfamiliar with Ada, much less generics.

: best jim

regards,
-- Carlos

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3b5ajt$n86@news-2.csn.net>, Carlos Perez <perez@oldcolo.com> wrote:

>Also Ada is associated with the "US military" and so "can't be appropriate"
>for commercial use.  I hope someday that Ada will be regarded like 
>TCP/IP and recognized as a military invention useful to the commercial world
>in a broad way.

If you run into people who think that anything developed by the "military"
- by which I take it you mean the Department of Defense - could not be
appropriate for commercial use, please remind them that not only was
TCP/IP a "military invention" (that is, _funded_ by DoD), but so were
Berkeley Unix, X Windows, and a raft of other software things over the
years. 

People's memories are very selective as to who funded what. Also,
there is monumental ignorance of the history of things we take for
granted. Like it or not, defense "spinoffs" are nearly everywhere in
technological life. That includes Ada, which needs to be evaluated
on its technical merits, not on ignorant "perceptions".

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Tarjei Jensen]

In article <3b5h8q$de0@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:


> People's memories are very selective as to who funded what. Also,
> there is monumental ignorance of the history of things we take for
> granted. Like it or not, defense "spinoffs" are nearly everywhere in
> technological life. That includes Ada, which needs to be evaluated
> on its technical merits, not on ignorant "perceptions".

Since technical merits migth not win the day we should face the realities and
use the other successful spinoffs to give Ada a better image.

"Coming soon from the people who brought you TCP/IP {insert list of other
goodies here} the programming language both the accountant and programmers
love. Ada the language for the new generation! Be in control!"


Greetings,

 
--
// Tarjei T. Jensen - if it ain't broken, fix it anyway!
//    tarjeij@ulrik.uio.no       || +47 51 563411
//   Support you local rescue centre: GET LOST!
// Working, but not speaking for the Norwegian Hydrographic Service.

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:

>Since technical merits migth not win the day we should face the realities and
>use the other successful spinoffs to give Ada a better image.

>"Coming soon from the people who brought you TCP/IP {insert list of other
>goodies here} the programming language both the accountant and programmers
>love. Ada the language for the new generation! Be in control!"

Bingo! And let's keep spreading the stories about where Ada has really
succeeded. Make sure everyone knows about the software behind all
those CRT's in the airliner cockpit. And in the air-traffic-control system
on the ground. And (in France, anyway) in the train system that gets
you from the airport to the Riviera at 300 km/h. And the Swiss PTT's
electronic-funds-exchange system. And, and...

And all those systems are written in Ada 83...

And not only that, it'll run on a PC. And not only that, GNAT is _free_.

How about "Ada - use it when your software really has to WORK."

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Matt Kennel]

: In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
: Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:

: >Since technical merits migth not win the day we should face the realities and
: >use the other successful spinoffs to give Ada a better image.

: >"Coming soon from the people who brought you TCP/IP {insert list of other
: >goodies here} the programming language both the accountant and programmers
: >love. Ada the language for the new generation! Be in control!"

The problem is that too many people see Ada as the OSI protocols
of computer languages instead of TCP/IP.

: Mike Feldman
--
-Matt Kennel  		mbk@inls1.ucsd.edu
-Institute for Nonlinear Science, University of California, San Diego
-*** AD: Archive for nonlinear dynamics papers & programs: FTP to
-***     lyapunov.ucsd.edu, username "anonymous".

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3befcg$grh@network.ucsd.edu>,
Matt Kennel <mbk@inls1.ucsd.edu> wrote:

>: >"Coming soon from the people who brought you TCP/IP {insert list of other
>: >goodies here} the programming language both the accountant and programmers
>: >love. Ada the language for the new generation! Be in control!"

>The problem is that too many people see Ada as the OSI protocols
>of computer languages instead of TCP/IP.

Can you elaborate on this a bit? I thik I'm missing the point.

Mike Feldman

Re: Why don't large companies use Ada?

[Tarjei Jensen]

Well, I thougth that the slogan "Be in control!" would capture the essence of
Ada.

If one want to stress the safety critical aspects of Ada one could use "Ada -
when YOUR life is on the line." However I think the former is better and more
general.


Greetings,
 
 
--
// Tarjei T. Jensen - if it ain't broken, fix it anyway!
//    tarjeij@ulrik.uio.no       || +47 51 563411
//   Support you local rescue centre: GET LOST!
// Working, but not speaking for the Norwegian Hydrographic Service.

Re: Why don't large companies use Ada?

[Dave Vernest]

Michael Feldman (mfeldman@seas.gwu.edu) wrote:
: In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
: Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:

: >Since technical merits migth not win the day we should face the realities and
: >use the other successful spinoffs to give Ada a better image.

: Bingo! And let's keep spreading the stories about where Ada has really
: succeeded. Make sure everyone knows about the software behind all
: those CRT's in the airliner cockpit. And in the air-traffic-control system
: on the ground. And (in France, anyway) in the train system that gets
: you from the airport to the Riviera at 300 km/h. And the Swiss PTT's
: electronic-funds-exchange system. And, and...

Not sure if I'm violating the c.p charter or not...

With the use of C++, which supports generics and exception-
handling, is there any reason to code mission-critical systems
in Ada anymore?? Advantages / disadvantages??

Dave

             /* DISCLAIMER: MyViews != ThoseViewsOf(*UW,&COGNOS) */
This article dedicated to Canada's greatest folk artist, Stan Rogers, R.I.P.'83
    VP, Early-Starters-Yet-All-Nighters-Yet-Creamed-By-CS-Assignments Society

	"Data integrity is like virtue -- it is lost only once."
			--	Mark Terribile from "Practical C++"

Re: Why don't large companies use Ada?

[Ole-Hjalmar Kristensen TF.DT/DELAB]

In article <1994Nov29.154220.27952@cognos.com>, vernestd@cognos.COM (Dave Vernest) writes:
|> Michael Feldman (mfeldman@seas.gwu.edu) wrote:
|> : In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
|> : Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:
|> 
|> : >Since technical merits migth not win the day we should face the realities and
|> : >use the other successful spinoffs to give Ada a better image.
|> 
|> : Bingo! And let's keep spreading the stories about where Ada has really
|> : succeeded. Make sure everyone knows about the software behind all
|> : those CRT's in the airliner cockpit. And in the air-traffic-control system
|> : on the ground. And (in France, anyway) in the train system that gets
|> : you from the airport to the Riviera at 300 km/h. And the Swiss PTT's
|> : electronic-funds-exchange system. And, and...
|> 
|> Not sure if I'm violating the c.p charter or not...
|> 
|> With the use of C++, which supports generics and exception-
|> handling, is there any reason to code mission-critical systems
|> in Ada anymore?? Advantages / disadvantages??
|> 
|> Dave
|> 
|>              /* DISCLAIMER: MyViews != ThoseViewsOf(*UW,&COGNOS) */
|> This article dedicated to Canada's greatest folk artist, Stan Rogers, R.I.P.'83
|>     VP, Early-Starters-Yet-All-Nighters-Yet-Creamed-By-CS-Assignments Society
|> 
|> 	"Data integrity is like virtue -- it is lost only once."
|> 			--	Mark Terribile from "Practical C++"


Well, I think this illustrates some differences between C++ and Ada :-)
(anyone knows who is the author of this amusing piece?)

Shooting Yourself in the Foot
-----------------------------

The proliferation of modern programming languages (all of which seem to
have stolen countless features from one another) sometimes makes it
difficult to remember what language you're currently using.  This guide
is offered as a public service to help programmers who find themselves
in such dilemmas.

C: You shoot yourself in the foot.

C++: You accidentally create a dozen instances of yourself and shoot
them all in the foot.  Providing emergency medical assistance is
impossible since you can't tell which are bitwise copies and which are
just pointing at others and saying, "That's me, over there."

< Some languages deleted >


Ada: After correctly packaging your foot, you attempt to concurrently
load the gun, pull the trigger, scream and shoot yourself in the foot.
When you try, however, you discover that your foot is of the wrong
type.

Re: Why don't large companies use Ada?

[An Amoeba]

Why should they bother??

Given that most students coming out of college/University will have some
experience in programming some type of language Pascal, Modula-2/3, C
and possibly BASIC plus any others and also most colleges/Universities
provide courses in C, it is totally logical, and more important cost
saving to send/train employees on C++ courses, and thus program systems
in C++.

Why?

i) Cost. If the user has a basic knowledge of C or other programming language
then converting to C++ is relatively easy.

ii) Ada's complexity. Most new users will quickly come across all Ada's *nice*
type checking, or special constructs which require a different syntax to normal.
In other languages these are not required.

iii) Support. Virtually every computer system in the world has a C++ compiler,
and it is being updated continually. Because that is where all the money is.
Not every computer system in the world has Ada on it, plus most will not be
powerful enough to run it. Plus the cost of the respective compilers. C++ 
compilers, are cheaper than Ada ones. On all if not the majority of platforms.

iv) Libraries, Code already written and available experts. C++ has access straight
away to all the C code libraries written. Not just a few nice ones, but all
C library code ever written, one reason the famous COBOL is still around!
Lots and lots of code has already been written in C++, and also any company can
go out and get any number of C++ programmers, or even C ones and train them,
whereas there are fewer Ada programmers.

v) OS Support. All modern OS have C and C++ support for there libraries. For Ada
to access the same stuff as C and C++ someone has to write specfications for the
C and C++ code in the OS. It is not supported straight out by OS Vendors.

This is just my .005 pence worth, and feel totally free to totally disagree.

--Gareth
+-------------------------------------------------------------------------+
| Gareth Bromley            | Erotic (adj): Using a feather as a sex aid. |
| bromlegb@sun.aston.ac.uk  | Kinky (adj):  Using the whole duck.         |
+-------------------------------------------------------------------------+

Re: Why don't large companies use Ada?

[Jules]

In article <D03AL2.5xw@aston.ac.uk>,
	bromlegb@aston.ac.uk writes:
>Why should they bother??
>
>Given that most students coming out of college/University will have some
>experience in programming some type of language Pascal, Modula-2/3, C
>and possibly BASIC plus any others and also most colleges/Universities
Oh come off it, nobody teaches BASIC anymore, do they?
Oh, and how many of them are teaching Ada now? Quite a lot I believe. As
a large proportion of those graduates know Ada, doesn't it make sense to
use it?

>provide courses in C, it is totally logical, and more important cost
>saving to send/train employees on C++ courses, and thus program systems
>in C++.
>
>Why?
>
>i) Cost. If the user has a basic knowledge of C or other programming language
>then converting to C++ is relatively easy.

Converting to C++ is astoundingly difficult if you want to do it well. See the
thread on learning OOP for more details.
Not retraining someone to use a different language to the one they learnt
at university seems like the cheaper option to me.

>ii) Ada's complexity. Most new users will quickly come across all Ada's *nice*
>type checking, or special constructs which require a different syntax to normal.
>In other languages these are not required.
A language without type checking is a recipe for catastrophe. Perhaps it is
C's loose type checking that has earnt it it's reputation as being the language
to code in if you don't mind the odd crash every so often.

>iii) Support. Virtually every computer system in the world has a C++ compiler,
>and it is being updated continually. Because that is where all the money is.
>Not every computer system in the world has Ada on it, plus most will not be
>powerful enough to run it. Plus the cost of the respective compilers. C++ 
>compilers, are cheaper than Ada ones. On all if not the majority of platforms.
Ever tried GNAT, the free Ada 9x compiler? Certainly GNU C++ is not cheaper 
than this.

>iv) Libraries, Code already written and available experts. C++ has access straight
>away to all the C code libraries written. Not just a few nice ones, but all
>C library code ever written, one reason the famous COBOL is still around!

Yes, but is it still a good reason to carry on using it, when it is perfectly
possible to interface another language anyway (and it's already been done for
you - most useful libraries have interfaces for many languages, and why would
you want to use anything that isn't one of the 'few nice ones', when you
could use a library that is much better known, and has much better support for
the same purpose?)

>Lots and lots of code has already been written in C++,and also any company can
>go out and get any number of C++ programmers, or even C ones and train them,
>whereas there are fewer Ada programmers.
Before high level languages were designed, lots and lots of code was written
in assembler. Once the high levels were available, everyone had to be retrained
to use them. But nobody stuck to assembler, did they?

>v) OS Support. All modern OS have C and C++ support for there libraries. For Ada
>to access the same stuff as C and C++ someone has to write specfications for the
>C and C++ code in the OS. It is not supported straight out by OS Vendors.

Does this really make any difference? You just get an interface library for the
OS with the compiler. Thats it - no problem.

>This is just my .005 pence worth, and feel totally free to totally disagree.
Have done.


-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[David Emery]

>v) OS Support. All modern OS have C and C++ support for there
>libraries. For Ada to access the same stuff as C and C++ someone has
>to write specfications for the C and C++ code in the OS. It is not
>supported straight out by OS Vendors.

OK, how many operating systems (or anything else) have 'native' C++
interfaces, and how many provide access for C++ through an existing C
interface?  

In some cases (e.g. X Windows/Fresco) there are C++ interfaces that
have evolved from pre-existing C interfaces.  But I suspect the great
majority of interfaces used by C++ programmers are 'reused' C
interfaces.  

				dave
--
--The preceeding opinions do not necessarily reflect the opinions of
--The MITRE Corporation or its sponsors. 
-- "A good plan violently executed -NOW- is better than a perfect plan
--  next week"                                      George Patton
-- "Any damn fool can write a plan.  It's the execution that gets you
--  all screwed up"                              James Hollingsworth
-------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[R. William Beckwith]

David Emery (emery@goldfinger.mitre.org) wrote:

: In some cases (e.g. X Windows/Fresco) there are C++ interfaces that
: have evolved from pre-existing C interfaces.  But I suspect the great
: majority of interfaces used by C++ programmers are 'reused' C
: interfaces.  

Actually Fresco's interface is defined in CORBA's IDL.

Gee, I hate to take away the only example of an OS or
a windowing system interface defined in C++.  ;-)

Hmmm.  How about InterViews.  I can't think of any others.

... Bill

------------------------------------------------------
e-mail: Bill.Beckwith@ois.com        |    Team Ada
Objective Interface Systems, Inc.    | dist, full O-O
1895 Preston White Drive, Suite 250  | multithreading
Reston, VA  22091-5448  U.S.A.       |    built in

Re: Why don't large companies use Ada?

[John Goodsen]

In article <3bjdhm$9q1@gamma.ois.com> beckwb@ois.com (R. William Beckwith) writes:

   : In some cases (e.g. X Windows/Fresco) there are C++ interfaces that
   : have evolved from pre-existing C interfaces.  But I suspect the great
   : majority of interfaces used by C++ programmers are 'reused' C
   : interfaces.  

   Actually Fresco's interface is defined in CORBA's IDL.

   Gee, I hate to take away the only example of an OS or
   a windowing system interface defined in C++.  ;-)

   Hmmm.  How about InterViews.  I can't think of any others.

Interviews is pure C++, not C.  And also, Fresco is based upon
Interviews, which makes it inherently C++.  Not sure what you
mean when you say:

    "Actually Fresco's interface is defined in CORBA's IDL."

Could explain this? thanks.

-- 
--
John Goodsen                         Currently on-site at:
The Dalmatian Group                       JP Morgan 
User Interface Specialists                60 Wall St., New York City
jgoodsen@radsoft.com                      jgoodsen@jpmorgan.com

Re: Why don't large companies use Ada?

[R. William Beckwith]

John Goodsen (jgoodsen@trinidad.radsoft.com) wrote:
: In article <3bjdhm$9q1@gamma.ois.com> beckwb@ois.com (R. William Beckwith) writes:

:    : In some cases (e.g. X Windows/Fresco) there are C++ interfaces that
:    : have evolved from pre-existing C interfaces.  But I suspect the great
:    : majority of interfaces used by C++ programmers are 'reused' C
:    : interfaces.  

:    Actually Fresco's interface is defined in CORBA's IDL.

:    Gee, I hate to take away the only example of an OS or
:    a windowing system interface defined in C++.  ;-)

:    Hmmm.  How about InterViews.  I can't think of any others.

: Interviews is pure C++, not C.  And also, Fresco is based upon
: Interviews, which makes it inherently C++.  Not sure what you
: mean when you say:

:     "Actually Fresco's interface is defined in CORBA's IDL."

: Could explain this? thanks.

Fresco is loosely based on InterViews.  Many of Fresco's
graphical concepts are derived from InterViews.  But Fresco
is a new development effort.

Fresco's API or interface was designed in CORBA IDL, not C++.
This provides not only the definition of a distributed GUI
API, but a multi-language GUI API.  The C++ X11R6 Fresco
distribution is a sample implementation.

We're working on an Ada 95 implementation.  The definition
of the interfaces (IDL) doesn't change, only the implementation
language.  It wouldn't surprise me if some enterprising
Smalltalk tools vendor might want to write a Smalltalk version.

... Bill

Re: Why don't large companies use Ada?

[Akopov Mikhail]

Jules (csusb@csv.warwick.ac.uk) wrote:
J>Converting to C++ is astoundingly difficult if you want to do it well. See the
J>thread on learning OOP for more details.
J>Not retraining someone to use a different language to the one they learnt
J>at university seems like the cheaper option to me.

Note there are not only university graduators but programmers who
works already for some years too. C++ gives them the chance to smooth
derivating to OOP _not_stopping_their_work_. Isn't it important?[H
'Best is the enemy of good'.

Vale! -Michael Akopov

Re: Why don't large companies use Ada?

[Jules]

In article <3bndsc$jlq@harpy.skfgvc.pyatigorsk.su>,
	amm@skfgvc.pyatigorsk.su (Akopov Mikhail) writes:
>Jules (csusb@csv.warwick.ac.uk) wrote:
>J>Converting to C++ is astoundingly difficult if you want to do it well. See the
>J>thread on learning OOP for more details.
>J>Not retraining someone to use a different language to the one they learnt
>J>at university seems like the cheaper option to me.
>
>Note there are not only university graduators but programmers who
>works already for some years too. C++ gives them the chance to smooth
>derivating to OOP _not_stopping_their_work_. Isn't it important?[H
>'Best is the enemy of good'.
>
>Vale! -Michael Akopov

The point that was made on the other thread is that even for those programmers
who already know C, It is easier to learn OOP through a language such as Ada than
by learning C++. C++'s class system is confusing and difficult to learn. The
systems of Ada, and Object Pascal, etc. are much easier to learn, and still can
be used to write useful programs. Admittedly, If you are not worried about the
quality of the OO code you are writing, it is easier for a C programmer to learn
C++, but he would tend NOT to use the object-oriented features of the language, as
he can still use all the methods he used in the past. This temptation means that
very often, the code an alleged C++ programmer produces is, to quote some other
poster (I've lost the article, so I can't be more specific with the attribution)
"Bad C with OOP sugar", simply because he has learnt C beforehand. It is therefore
a good idea to start working in a completely new language when you start to learn
OOP, and perhaps later move back onto C++, once you fully understand the ideas
of OOP. I understand that this may take some time, and that during this time
period a programmer would not really be able to work, but this is IMHO the only
way to learn good OOP practices.
-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Ada as intro language

[un033144]

In article <3bidce$a83@borage.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk (Jules) writes:
> Oh come off it, nobody teaches BASIC anymore, do they?
> Oh, and how many of them are teaching Ada now? Quite a lot I believe. As
> a large proportion of those graduates know Ada, doesn't it make sense to
> use it?

	Well, I'm an undergraduate at West Virginia University, and Ada is the
intro language that is taught here.

	As far as other schools that teach Ada as there intro language, I don't
remember the exact number, but I think I heard that it's somewhere under 20.

Re: Ada as intro language

[un033144]

In article <1994Dec7.151830.12332@wvnvms>, un033144@wvnvms.wvnet.edu writes:
> In article <3bidce$a83@borage.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk (Jules) writes:
>> Oh come off it, nobody teaches BASIC anymore, do they?
>> Oh, and how many of them are teaching Ada now? Quite a lot I believe. As
>> a large proportion of those graduates know Ada, doesn't it make sense to
>> use it?
> 
> 	Well, I'm an undergraduate at West Virginia University, and Ada is the
> intro language that is taught here.
> 
> 	As far as other schools that teach Ada as there intro language, I don't
> remember the exact number, but I think I heard that it's somewhere under 20.
> 
>

	Well, I have egg on my face....  That figure is way off base...  I'll
have to check my sources more closely next time..

<slinking out of here>
 

Re: Ada as intro language

[Michael Feldman]

In article <1994Dec7.151830.12332@wvnvms>,  <un033144@wvnvms.wvnet.edu> wrote:
>In article <3bidce$a83@borage.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk (Jules) writes:
>> Oh come off it, nobody teaches BASIC anymore, do they?
>> Oh, and how many of them are teaching Ada now? Quite a lot I believe. As
>> a large proportion of those graduates know Ada, doesn't it make sense to
>> use it?
>
>	Well, I'm an undergraduate at West Virginia University, and Ada is the
>intro language that is taught here.
>
>	As far as other schools that teach Ada as there intro language, I don't
>remember the exact number, but I think I heard that it's somewhere under 20.

Actually, it's nearly 80. See below.

Mike Feldman

--- cut here ---

Ada as a Foundation Programming Language

Michael B. Feldman
Chair, ACM-SIGAda Education Working Group
Department of Electrical Engineering and Computer Science
The George Washington University
Washington, DC 20052
(202) 994-5919 (voice) - (202) 994-0227 (fax)
mfeldman@seas.gwu.edu

November 1994

This report lists the colleges and universities known -- or at least credibly 
believed -- to have adopted Ada as a "foundation language," at CS1, CS2, 
or CS7 level in their computer science (or similar) curricula. I choose
to keep track of precisely these courses because they are taken by 
students in either first or second year, and thus early enough to serve
as a foundation upon which to build a large portion of the undergraduate 
software curriculum.

This data is, for the most part, verifiable, based on first-hand reports 
from the teachers of those courses. In a few cases, publishers' textbook 
adoption data was used. I would like to keep this list as complete and up to 
date as possible, so please let me know of any additions or corrections.

Group 1: Colleges and Universities Introducing Ada as the First Language
         Taught in an Undergraduate Computing Curriculum.

Allan Hancock College, California
Auburn University
Australian Defence Force Academy, Canberra
Basque University, San Sebastian, Spain
Birmingham Southern College, Alabama
California State University, Long Beach
California Polytechnic State University, San Luis Obispo
Chalmers University of Technology, Sweden
Clemson University, South Carolina
Conservatoire National des Arts et Metiers, Paris,  France
Cranfield Institute of Technology, United Kingdom
Cypress College, California
Ecole Nationale Superieure des Telecommunications de Bretagne, France
Ecole Superieure en Sciences Informatiques, Nice, France
Edith Cowan University, Perth, Australia
Embry-Riddle Aeronautical University, Arizona and Florida campuses
Florida Institute of Technology
Fayetteville State University, North Carolina
Flinders University, Australia
The George Washington University, Washington, DC
Indiana-Purdue University, Ft. Wayne
Institut National des Sciences Appliquees, Toulouse, France
Jerusalem College of Technology, Israel
LeMoyne College, New York
Lenoir-Rhyne College, North Carolina
Loyola Marymount University, California
Marion County Technical Center, West Virginia
Marshall University, West Virginia
Montana State University
Muskingum College, Ohio
National Defense Management College at Taiwan, Republic of China
Northeast Missouri State University
Northern Arizona University
Norwich University, Vermont
Otterbein College, Ohio
Portsmouth Polytechnic, United Kingdom
Royal Melbourne Institute of Technology, Australia
Royal Military Academy, Belgium
Sacred Heart College (Connecticut)
Saint Mary College, Kansas
Sam Houston State University, Texas
San Diego Mesa College, California
Santa Clara University
Seattle University
South Bank University, United Kingdom
Southampton Institute of Higher Education, United Kingdom
Southern Arkansas University
St. Cloud State Univ., Minnesota
State University of New York at Plattsburgh
Stockton College of New Jersey
Swinburne University of Technology, Australia
Technical College of Berlin, Germany
Technical University of Madrid (School of Telecommunications), Spain
United States Air Force Academy
University of Adelaide, Australia
University of Aston, United Kingdom
University of Bradford, United Kingdom
University of Canberra, Australia
University of Dayton, Ohio
University of Glamorgan, Wales, United Kingdom
University of Hertfordshire, United Kingdom
University of Lancaster, United Kingdom
University of Liverpool, United Kingdom
University of Maryland (University College)
University of Muenster, Germany
University of New Orleans
University of North Dakota
University of Paisley, United Kingdom
University of Rome at Tor Vergata, Italy
University of Salzburg, Austria
University of Sofia, Bulgaria
University of South Dakota
University of South Florida
University of Stafford, United Kingdom
University of Wales, Aberystwyth, United Kingdom
University of York, United Kingdom
Valparaiso University, Indiana
West Virginia University

Group 2: Colleges and Universities Not Introducing Ada as the First Language
         but Rather in a CS2 or Data Structures Course (or equivalent)

Briar Cliff College, Iowa
California Polytechnic State University, Pomona
California State University, Fullerton
College of West Virginia, Beckley
Daniel Webster College, New Hampshire
Davis and Elkins College, West Virginia
Ecole d'Ingenieurs de l'Etat de Vaud, Switzerland
Florida International University
Gallaudet University, Washington, DC
Georgia State University
Indiana University, New Albany
Mesa State College, Colorado
Monterey Peninsula College, California
Mount Mercy College, Iowa
Murray State University, Kentucky
National University, California
Nicholls State University, Louisiana
Northern Arizona University
Northern Kentucky University
Oglethorpe University, Georgia
Ohio University, Athens
Pennsylvania State University, Harrisburg
Rose Hulman Institute of Technology, Indiana
Southwest Baptist College, Missouri
Shippensburg University, Pennsylvania
State University of New York at Fredonia
Swiss Federal Institute of Technology, Lausanne
University of Alaska, Fairbanks
University of Brighton, United Kingdom
University of Geneva, Switzerland
University of Missouri, Columbia
University of Richmond, Virginia
University of Scranton, Pennsylvania
University of Seville, Spain
University of Texas, Austin
University of Texas, Permian Basin
Western New England College, Massachusetts

Re: Why don't large companies use Ada?

[David Weller]

In article <D03AL2.5xw@aston.ac.uk>, An Amoeba <bromlegb@aston.ac.uk> wrote:
>Why should they bother??
>This is just my .005 pence worth, and feel totally free to totally disagree.
>
I do.  Put some jam in your pockets, you're about to be toasted :-)

>Why?
>
>i) Cost. If the user has a basic knowledge of C or other programming language
>then converting to C++ is relatively easy.
>
Wrong.  That's not been my observations nor the observations of
others.  Cite evidence to support this claim.

>ii) Ada's complexity. Most new users will quickly come across all Ada's *nice*
>type checking, or special constructs which require a different syntax to normal.
>In other languages these are not required.
>
Define complexity.  This is such a vague claim I can't even find a
good way to flame back :-)  Why is that Ada, a strongly typed
language, somehow finds disfavour in your eyes, while the same claim
to C++ appears to be revered by you?

>iii) Support. Virtually every computer system in the world has a C++ compiler,
>and it is being updated continually. Because that is where all the money is.
>Not every computer system in the world has Ada on it, plus most will not be
>powerful enough to run it. Plus the cost of the respective compilers. C++ 
>compilers, are cheaper than Ada ones. On all if not the majority of platforms.
>
I didn't pay a thing for my Ada compiler, and it works quite nicely.
Runs on most OS's I know of: MickeySoft-DOS, OS/2, NeXT, Amiga,
Linux, SunOS, Solaris, HP, etc., etc.  I think your info is a little
out of date.

>iv) Libraries, Code already written and available experts. C++ has access straight
>away to all the C code libraries written. Not just a few nice ones, but all
>C library code ever written, one reason the famous COBOL is still around!
>Lots and lots of code has already been written in C++, and also any company can
>go out and get any number of C++ programmers, or even C ones and train them,
>whereas there are fewer Ada programmers.
>
Ada also has access "straight away" to C libraries.  *poof*  There
goes another one of your arguments.  Tsk, tsk.  YOu need to work
harder at this.

>v) OS Support. All modern OS have C and C++ support for there libraries. For Ada
>to access the same stuff as C and C++ someone has to write specfications for the
>C and C++ code in the OS. It is not supported straight out by OS Vendors.
>
When I get an Ada compiler from a vendor, I typically get complete OS
bindings.



-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
	|"Quitting C++ isn't so difficult, provided you show as much |
	|	persistence stopping as you did starting." dweller   |

Re: Why don't large companies use Ada?

[Robert I. Eachus]

In article <D03AL2.5xw@aston.ac.uk> bromlegb@aston.ac.uk (An Amoeba) writes:

 > Why should they bother??

   Hmmm.... Should I bother replying?  Only if I can do so politely,
this post will certain attract its share of flames.

 > Given that most students coming out of college/University will have some
 > experience in programming some type of language Pascal, Modula-2/3, C
 > and possibly BASIC plus any others and also most colleges/Universities
 > provide courses in C, it is totally logical, and more important cost
 > saving to send/train employees on C++ courses, and thus program systems
 > in C++.

   Let me recast this.  Students who have learned Pascal in school
will have little trouble transitioning to Ada, and those who started
out with the second most popular introductory computing language, Ada,
will have even less trouble.  (As I remember the most recent survey
Scheme is third...is that right Mike?)  Those who learned to program
on their own have most likely learned BASIC or C, and should be sent
to school to learn better practices.  Of course, well trained software
engineers can use any reasonable language.

 > i) Cost. If the user has a basic knowledge of C or other
 > programming language then converting to C++ is relatively easy.

   Reminds me of the two part plan for converting lead into gold.
Part one is to mine the lead. Part two is still in the planning
stages.  Seriously, learning OOP is NOT easy, writing C in C++ is
easy.  If you are planning to use C, use C.

 > ii) Ada's complexity. Most new users will quickly come across all
 > Ada's *nice* type checking, or special constructs which require a
 > different syntax to normal.  In other languages these are not
 > required.

   Yes, strong type checking is very nice, and not very complex.
Special constructs?  Are you referring to the language constructs for
task, exceptions, and/or generics?  But these are nice features to
have available, even C++ has adopted two of them, and if you use them
you do need special syntax.  But most Ada programmers never need to
use features from chapters 9, 12 and 13, so I guess most of that
syntax is not normally required in Ada either.  It's just that in Ada,
it is there when you need it.

  > iii) Support. Virtually every computer system in the world has a
  > C++ compiler, and it is being updated continually. Because that is
  > where all the money is.  Not every computer system in the world
  > has Ada on it, plus most will not be powerful enough to run it.
  > Plus the cost of the respective compilers. C++ compilers, are
  > cheaper than Ada ones. On all if not the majority of platforms.

    Actually if you check, there are Ada compilers available for many
more machines than C++ compilers.  (Especially if you count "bare
boards" and embedded processors.)  On the other hand C is probably the
most popular language in terms of number of targets.  There are many
small chips where the only programming languages available are
assembler, C, and perhaps PL/M or Forth.

 > iv) Libraries, Code already written and available experts. C++ has
 > access straight away to all the C code libraries written. Not just
 > a few nice ones, but all C library code ever written, one reason
 > the famous COBOL is still around!  Lots and lots of code has
 > already been written in C++, and also any company can go out and
 > get any number of C++ programmers, or even C ones and train them,
 > whereas there are fewer Ada programmers.

   It is a LOT easier to build a good Ada interface to a C library
than to build a good C++ interface to the same library.  In either
language a mediocre thin binding can be whipped up quickly.

 > v) OS Support. All modern OS have C and C++ support for there
 > libraries. For Ada to access the same stuff as C and C++ someone
 > has to write specfications for the C and C++ code in the OS. It is
 > not supported straight out by OS Vendors.

   Hmmm...  I must use a lot of ancient OS's.  Seriously, POSIX Ada
bindings (a very good thick binding) are available on most Unix
platforms, and I'm in the process of upgrading the Ada 9X bindings
that Pass Travis developed on the Amiga.  I'm not aware of any real
C++ bindings, on the Amiga or elsewhere.  Am I missing something?

 > This is just my .005 pence worth, and feel totally free to totally disagree.
 
   Not even your full two cents' worth?  I predict you will get a
large on your investment.
--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: Why don't large companies use Ada?

[Robert Dewar]

Gareth's idea that knowing C means that learning C++ is easy, and his
view that Ada is a much more complex language than C++ are laughable, but
the programming language field is one in which people seem happy to say
stupid things about programming lanuages (in all directions) without
knowing anything about the languages they are talking about.

What is for sure is that most students coming out of universities are
woefully unequipped for writing serious, realiable software. THe idea
that knowing how to hack around in C or Basic (or any other language
for that matter) has anything to do with serious software engineering
only goes to show the lack of awareness.

Sometimes I wonder whether, regardless of the language you intend to use,
you would do better to specifically hire people who do *not* know the
language so that they don't bring a load of misconceptions to the table.
Maybe the best thing is if they don't know any languages at all. The
teaching of programming is often so bad that it has definitely overall
negative value.

The arguments that one should use language X because that's what everyone
knows and what systems support most easily etc. is a recipe for using
obsolete technology. For years, Fortran marched on in the US under this
banner.

I am happy to see technical points in comparison of languages, but this
argument is non-technical and in the long run less relevant than people
imagine. Note that the only technical point that Gareth made was that
somehow all the strong typing etc. in Ada would make people's life
harder, which is of course nonsense, and presumably is nonsense borne
of ignorance.

It is amazing how programming languages get characterized by second
hand opinions. For example, looking at the cross-posting of this thread,
how many of out out there don't know COBOL at all, but are sure that it
is extremely verbose and this is one of its disadvantages. Totally false
of course, COBOL programs are often considerably more compact than Pascal
programs doing similar things because of some very nice concise COBOL
grammar. Also of course totally irrelevant, counting the number of characters
on the page is NOT the way to determine the best language (otherwise we
would automatically conclude that a language that disallowed comments
was best :-)

Re: Why don't large companies use Ada?

[Scott McLoughlin]

dewar@cs.nyu.edu (Robert Dewar) writes:

> 
> What is for sure is that most students coming out of universities are
> woefully unequipped for writing serious, realiable software. THe idea
> that knowing how to hack around in C or Basic (or any other language
> for that matter) has anything to do with serious software engineering
> only goes to show the lack of awareness.

I don't think it appropriate to knock on C (or Basic, but bleck ;-).
While not the one true language, C has many strengths that I think
we can all recognize. It's very flexible,powerful and simple. It's 
often the first horse out of the gate on new platforms (after an
assembler) for many reasons. 

C allows one to get fairly close to CPU data representations/op
codes using standard operators. C seems to be good at generating
space efficient programs.

Furthermore, I am sure that much "serious software engineering"
has been conducted using C as the implementation language. 

> Sometimes I wonder whether, regardless of the language you intend to use,
> you would do better to specifically hire people who do *not* know the
> language so that they don't bring a load of misconceptions to the table.
> Maybe the best thing is if they don't know any languages at all. The
> teaching of programming is often so bad that it has definitely overall
> negative value.

In my experience, this is not accurate. I don't know what you mean
be "misconceptions", but I think all of us know what I mean when I
say "style, idiom, convention" - the stuff you can only pick up
by banging on some problems with a language, making mistakes, 
seeing what others have done, etc.  A sort of ideal tabula rasa
is frequently described here on the net (usually with regard to
teaching language X) and I think that this is rather naive.

Egads, I don't believe that I'm actually defending C! Look, I 
sincerely hope that the programming community, economic
environment, or whatever it takes changes so that other
non-C languages can flourish.  While I can write cryptic
for() loops in my sleep, I've taken a shine to Lisp and Forth
in the past two years or so (neither typed, BTW). I've been
reading the Ada, Modula-3 usenet groups and they look 
interesting. Sather/Eiffel also sound interesting.

OTOH, I don't see any budding Phillipe Kahn's of the Ada/
Modula3/Sather/Eiffel world marshalling marketing muscle
behind these languages.  I don't see anyone out there
besting GNU with a free implementation of these languages
that will kick gcc's ass in terms of features, debugging,
support, ports (Unix,VMS,DOS,Windows,Mac,OS/2...), 
packaging, RAM consumption, footprint, etc.

While some folks _love_ C, many (most?) who have used the
language long enough for the "glow" to wear off, don't
love it and would appreciate a better tool, thank you very
much.  But when it comes to shopping around time, there
aren't many clear alternatives with a clear future or 
momentum behind them, or the price isn't right, or imps
don't support the box/os you have, or the executable
images produced are too large, or whatever.

Time for the non-C language X advocates to stand and
deliver. To overcome the force of habit, there must
be an alternative that is readily percieved as better
for the types of programs folks want to write on the
platform they need to write them for.

=============================================
Scott McLoughlin
Conscious Computing
=============================================

Re: Why don't large companies use Ada?

[Andrew Koenig]

In article <qa6owc1w165w@sytex.com> smcl@sytex.com (Scott McLoughlin) writes:

> dewar@cs.nyu.edu (Robert Dewar) writes:

> > What is for sure is that most students coming out of universities are
> > woefully unequipped for writing serious, realiable software. THe idea
> > that knowing how to hack around in C or Basic (or any other language
> > for that matter) has anything to do with serious software engineering
> > only goes to show the lack of awareness.

> I don't think it appropriate to knock on C (or Basic, but bleck ;-).

I don't think he is.  Why do you think so?
-- 
				--Andrew Koenig
				  ark@research.att.com

Re: Why don't large companies use Ada?

[Robert I. Eachus]

In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:

  > Also of course totally irrelevant, counting the number of characters
  > on the page is NOT the way to determine the best language (otherwise we
  > would automatically conclude that a language that disallowed comments
  > was best :-)

    Nonsense! If you programmed in APL, you would know that APL
permits comments. 

   (And of course APL has a special character to begin comments... ;-)

   (For those not familiar with APL, it is an extremely terse
programming language.  For example, I know of three extremely short
programs which the user to enter a number, and print all primes less
than or equal to that number.  Each implements a different definition
of prime numbers, and the shortest which is sixteen charactes long, is
based on Wilson's theorem.  The "standard" software engineering
approach for APL is called throw-away coding.  The idea is to write
everything as one-line functions, with several lines of comments in
each which explain the interface and intended use.  If you need to
modify the function, you begin by throwing away the code.)

--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: Why don't large companies use Ada?

[Mike Chapman]

Robert I. Eachus (eachus@spectre.mitre.org) wrote:
: In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:

:   > Also of course totally irrelevant, counting the number of characters
:   > on the page is NOT the way to determine the best language (otherwise we
:   > would automatically conclude that a language that disallowed comments
:   > was best :-)

:     Nonsense! If you programmed in APL, you would know that APL
: permits comments. 

:    (And of course APL has a special character to begin comments... ;-)

:    (For those not familiar with APL, it is an extremely terse
: programming language.  For example, I know of three extremely short
: programs which the user to enter a number, and print all primes less
: than or equal to that number.  Each implements a different definition
: of prime numbers, and the shortest which is sixteen charactes long, is
: based on Wilson's theorem.  The "standard" software engineering
: approach for APL is called throw-away coding.  The idea is to write
: everything as one-line functions, with several lines of comments in
: each which explain the interface and intended use.  If you need to
: modify the function, you begin by throwing away the code.)

The last time I wrote any APL code (ca 15 years ago), as an
optimization step I had to throw away all the comments to allow
enough space for the program's data in memory!!

This was ca. 1000 lines of APL code.
I bet the guy who had to maintain this had real fun!

The aspect of throw-away functions in APL I can readily confirm.
You could spend a whole day writing a line of code.
Two days later you would not understand it any more.

-- 
Mike Chapman <Mike.Chapman@muc.de>
    As far as the laws of mathematics refer to reality, they are 
    not certain; and as far as they are certain, they do not refer 
    to reality.  -- Albert Einstein --

Re: Why don't large companies use Ada?

[Norman H. Cohen]

In article <1994Dec2.225748.1065@isis.muc.de>, Mike.Chapman@muc.de
(Mike Chapman) writes: 

|> The last time I wrote any APL code (ca 15 years ago), as an
|> optimization step I had to throw away all the comments to allow
|> enough space for the program's data in memory!!

It makes me appreciate the ZAP student assembler (that's an assembler FOR
students, not an assembler OF students!) that we used at Cornell 20 years
ago.  (20 years??!!!! Where has the time gone?)

The assembler had a first pass in which it read card images containing
assembly-language source into a buffer, enlarging the buffer as necessary
as more cards were read in.  The second pass translated assembly language
into machine code and placed the machine code in the same buffer, since
a typical 80-byte line in the buffer generated from two to six bytes
worth of machine code.  But if you had a large DS (Define Storage)
directive near the beginning of your program, that one line could easily
take up more space in the machine-language image than had yet been made
available in the buffer from fully processed card images.  The solution
was to add more card images, typically comment cards, before the DS
directive, to ensure that there would be enough space in the buffer by
the time the directive was reached.

So, in contrast to Mike Chapman's APL translator, the ZAP assembler made
you ADD more comments when you ran out of memory!

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[D'Arcy J.M. Cain]

Mike Chapman (Mike.Chapman@muc.de) wrote:
: The aspect of throw-away functions in APL I can readily confirm.
: You could spend a whole day writing a line of code.
: Two days later you would not understand it any more.

I have always said that I won't use a language of which the proponents put
code in front of each other and say "Bet you can't guess what this does!"

-- 
D'Arcy J.M. Cain (darcy@druid.com)  |
Planix, Inc.                        |   Democracy is three wolves and a
Toronto, Ontario, Canada            |   sheep voting on what's for dinner.
+1 416 424 2871  (DoD#0082) (eNTP)  |

Re: Why don't large companies use Ada?

[M. J. Saltzman]

darcy@druid.com (D'Arcy J.M. Cain) writes:
>>[...APL is write-only...]

>I have always said that I won't use a language of which the proponents put
>code in front of each other and say "Bet you can't guess what this does!"

Umm, isn't that the point of the Obfuscated C Contest?
-- 
		Matthew Saltzman
		Clemson University Math Sciences
		mjs@clemson.edu

Re: Why don't large companies use Ada?

[Andrew Koenig]

In article <D0LvnA.8C3@druid.com> darcy@druid.com (D'Arcy J.M. Cain) writes:

> I have always said that I won't use a language of which the proponents put
> code in front of each other and say "Bet you can't guess what this does!"

But I know how to do that for any language.  Where does that leave you?
-- 
				--Andrew Koenig
				  ark@research.att.com

Re: Why don't large companies use Ada?

[John Goodsen]

In article <D0nqKF.Jyt@research.att.com> ark@research.att.com (Andrew Koenig) writes:

   In article <D0LvnA.8C3@druid.com> darcy@druid.com (D'Arcy J.M. Cain) writes:

   > I have always said that I won't use a language of which the proponents put
   > code in front of each other and say "Bet you can't guess what this does!"

   But I know how to do that for any language.  Where does that leave you?

				   --Andrew Koenig
				     ark@research.att.com

Out of touch with the real software development community.  This is an
example of the type of mindset that a lot of engineers (not all,
though) who use Ada and then start touting the age old, boring, "my
language is better than yours" crap.

IMOE, of course. :-)
-- 
--
John Goodsen                         Currently on-site at:
The Dalmatian Group                       JP Morgan 
User Interface Specialists                60 Wall St., New York City
jgoodsen@radsoft.com                      jgoodsen@jpmorgan.com

Re: Why don't large companies use Ada?

[Con Bradley]

D'Arcy J.M. Cain (darcy@druid.com) wrote:

:Mike Chapman (Mike.Chapman@muc.de) wrote:
:: The aspect of throw-away functions in APL I can readily confirm.
:: You could spend a whole day writing a line of code.
:: Two days later you would not understand it any more.

:I have always said that I won't use a language of which the proponents put
:code in front of each other and say "Bet you can't guess what this does!"

There are competitions for the Worst Book of the year, the Worst Movie 
(film) of the year etc. and it doesn't prevent you reading a book nor going 
to the movies(films). I think it's all part of a healthy language culture,

  Con Bradley.

Re: Why don't large companies use Ada?

[Ian S Nelson]

Excerpts from netnews.comp.programming: 12-Dec-94 Re: Why don't large
compani.. by Con Bradley@spitfire.pac 
> D'Arcy J.M. Cain (darcy@druid.com) wrote:
>  
> :Mike Chapman (Mike.Chapman@muc.de) wrote:
> :: The aspect of throw-away functions in APL I can readily confirm.
> :: You could spend a whole day writing a line of code.
> :: Two days later you would not understand it any more.
>  
> :I have always said that I won't use a language of which the proponents put
> :code in front of each other and say "Bet you can't guess what this does!"
>  
> There are competitions for the Worst Book of the year, the Worst Movie 
> (film) of the year etc. and it doesn't prevent you reading a book nor going 
> to the movies(films). I think it's all part of a healthy language culture,

Moreover, I'd have my doubts about a language that claimed you couldn't
make anything obfuscated in it.  I've seen some downright scary things
in Pascal and Smalltalk.  I've seen a few funky things in rexx too. 
Those are about the "simplest" or "easist" to read languages I know.  
--
Ian S. Nelson  <bonovox@cmu.edu>                  finger for PGP key
Carnegie Mellon Math/Computer Science 
My opinions are not the school's, although they should be!

Re: Why don't large companies use Ada?

[Ross Mather]

Con Bradley (ceb@spitfire.pact.srf.ac.uk) wrote:

: D'Arcy J.M. Cain (darcy@druid.com) wrote:

: :Mike Chapman (Mike.Chapman@muc.de) wrote:
: :: Two days later you would not understand it any more.

: :I have always said that I won't use a language of which the proponents put
: :code in front of each other and say "Bet you can't guess what this does!"

: There are competitions for the Worst Book of the year, the Worst Movie 
: (film) of the year etc. and it doesn't prevent you reading a book nor going 
: to the movies(films). I think it's all part of a healthy language culture,

But we don't have competitions for who can build the worst bridge / worst
building / worst anything else that is meant to be *engineered*.

Computer programming languages are Engineering Tools designed to let
engineers do their jobs. Yeah sure we have to be a bit artistic now and
then, but that doesn't equate programming with filmmaking. 

cheers

Ross
-- 
Ross M. Mather                         mather@mscn.nl, rmt@dasc.nl
I speak for me, and me alone. Canter & Siegel please take note.
Rugby isn't a way of life, it's more important than that. 

Re: Why don't large companies use Ada?

[Peter Seebach]

In article <D0qwIu.A9q@mscn.nl> mather@mscn.nl (Ross Mather) writes:
>But we don't have competitions for who can build the worst bridge / worst
>building / worst anything else that is meant to be *engineered*.

Not all programs are meant to be engineered.  Tools are.  Works of art
aren't.  The situation with natural language parallels this; there are
competitions for worst thing for artistic purposes, but these do not carry
over into engineering work.  In fact, people who can write particularly badly
on purpose generally write better when they do* need quality.

>Computer programming languages are Engineering Tools designed to let
>engineers do their jobs. Yeah sure we have to be a bit artistic now and
>then, but that doesn't equate programming with filmmaking. 

It's not that we have* to, it's that there's absolutely no point if
we aren't.  Engineering is an art form; no one who believes otherwise has
ever been a great engineer.

To understand how to make a truly bad program, you have to know what the bad
and good things about it are.  This is a good way to study quality.

>cheers

>Ross M. Mather                         mather@mscn.nl, rmt@dasc.nl

-seebs
-- 
Peter Seebach - seebs@solutions.solon.com  -- seebs@intran.xerox.com
C/Unix proto-wizard -- C/Unix questions? Send mail for help.

Re: Why don't large companies use Ada?

[Markus Freericks]

[follow-up set to comp.programming]

In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:
> Gareth's idea that knowing C means that learning C++ is easy, and his
> view that Ada is a much more complex language than C++ are laughable, 
[...]

When programming C++, you must really distinguish between "being able to
use it" and "mastering it in every last nitty-gritty detail". I know C
programmers who have been programming correct C++ after a two-hour intro,
simply by ignoring most of the "OO"-features. There is no single "correct"
paradigm for programming C++, and some of the simplest ones employ classes
as better structs, or as modules.  You can use C++ profitably without ever
knowing what a copy constructor (or even a reference!) is, especially if
you employ third-party class packages that provide objects that "work just
like the basic types".

C++ _is_ incredibly complex, but most of that complexity is provided for
_library writers_. The "end user programmer", if there ever is such a
beast, _can_ use C++ as better-C-with-typechecking.

All with a big IMHO,
	Markus 

PS: I'd rather programming  in SML, or  Clean, or Self, or evern  Modula-3,
but we are talking C-vs-Pascal here...

Re: Why don't large companies use Ada?

[James Kanze US/ESC 60/3/141 #40763]

In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert
Dewar) writes:

|> I am happy to see technical points in comparison of languages, but this
|> argument is non-technical and in the long run less relevant than people
|> imagine. Note that the only technical point that Gareth made was that
|> somehow all the strong typing etc. in Ada would make people's life
|> harder, which is of course nonsense, and presumably is nonsense borne
|> of ignorance.

I can't resist pointing it out.  Using strict type-checking *does*
make writing programs harder.  You actually have to think about what
you are doing, and maybe even do a bit of design before writing,
rather than just banging in a bunch of code, and dropping into the
debugger at the first sign of a core dump.

I might also point out that used correctly, C++ has relatively strict
type checking.  (When discussing languages, it is important to know
what is being compared to what.  In most of my work, the alternative
to C++ is C, so I have become somewhat of a C++ fan.)
-- 
James Kanze         Tel.: (+33) 88 14 49 00        email: kanze@gabi-soft.fr
GABI Software, Sarl., 8 rue des Francs-Bourgeois, F-67000 Strasbourg, France
Conseils en informatique industrielle --
                              -- Beratung in industrieller Datenverarbeitung

Re: Why don't large companies use Ada?

[Bart_van_der_Worp]

In article <KANZE.94Dec5174315@slsvhdt.us-es.sel.de> kanze@us-es.sel.de (James Kanze US/ESC 60/3/141 #40763) writes:
>From: kanze@us-es.sel.de (James Kanze US/ESC 60/3/141 #40763)
>Subject: Re: Why don't large companies use Ada?
>Date: 05 Dec 1994 16:43:14 GMT

>In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert
>Dewar) writes:

>|> I am happy to see technical points in comparison of languages, but this
>|> argument is non-technical and in the long run less relevant than people
>|> imagine. Note that the only technical point that Gareth made was that
>|> somehow all the strong typing etc. in Ada would make people's life
>|> harder, which is of course nonsense, and presumably is nonsense borne
>|> of ignorance.

>I can't resist pointing it out.  Using strict type-checking *does*
>make writing programs harder.  You actually have to think about what
>you are doing, and maybe even do a bit of design before writing,
>rather than just banging in a bunch of code, and dropping into the
>debugger at the first sign of a core dump.

Does not using strong typing make programming easier?
Of course not. Even using typedefs in C, without proper compiler checking,
allready makes programming easier in the long term. Indeed because
it is required to think, and because programs are more readable.

(Strong) typing gives most programming languages user friendly interfaces.
Our clients want them at their level, so why not use them for our own best
sake at our programming level?

Bart.

Re: Why don't large companies use Ada?

[Chris Dollin]

worp@euronet.nl (Bart_van_der_Worp) writes:

   (Strong) typing gives most programming languages user friendly interfaces.

Do you mean *strong* typing, or *static* typing?

--

Regards,    | ``"I can't suit myself," said Weinbaum, a little petulantly.
Kers.       | "I work for the Government".'' - Blish, "The Quincunx of Time".

Re: Why don't large companies use Ada?

[Bart_van_der_Worp]

In article <KERS.94Dec7094928@cdollin.hpl.hp.com> kers@hplb.hpl.hp.com (Chris Dollin) writes:
>From: kers@hplb.hpl.hp.com (Chris Dollin)
>Subject: Re: Why don't large companies use Ada?
>Date: Wed, 7 Dec 1994 17:49:28 GMT

>worp@euronet.nl (Bart_van_der_Worp) writes:

>   (Strong) typing gives most programming languages user friendly interfaces.

>Do you mean *strong* typing, or *static* typing?

I have to admit that I don't know the meaning of static typing. Does this
mean it is not possible to declare derived and/or subtypes?

What I ment in this posting was, that typing which is tailorred to the
requirements of a type makes reading, developing and maintaining
easier, dispite of the deficiencies in current languages.
An extreme is the C typedef with allmost no compiler checking, but
which adds to readability, to which I referred as 'user friendly interface'.

If some kind of typechecking is not flexible enough, like the previous
example of the dimension problem implemented as ADA types, my idea
is that it is allways possible to use a less strong mechanism, e.g.
derived types in ADA, which does the job. But of course, this will
be a *compromise* which allows operations which are functionally not
allowed. However, it will add to the readability of the code which is
also very important IMHO. A practical viewpoint.

The ideal typing has not yet been invented, if at all possible.

By the way, I ment customer in stead of client.....
Sorry for my English, wich is also far from ideal....

Bart.

Re: Why don't large companies use Ada?

[Alistair James Robert Young]

In article <KANZE.94Dec5174315@slsvhdt.us-es.sel.de>,
James Kanze US/ESC 60/3/141 #40763 <kanze@us-es.sel.de> wrote:
>
>I might also point out that used correctly, C++ has relatively strict
>type checking.  (When discussing languages, it is important to know
>what is being compared to what.  In most of my work, the alternative
>to C++ is C, so I have become somewhat of a C++ fan.)
>-- 

I'd agree that strict type-checking is a good thing, in that it can prevent
the many nasty errors that result from getting it wrong manually. However,
in some cases, it can be a nuisance. To quote a specific example, I am
in the throes of learning C++ m'self, and have a problem using the OS-specific
header files with C++.

These header files use a type PSZ to pass strings to functions. This type
(pointer to string terminated by zero) is exactly the same as a char *,
but the compiler refuses to acknowledge this, forcing me to use three or
four typecasts in every OS call.

Strict type-checking's all very well, but you need to be able to turn it off
for specific cases. BTW, if anyone knows a way to do this for PSZ/char *, I'd
really appreciate it!

Alistair

--------------------------------------------------------------------------
Alistair Young - Arkane Systems Software Development & PC Consultancy
The opinions above are my company's, because I OWN it!
[Development for OS/2 only!] Contact: ajry@st-and.ac.uk
"Time flies like an arrow, fruit flies like a banana" - Groucho Marx
GAT/CS/C/S d++ H s++:- g+ !p+ !au a- w+++ v+++ C++++ O++++$ P+ E+++
N+++ W--- M-- !V -po+(++) Y+ t+ R++ tv b+++ D++ e+ u** h--- f+ r++ n+ y?
PGP key fingerprint = 5C 27 43 25 6B 32 4D 6E  E7 69 EC 7B 77 46 13 92
finger ajry@st-and.ac.uk for full public key - encrypted mail welcome

Re: Why don't large companies use Ada?

[Jules]

In article <3c873t$lut@calvin.st-and.ac.uk>,
	ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
>In article <KANZE.94Dec5174315@slsvhdt.us-es.sel.de>,
>James Kanze US/ESC 60/3/141 #40763 <kanze@us-es.sel.de> wrote:
>>
>>I might also point out that used correctly, C++ has relatively strict
>>type checking.  (When discussing languages, it is important to know
>>what is being compared to what.  In most of my work, the alternative
>>to C++ is C, so I have become somewhat of a C++ fan.)
>>-- 
>
>I'd agree that strict type-checking is a good thing, in that it can prevent
>the many nasty errors that result from getting it wrong manually. However,
>in some cases, it can be a nuisance. To quote a specific example, I am
>in the throes of learning C++ m'self, and have a problem using the OS-specific
>header files with C++.
>
>These header files use a type PSZ to pass strings to functions. This type
>(pointer to string terminated by zero) is exactly the same as a char *,
>but the compiler refuses to acknowledge this, forcing me to use three or
>four typecasts in every OS call.
>
>Strict type-checking's all very well, but you need to be able to turn it off
>for specific cases. BTW, if anyone knows a way to do this for PSZ/char *, I'd
>really appreciate it!

This is one of the problems of adding strict typing onto the top of an existing
language, as was done for C++. Basically, the problem is the 'PSZ' type should
either be used in all cases, or not at all. It's a good idea to have the
separate type (in some cases you may also be using Pascal style strings,
which would have a different type [probably 'PST', following the hungarian
convention], also defined as being a char *, thus preventing you from mixing
the two types up), but the HUGE problem is the large number of C libraries
that use 'char *' for all strings, thus requiring type-casting between the
types. What you need is a new language, rather than an extension of an existing
one, which forces all users to completely rewrite their libraries to be
more compatible, rather than mindlessly attempting to write patch after patch
to interface their C++ code with existing C libraries. Enter Ada...



-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Brian Stern]

In article <3c9jle$796@thyme.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk
(Jules) wrote:

< This is one of the problems of adding strict typing onto the top of an
existing
< language, as was done for C++. Basically, the problem is the 'PSZ' type should
< either be used in all cases, or not at all. It's a good idea to have the
< separate type (in some cases you may also be using Pascal style strings,
< which would have a different type [probably 'PST', following the hungarian
< convention], also defined as being a char *, thus preventing you from mixing
< the two types up), 

Pascal strings should be defined as unsigned char *.  You'll have trouble
getting or setting the length if it's greater than 128 otherwise.  This
also prevents you from mixing them up with C strings (null terminated).

< 
< 
< -- 
< /* Julian R Hall                                csusb@csv.warwick.ac.uk
<    
<    Flames should be redirected to /dev/null - I don't know what
<    I'm saying myself so don't expect it to make sense all the time!         */

-- 
Brian  Stern  :-{)}
Toolbox commando and Menu bard
Jaeger@fquest.com

Re: Why don't large companies use Ada?

[Pete Gontier]

In article <3c9jle$796@thyme.csv.warwick.ac.uk>,
csusb@csv.warwick.ac.uk (Jules) wrote:

> In article <3c873t$lut@calvin.st-and.ac.uk>,
> ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
>
>> I am in the throes of learning C++ m'self, and have a problem using
>> the OS-specific header files with C++. These header files use a type
>> PSZ to pass strings to functions. This type is exactly the same as a
>> char *, but the compiler refuses to acknowledge this, forcing me to
>> use three or four typecasts in every OS call.

I think it's likely that this is the headers' fault. Perhaps they are
failing to use 'const' in the right places. Or you're on an Intel machine
and all that silly 'near' and/or 'far' business hasn't been done perfectly
(in which case brute-force type-casting is downright dangerous). The
following code compiles without a warning under Metrowerks CodeWarrior C++
version CW4:

        typedef char *foo;
        typedef char *bar;
        
        static void fooFunc (foo) { }
        static void barFunc (bar) { }
        static void charFunc (char *) { }

        void main (void)
        {
                foo baz = 0;
                bar quux = 0;
                char *cp1 = baz;
                char *cp2 = quux;

                barFunc (baz); fooFunc (quux);
                charFunc (baz); charFunc (quux);
        }

> This is one of the problems of adding strict typing onto the top of an
> existing language, as was done for C++... What you need is a new language...

Bzzzzt! Learn C++ before you bash it.

Don't get me wrong; I wish there were a Macintosh Ada compiler worth
using, because then I could get a real perspective on the language.

In the meantime, I'd rather use C++ than anything else, and I will
certainly avoid bashing Ada until I can do so from an informed
perspective.

-- 

 Pete Gontier // MacZealotry, Ink. // gurgle@dnai.com

 Where do I want go today? Anywhere but Chicago, thanks.

Re: Why don't large companies use Ada?

[Dr. Richard Botting]

Pete Gontier (gurgle@dnai.com) wrote:
: In article <3c9jle$796@thyme.csv.warwick.ac.uk>,
: csusb@csv.warwick.ac.uk (Jules) wrote:

: > In article <3c873t$lut@calvin.st-and.ac.uk>,
: > ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
: >
: >> I am in the throes of learning C++ m'self, and have a problem using
: >> the OS-specific header files with C++. These header files use a type
: >> PSZ to pass strings to functions. This type is exactly the same as a
: >> char *, but the compiler refuses to acknowledge this, forcing me to
: >> use three or four typecasts in every OS call.

: I think it's likely that this is the headers' fault. Perhaps they are
[...snip...]

Now PSZ is Hungarian for a pointer to Zero terminated string, isn't it?
Perhaps they planned to make a safe string type?

Could you add another header that declares inline functions that overload
the OS functions you might call that then call the real OS functions
with the necessary type casts... then let C++ do the hard work for you.

Sorry if someone already said this.

--
EMail: dick@csci.csusb.edu=rbotting@wiley.csusb.edu.
<a href="http://www.csci.csusb.edu/dick/">WWW</a>
Disclaimer::=`CSUSB may or may not agree with this message`.
Copyright(1994)::=`Copy as long as you include this copyright and signature`.

Re: Why don't large companies use Ada?

[Robert Firth]

In article <3c873t$lut@calvin.st-and.ac.uk> ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:

>These header files use a type PSZ to pass strings to functions. This type
>(pointer to string terminated by zero) is exactly the same as a char * ...

No, it isn't.  The two types are structurally equivalent, but they are
not therefore "exactly the same".  For some of the reasons why this
point is of considerable importance in the design of a type system,
I refer you to the Ada Rationale (Ada83), Section 4.3.

Re: Why don't large companies use Ada?

[James Kanze US/ESC 60/3/141 #40763]

In article <3blinp$8dm@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert
Dewar) writes:

|> I am happy to see technical points in comparison of languages, but this
|> argument is non-technical and in the long run less relevant than people
|> imagine. Note that the only technical point that Gareth made was that
|> somehow all the strong typing etc. in Ada would make people's life
|> harder, which is of course nonsense, and presumably is nonsense borne
|> of ignorance.

I can't resist pointing it out.  Using strict type-checking *does*
make writing programs harder.  Writing correct programs is difficult.
You actually have to think about what you are doing, and maybe even do
a bit of design before writing, rather than just banging in a bunch of
code, and dropping into the debugger at the first sign of a core dump.
-- 
James Kanze         Tel.: (+33) 88 14 49 00        email: kanze@gabi-soft.fr
GABI Software, Sarl., 8 rue des Francs-Bourgeois, F-67000 Strasbourg, France
Conseils en informatique industrielle --
                              -- Beratung in industrieller Datenverarbeitung

Re: Why don't large companies use Ada?

[Stephen Benson]

 
In article <D03AL2.5xw@aston.ac.uk>, An Amoeba (bromlegb@aston.ac.uk) writes:
>Why should they bother??
>
.
.
.
.
>iii) Support. Virtually every computer system in the world has a C++
compiler, 
>and it is being updated continually. Because that is where all the money is.
>Not every computer system in the world has Ada on it, plus most will not be
>powerful enough to run it. Plus the cost of the respective compilers. C++ 
>compilers, are cheaper than Ada ones. On all if not the majority of platforms.
>

I know little of Ada... but isn't Gnu/gcc about to support it? I think I've
seen news to this effect. Which would make it cheap, widely disseminated and
constantly updated... 



--                                             
: stephen benson  :   :   :   :   :   :   :   :  stephenb@scribendum.win-uk.net

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <589@scribendum.win-uk.net>,
Stephen Benson <stephenb@scribendum.win-uk.net> wrote:

>I know little of Ada... but isn't Gnu/gcc about to support it? I think I've
>seen news to this effect. Which would make it cheap, widely disseminated and
>constantly updated... 

It exists already and is called GNAT; you can ftp GNAT from cs.nyu.edu,
directory pub/gnat. It is integrated into the gcc toolchain. Not quite
finished, but ports exist for SunOS, Solaris, Linux, OS/2, DOS, SGI,
DEC Alpha. It's written in itself, not in C, so porting to new
platforms requires cross-compiling the first version. An adapted
gcc back-end is used. 

As is always the case with GNU software, full source code can be ftp-ed, 
and is fascinating to read. The GNAT team (at NYU, Florida State,
and elsewhere) has done a great job!

At the moment, GNAT supports nearly all of Ada 9X and supports more and
more at each release. Releases occur roughly monthly these days. Try it.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Jules]

In article <1994Nov29.154220.27952@cognos.com>,
	vernestd@cognos.COM (Dave Vernest) writes:
>Michael Feldman (mfeldman@seas.gwu.edu) wrote:
>: In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
>: Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:
>
>: >Since technical merits migth not win the day we should face the realities and
>: >use the other successful spinoffs to give Ada a better image.
>
>: Bingo! And let's keep spreading the stories about where Ada has really
>: succeeded. Make sure everyone knows about the software behind all
>: those CRT's in the airliner cockpit. And in the air-traffic-control system
>: on the ground. And (in France, anyway) in the train system that gets
>: you from the airport to the Riviera at 300 km/h. And the Swiss PTT's
>: electronic-funds-exchange system. And, and...
>
>Not sure if I'm violating the c.p charter or not...
>
>With the use of C++, which supports generics and exception-
>handling, is there any reason to code mission-critical systems
>in Ada anymore?? Advantages / disadvantages??
>
>Dave

A programmer is still much more likely to make a mistake when working in a
language such as C++, which is very highly unreadable, than when working in
a language such as Ada 9x, which is a highle readable language. The main
problems with C++ come from it's C ancestry, and it's unusual constructs
that are difficult to remember (now is the second or third statement in a
for loop the test condition...?, etc).

Okay, maybe I am exaggerating a little. A C++ program will not crash and burn
like a C program could, but there are still many areas where mistakes could
be made, where the increased readability of Ada may save the day. And these
mistakes are likely to be the small mistake that isn't noticed during
testing...

-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Norman H. Cohen]

In article <1994Nov29.154220.27952@cognos.com>, vernestd@cognos.COM
(Dave Vernest) writes: 

|> With the use of C++, which supports generics and exception-
|> handling, is there any reason to code mission-critical systems
|> in Ada anymore?? Advantages / disadvantages??

Just a few reasons that spring to mind quickly: 

validated compilers
multitasking
cross-compilation-unit checks that don't depend on the programmer using
   #include "properly"
version-consistency checks automatically built into the construction of a
   program
a type system strong enough to distinguish between pointers to a T and an
   array of T's, and thus to prevent at compile time mistakes like
   "delete X" when "delete [] X" was meant
a type system strong enough to distinguish between the low-level notion
   of a machine address and the high-level notion of indirect access
a type system strong enough to distinguish among integer types used for
   different purposes
the ability to declare integer data portably, based on expected range
   rather than the expected size of a given predefined type on a given
   machine
language-defined run-time checks, integrated with exception handling
compile-time and run-time checks to prevent dangling pointers
extraordinary difficulty of clobbering memory accidently in units that do
   not explicitly import low-level mechanisms, so that debuggers are
   rarely necessary in debugging high-level Ada programs
EFFICIENT generics, whose implementation is based on a decade and a half
   of lessons learned and the presence of an explicit instantiation
encapsulation of method implementations in a package body
compile-time consistency checks for templates and instantiations
in Ada 94, a standard set of string-manipulation routines that (unlike
   the C-library string routines) free the programmer from low-level
   storage-management concerns
a standard math library
language rules about numerical accuracy (applicable when the target
   machine is not a Pentium :-) )
in Ada 94, strong support for commercial applications, including
   picture-based formatting and exact scaled arithmetic (decimal
   fixed-point types)
explicit control over elaboration order (roughly analogous to invocation
   of C++ constructors for static and extern objects)
a standardized and universally implemented approach to namespace
   management (packages and, in Ada 94, child units)
rigorous control over bit-by-bit layout, when needed
standard mechanisms for interrupt handling
readable notation ("is abstract" vs. "= 0", ...)
coherent notation (if int *a[] means that *a[i] is an int, int &a[]
   should mean that &a[i] is an int, but it means nothing of the kind)
symmetric handling of binary operations and enforced consistency of
   operand tags when appropriate
explicit declaration of variables whose address may be taken,
   facilitating program understanding and code optimization
lots of technical advantages that are minor by themsleves but add up to
   signficantly greater expressiveness and/or reliability, for example: 
   named notation in aggregates and subprogram calls (understood by the
   compiler); aggregates allowed in places other than initial values;
   overloading that can be resolved based on function result type; rules
   for identifiers (case insensitivity, no consecutive underscores) that
   make typographical errors less likely; case statements that don't
   require a break statement after each alternative; if statements that
   don't require braces to be added when an arm is extended from one
   statement to two; distinct opening and closing brackets for different
   constructs that can be checked by the compiler for consistency; end
   brackets required for all loops, thus avoiding traps like

      while(condition);
         statement_to_be_repeated;

   binary integer literals when appropriate; ability to default
   subprogram parameters while still providing values for later
   parameters; etc., etc., etc.

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[Stef Van Vlierberghe]

In article <3bimkh$149q@watnews1.watson.ibm.com> ncohen@watson.ibm.com (Norman H. Cohen) writes:

   Just a few reasons that spring to mind quickly: 

Very impressive list Norman !
Perhaps it could be maintained in the FAQ ?

A few more good reasons for considering Ada :

- Ada has a generic contract model: the implementor and the user
  are isolated by the compiler, each dealing with his own errors.
  Some C++ compilers generate error messages stemming from templates
  or instances when the program is *linked*.

- Classical C(++)-errors that aren't trapped at compile or run time  
  simply have no Ada equivalent. Just a few that I'm familiar with :

    if (a=0) oops_meant_equality;

    while (b /= a) oops_meant_inequality;

    switch (a) 
    { case 1: one; 
      case 2: oops_forgot_the_break;

  (I have the impression that the rate at which I get bitten by these
   doesn't decline over the years, mostly because I'm getting spoiled by
   Ada between occasional C usage, but still I think this is an issue).

--
Stef VAN VLIERBERGHE            Eurocontrol - Central Flow Management Unit
stef@cfmu.eurocontrol.be        Avenue des Arts 19H
Tel: +32 2 729 33 42            B-1040 BRUSSELS
Fax: +32 2 729 32 16            Belgium

Re: Why don't large companies use Ada?

[Geens Ronald]

In article <STEF.94Dec5225406@heron.cfmu.eurocontrol.be>, stef@heron.cfmu.eurocontrol.be (Stef Van Vlierberghe) writes:
|> In article <3bimkh$149q@watnews1.watson.ibm.com> ncohen@watson.ibm.com (Norman H. Cohen) writes:
|> 
|>    Just a few reasons that spring to mind quickly: 
|> 
|> Very impressive list Norman !
|> Perhaps it could be maintained in the FAQ ?
|> 
|> A few more good reasons for considering Ada :
|> 
|> - Ada has a generic contract model: the implementor and the user
|>   are isolated by the compiler, each dealing with his own errors.
|>   Some C++ compilers generate error messages stemming from templates
|>   or instances when the program is *linked*.
|> 
This is true ...

|> - Classical C(++)-errors that aren't trapped at compile or run time  
|>   simply have no Ada equivalent. Just a few that I'm familiar with :
|> 
|>     if (a=0) oops_meant_equality;

This is not true any more, most modern compilers will give a warning when you 
use = in an if.

Ron.

Re: Why don't large companies use Ada?

[Jules]

In article <STEF.94Dec5225406@heron.cfmu.eurocontrol.be>,
	stef@heron.cfmu.eurocontrol.be (Stef Van Vlierberghe) writes:
>In article <3bimkh$149q@watnews1.watson.ibm.com> ncohen@watson.ibm.com (Norman H. Cohen) writes:

>- Classical C(++)-errors that aren't trapped at compile or run time  
>  simply have no Ada equivalent. Just a few that I'm familiar with :
>
>    if (a=0) oops_meant_equality;
>
>    while (b /= a) oops_meant_inequality;

Many compilers actually issue warnings in these cases, they are quite easy
to trap. Admittedly, the warnings can get irritating if you want to do
an assignment during an if statement.
>
>    switch (a) 
>    { case 1: one; 
>      case 2: oops_forgot_the_break;
Again, warnings can be quite easily produced in this situation.

>  (I have the impression that the rate at which I get bitten by these
>   doesn't decline over the years, mostly because I'm getting spoiled by
>   Ada between occasional C usage, but still I think this is an issue).

I must admit, these are very frequent errors for many C programmers (myself
included), especially once one has been using another language for a little
while.

Other frequent errors are caused by C's unusual syntax. Things that are
hard to remember (such as the order of the arguments passed to main() -
I always get them the wrong way around first time when I'm starting a new
project) often just cause plain and simple irritation.



-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Tucker Taft]

In article <1994Nov29.154220.27952@cognos.com>,
Dave Vernest <vernestd@cognos.COM> wrote:
>Michael Feldman (mfeldman@seas.gwu.edu) wrote:
>: In article <TARJEIJ.94Nov28084714@ulrik.uio.no>,
>: Tarjei Jensen <tarjeij@ulrik.uio.no> wrote:
>
>: >Since technical merits migth not win the day we should face the realities and
>: >use the other successful spinoffs to give Ada a better image.
>
>: Bingo! And let's keep spreading the stories about where Ada has really
>: succeeded. Make sure everyone knows about the software behind all
>: those CRT's in the airliner cockpit. And in the air-traffic-control system
>: on the ground. And (in France, anyway) in the train system that gets
>: you from the airport to the Riviera at 300 km/h. And the Swiss PTT's
>: electronic-funds-exchange system. And, and...
>
>Not sure if I'm violating the c.p charter or not...
>
>With the use of C++, which supports generics and exception-
>handling, is there any reason to code mission-critical systems
>in Ada anymore?? Advantages / disadvantages??

Although C++ and Ada 9X have very similar functionality, they
have quite different philosophies relative to safety.

In Ada 9X, all normal language features are by default safe, using a
combination of compile-time and run-time consistency checking;
you can circumvent checking if you need to in specific circumstances
using explicitly "unchecked" programming features.

In C++, largely because of its C heritage, many "normal" language features
are by default unsafe (e.g. pointer indirection, array indexing,
numeric calculations, casting), but if you work at it, you can create
and then restrict yourself to safe abstractions (e.g. safe pointers,
checked arrays, overflow-checking arithmetic, checked casts).

I realize there are advantages and disadvantages to both
the default safe and the default unsafe approaches. 
When trying to verify a mission critical system, the net advantage
seems to go to the default safe approach.  In Ada 9X, the places 
where consistency checks are being circumvented are quite visible, 
and easily localizable.  In C++, one must look at essentially every 
line of the program to check whether one of the potentially unsafe 
features of the language is being used in an unsafe way.

>Dave

S. Tucker Taft   stt@inmet.com
Intermetrics, Inc.

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <house.786178243@helios>, ron house <house@helios.usq.EDU.AU> wrote:
>mfeldman@seas.gwu.edu (Michael Feldman) writes:
>
>>And not only that, it'll run on a PC. And not only that, GNAT is _free_.
>
>>How about "Ada - use it when your software really has to WORK."
>
>How about "Ada - use it when you want to multiply a length by a length
>and get a length, or when you want to be prevented from dividing an area
>by a length because they have different types."

This is NOT my favorite aspect of the Ada type system, and I seriously 
wish the language directly supported dimensions. As it is, I try to 
avoid using derived numeric types, preferring to stick to subtypes 
(subrange types) which give me the necessary range checking but don't 
get in the way otherwise.

>(PS: yes, you can fix that by defining zillions of additional *,/,- and +
>functions, but good grief...)

Well, there are also other solutions. One such is to use a variant
record with 3 tags for the physical dimensions. This makes up into
a very simple package and is bulletproof in the sense that mangled
dimension-analysis will get caught either at compile time or (at worst)
at run time with a nicely raised exception. My CS2-level students
use and modify this package in their class projects.

It does require a bit of runtime overhead, i.e. space for the tags
and a bit of runtime checking, and I agree that it would be
preferable to do this at compile time. But variant records are
done so type-safely in Ada that this very simple solution works well.

(My guess is that the runtime overhead is no more than is normally
tolerated by folks who like dynamic dispatching.)

I'll send the code by e-mail to anyone who wants it.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Matt Kennel]

Michael Feldman (mfeldman@seas.gwu.edu) wrote:
: >How about "Ada - use it when you want to multiply a length by a length
: >and get a length, or when you want to be prevented from dividing an area
: >by a length because they have different types."

: This is NOT my favorite aspect of the Ada type system, and I seriously 
: wish the language directly supported dimensions. As it is, I try to 
: avoid using derived numeric types, preferring to stick to subtypes 
: (subrange types) which give me the necessary range checking but don't 
: get in the way otherwise.

: >(PS: yes, you can fix that by defining zillions of additional *,/,- and +
: >functions, but good grief...)

: Well, there are also other solutions. One such is to use a variant
: record with 3 tags for the physical dimensions. This makes up into
: a very simple package and is bulletproof in the sense that mangled
: dimension-analysis will get caught either at compile time or (at worst)
: at run time with a nicely raised exception.

Does these include lepton number?  Charge?  :-)

Atomic concentrations?  Temperature?  (Do you keep that separate
or consider it as energy units through boltzmann's constant?  Usually
that's wrong, but sometimes right?)   Units of electric field?

What system of lengths/time/masses? 

: My CS2-level students
: use and modify this package in their class projects.

: It does require a bit of runtime overhead, i.e. space for the tags
: and a bit of runtime checking, and I agree that it would be
: preferable to do this at compile time. But variant records are
: done so type-safely in Ada that this very simple solution works well.

I think it's a nice student project, but in real physical computations, the
right way is often to go into a system of units that is natural for your
problem, and work with these nondimensional variables.  

You usually gain more insight into the important terms in your computations
and you don't end up with insanely small or insanely large exponents (no
overflow) unless things are physically negligable or dominant.

cheers
matt

: Mike Feldman

--
-Matt Kennel  		mbk@inls1.ucsd.edu
-Institute for Nonlinear Science, University of California, San Diego
-*** AD: Archive for nonlinear dynamics papers & programs: FTP to
-***     lyapunov.ucsd.edu, username "anonymous".

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3blc1s$8gr@network.ucsd.edu>,
Matt Kennel <mbk@inls1.ucsd.edu> wrote:

>Does these include lepton number?  Charge?  :-)

No, but could. My little package is only length/mass/time, but could
be extended to the fourth charge dimension.
>
>Atomic concentrations?  Temperature?  (Do you keep that separate
>or consider it as energy units through boltzmann's constant?  Usually
>that's wrong, but sometimes right?)   Units of electric field?
>
>What system of lengths/time/masses? 

Could be adapted to MKS, CGS, or both, with conversion constants.
It's really intended to show the technique, not to be exhaustive.
>
>I think it's a nice student project, but in real physical computations, the
>right way is often to go into a system of units that is natural for your
>problem, and work with these nondimensional variables.  

Well, this is a matter of taste, I think. That my CS2 classes can handle
it shows that it's conceptually simple. Lots of engineers have seen it
and said they could use it in their work, with some "localization".

BTW - the variant-record solution is not at all original with me;
it's been sketched out several times in the literature. I take credit
only for boiling it down and making a pretty complete package out of
it, so that it shows how to handle dimensions and also how Ada does
safe variant records.

Your "nondimensional" variables are fine, and in Ada would be done,
presumably, by just creating a bunch of nicknames for Float (say):

  SUBTYPE Length IS Float;
  SUBTYPE Mass IS Float;

etc. I have no problem with this approach. Recall that the thread
started with a guy who said Ada's type system gets in the way of
declaring _new_ or _derived_ types to handle units, e.g.

  TYPE Length IS DIGITS 6;
  TYPE Mass iS DIGITS 6;

He correctly pointed to the need to define "zillions" of operators to
cover meaningful expressions using this style. I simply responded that
there are alternative "bulletproof" ways of handling dimensions.

>You usually gain more insight into the important terms in your computations
>and you don't end up with insanely small or insanely large exponents (no
>overflow) unless things are physically negligable or dominant.

I quite agree. What leads us to discuss dimensioned variables is the
number of "urban myths" telling horror tales of botched rocket launches
and the like, caused by mangled dimension analysis. Obviously with great
care one can avoid mangling the dimension analysis. :-)

Mike Feldman

Re: Why don't large companies use Ada?

[gamache]

In article <3btf4n$jma@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
> In article <3blc1s$8gr@network.ucsd.edu>,
> Matt Kennel <mbk@inls1.ucsd.edu> wrote:
> 
[snip]
>>
>>What system of lengths/time/masses? 
> 
> Could be adapted to MKS, CGS, or both, with conversion constants.
> It's really intended to show the technique, not to be exhaustive.
>>

[more snip]

> BTW - the variant-record solution is not at all original with me;
> it's been sketched out several times in the literature. I take credit
> only for boiling it down and making a pretty complete package out of
> it, so that it shows how to handle dimensions and also how Ada does
> safe variant records.
> 
> Your "nondimensional" variables are fine, and in Ada would be done,
> presumably, by just creating a bunch of nicknames for Float (say):
> 
>   SUBTYPE Length IS Float;
>   SUBTYPE Mass IS Float;
> 
> etc. I have no problem with this approach. Recall that the thread
> started with a guy who said Ada's type system gets in the way of
> declaring _new_ or _derived_ types to handle units, e.g.
> 
>   TYPE Length IS DIGITS 6;
>   TYPE Mass iS DIGITS 6;
> 
This seemingly innocuous example hits one of my hot buttons. 

[climb soapbox]

So much damage can be done with a bad example it never ceases to amaze me.  
(ok, so this is why I'd rather mostly read CLA than post...!)  I've always
thought that K&R did so much harm to this profession with the examples it
contained.  I don't know about anyone else, but I still see people who think
it is acceptable to name file pointers "fp".

How does this relate to the above?  Well, I always view user defined types as
not only a way to provide contraints, but also as a means to provide 
*information* to some future reader of the code (which could possibly even
be the author his/her self).  This example, which *some* people will think
is an okay way to code fails to capitalize on this opportunity to convey
some information.  For instance, object definitions might look like:

	Obj1_Mass	: Mass := 0.001;
	Obj2_Mass	: Mass := 3.5;
the type definition fails to provide readers with any useful information.  This
is particularly noticable when people use _type on type definitions.  Then you
frequently see:

	Mass	: Mass_Type;

Few such simple issues get to me as much.  I even wish compilers would disallow
this construct altogther.  What would a future reader like to
know?  Is mass in kilograms, grams, milligrams?  Is length feet, meters, miles?
Code could then look like:

	Obj1_Mass	: Grams := 0.001;

which conveys much different information than:

	Obj2_Mass	: Kilograms := 0.001;

IMO, the distinction between this and the above is important.  Unfortunately, 
too much code that I see uses types such as length, mass etc.
[ok, climb down from soapbox...]

-------------------------------------------------------------------------------
Joe Gamache
Lockheed Sanders Inc.
with all_disclaimers_standard_or_otherwise;

 _______  ______     /     |     |         /     ___        /
 |     |  | ___|    / \    |\   /|        / \    |   \     / \
 --   --  | |      / / \   | \ / |       / / \   | |\ \   / / \
   | |    | --    / / \ \  |  \  |      / / \ \  | | \ | / / \ \
   | |    | --    | --- |  | \ / |      | --- |  | | / / | --- |
   | |    | |     | --- |  | |\| |      | --- |  | |/ /  | --- |
   | |    | ---|  | | | |  | | | |      | | | |  |   /   | | | |
   |_|    |____|  |_| |_|  |_| |_|      |_| |_|  |_/     |_| |_|

Re: Why don't large companies use Ada?

[Peter Hermann]

gamache@rapnet.sanders.lockheed.com wrote:
[snip]
: is particularly noticable when people use _type on type definitions.
: Then you frequently see:

: 	Mass	: Mass_Type;

this is very useful to identify an identifier as a type.

: Few such simple issues get to me as much.
: I even wish compilers would disallow
: this construct altogther.

disagree

: What would a future reader like to
: know?  Is mass in kilograms, grams, milligrams?
: Is length feet, meters, miles?
: Code could then look like:

: 	Obj1_Mass	: Grams := 0.001;

: which conveys much different information than:

: 	Obj2_Mass	: Kilograms := 0.001;

: IMO, the distinction between this and the above is important.
: Unfortunately, 
: too much code that I see uses types such as length, mass etc.
: [ok, climb down from soapbox...]

You are intermingling two different things:
One is the typing of variables, the other is the accompanying
(physical) dimension of the type.
Ada did not go that far to attach dimensions to the type.
But you can, as we discussed a couple of weeks ago, 
enforce those extras by means of programed extra mechanisms
such as private types (example ref: "Ada in Action" by Do-While Jones)
thereby enforcing the problem abstraction in the following way:
   example: 
		 area := length * length ;    -- "*" allowed
		 leng := length * length ;    -- "*" not allowed!

: -------------------------------------------------------------------------------
: Joe Gamache
: Lockheed Sanders Inc.
: with all_disclaimers_standard_or_otherwise;

:  _______  ______     /     |     |         /     ___        /
:  |     |  | ___|    / \    |\   /|        / \    |   \     / \
:  --   --  | |      / / \   | \ / |       / / \   | |\ \   / / \
:    | |    | --    / / \ \  |  \  |      / / \ \  | | \ | / / \ \
:    | |    | --    | --- |  | \ / |      | --- |  | | / / | --- |
:    | |    | |     | --- |  | |\| |      | --- |  | |/ /  | --- |
:    | |    | ---|  | | | |  | | | |      | | | |  |   /   | | | |
:    |_|    |____|  |_| |_|  |_| |_|      |_| |_|  |_/     |_| |_|

2 objections:    -- a friendly Ada Teamer would not waste bandwidth
		 -- it's Ada (not ADA  ;-):wq

--
Peter Hermann  Tel:+49-711-685-3611 Fax:3758 ph@csv.ica.uni-stuttgart.de
Pfaffenwaldring 27, 70569 Stuttgart Uni Computeranwendungen
Team Ada: "C'mon people let the world begin" (Paul McCartney)

Re: Why don't large companies use Ada?

[Paul Johnson]

Michael Feldman (mfeldman@seas.gwu.edu) wrote:
> In article <house.786178243@helios>, ron house <house@helios.usq.EDU.AU> wrote:

>> How about "Ada - use it when you want to multiply a length by a length
>> and get a length, or when you want to be prevented from dividing an area> 
>> by a length because they have different types.

> This is NOT my favorite aspect of the Ada type system, and I seriously 
> wish the language directly supported dimensions. As it is, I try to 
> avoid using derived numeric types, preferring to stick to subtypes 
> (subrange types) which give me the necessary range checking but don't 
> get in the way otherwise.

> >(PS: yes, you can fix that by defining zillions of additional *,/,- and +
> >functions, but good grief...)

I worked on an Ada project that tried to do exactly that.  We had
huge files of definitions, but the type system still kept getting in
the way.

We were also using a library versioning system, and of course this
basic stuff was in a library.  So when you changed it, everything else
had to be recompiled.  What with all the other book-keeping associated with
making a change, it held up the project for a full day.

We wound up repeatedly defining our own operators in code blocks
wherever they were needed.  This had several bad consequences

1: The source code become larger and less readable.

2: Type checking (the whole point of all this pain) was weakened
   because people were just defining operations ad-hoc.

3: We could not add the necessary ops to the library later on because
   they then conflicted with the ad-hoc declarations and the compiler
   complained about ambiguities.

Paul.

-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Re: Why don't large companies use Ada?

[Robert I. Eachus]

In article <5994@gec-mrc.co.uk> paj@gec-mrc.co.uk (Paul Johnson) writes:

 > We were also using a library versioning system, and of course this
 > basic stuff was in a library.  So when you changed it, everything
 > else had to be recompiled.  What with all the other book-keeping
 > associated with making a change, it held up the project for a full
 > day.

   This is Ada trying to tell you something, learn to listen.  When,
during development, you find yourself having to recompile the world,
it means that something your design regards as set in concrete isn't.
Even if you can get through development, the maintainers are going to
have to live with these design decisions for a long time.  Pity the
poor maintainer, who finds that the definition of "*" for length is
wrong for his part of the code--perhaps due to a requirements change.
How does he "fix" it?  Should he change the global definition, hoping
not to break something else, or should he add a local definition which
hides the global definition and creates a maintenance nightmare?

   How do you fix the problem?  Fix the design, and you start by
finding which design assumptions are invalid.  In this case it seems
to be that you can know all the possible combinations of dimensions
that will be needed.  The approach Mike uses, and which has been
discussed here before works, and has a lot fewer drawbacks, but it
still has some.  (Doing everything in Float has a different, and often
more serious set of drawbacks.)

   You have to compromise, and the compromise selected becomes part of
your design.  If you end up doing lots of global recompiles, the
compiler is telling you that that decision is wrong, and there will
never be a better time to fix it.


--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Design problems (was Re: Why don't large companies use Ada?)

[Paul Johnson]

Robert I. Eachus (eachus@spectre.mitre.org) wrote:
> In article <5994@gec-mrc.co.uk> paj@gec-mrc.co.uk (Paul Johnson) writes:

>  > We were also using a library versioning system, and of course this
>  > basic stuff was in a library.  So when you changed it, everything
>  > else had to be recompiled.  What with all the other book-keeping
>  > associated with making a change, it held up the project for a full
>  > day.

>    This is Ada trying to tell you something, learn to listen.

Yeah.  I know.  But I was working as contract labour on that project
so my opinions counted for zilch.

This wasn't just the dimension libraries.  All the data structure
libraries suffered from this.  The analysis document contained Data
Flow Diagrams and the bubbles communicated through types defined in a
Data Dictionary.  So all the data types were implemented in low-level
library packages for access by the procedures that implemented the
bubbles.  Every time a data type needed to be added or modified, a
low-level package had to be checked out, and this triggered a
recompile of everything above.  C and C++ suffer from similar problems
in "include" files.

If they had taken an "object based" approach to low level design
(packaging up data structures with the relevant operations) then
things would have been different.  But structured analysis doesn't
group things like that, so they didn't either.

To be fair, the management knew this was a problem as well, but they
were well into the lifecycle by then (this was the Version 2 release),
so the design was pretty much cast in concrete.

We were also working with two compilers (development platform and
target platform), one of which had a bug that caused it to reject the
dimension libraries.  So one version was coded with subtypes and one
with ranges.  Most of the time we managed to produce code that worked
with both, but not always.  So then some arbitrary module halfway up
the hierarchy had to have a version forked off and separately
maintained.  We had a paper-based change-control system running, and
we certainly needed it.

This was the first Ada project by this team.  They commented that
next time they used Ada they would know how to do it right.  But by
then something else would be flavour-of-the-year, and they would be
back at the bottom of the learning curve again.

I will say that, despite the problems, we produced code on-time and
on-budget.  It was expensive code, but we knew ahead of time how much
it would cost.  Code was reviewed before being checked back into the
development version, so the bug rate was fairly low.  It was just that
some bugs were very expensive.

Paul.

-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Re: Why don't large companies use Ada?

[Stephen J Bevan]

In article <5994@gec-mrc.co.uk> paj@gec-mrc.co.uk (Paul Johnson) writes:
   ... I worked on an Ada project that tried to do exactly that.  We had
   huge files of definitions, but the type system still kept getting in
   the way. ...

At the end of the project was it acknowledged that the design was
(probably) wrong or was Ada blamed for the problem?

Re: Why don't large companies use Ada?

[Paul Johnson]

Stephen J Bevan (bevan@cs.man.ac.uk) wrote:
> In article <5994@gec-mrc.co.uk> paj@gec-mrc.co.uk (Paul Johnson) writes:
>    ... I worked on an Ada project that tried to do exactly that.  We had
>    huge files of definitions, but the type system still kept getting in
>    the way. ...

> At the end of the project was it acknowledged that the design was
> (probably) wrong or was Ada blamed for the problem?

I left before the end of the project (contract labour).  The people I
was with agreed that it might have been better if the design had been
different.

Paul.

-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Re: Why don't large companies use Ada?

[Richard Riehle]

In article <house.786178243@helios> house@helios.usq.EDU.AU (ron house) writes:
>
>How about "Ada - use it when you want to multiply a length by a length
>and get a length, or when you want to be prevented from dividing an area
>by a length because they have different types."
>
This is a most peculiar example in this era of enlightened parsimonious
typing.

A mistake common to novice Ada designers is to promicuously define types
for every possible variable.  Prudent use of the typing mechansm will
serve the intelligent designer.  

That being said, even if the types are different, one can do type conversion
on numeric types. 

Now, if the writer is referring to the potential for constraint errors
from intermediate results in long expressions of mixed type, there is
a real issue. Even this issue is not a cause for alarm, and is trivially
solved by experienced Ada developers.

Richard Riehle
AdaWorks Software Engineering
Palo Alto, CA
adaworks@netcom.com

Re: Why don't large companies use Ada?

[Michael Coburn]

house@helios.usq.EDU.AU (ron house) writes:

>How about "Ada - use it when you want to multiply a length by a length
>and get a length, or when you want to be prevented from dividing an area
>by a length because they have different types."

total_area := area(length_one) * area(length_two);

yawn....

>(PS: yes, you can fix that by defining zillions of additional *,/,- and +
>functions, but good grief...)

Do you enjoy doing things the hard way?

:-)
--
|/-\| Michael Coburn              email: s9409498@yallara.cs.rmit.oz.au |/-\|
|\-/|                                                                   |\-/|

Re: Why don't large companies use Ada?

[Bob Duff]

In article <3bjf2q$i69@goanna.cs.rmit.oz.au>,
Michael Coburn <s9409498@yallara.cs.rmit.OZ.AU> wrote:
>house@helios.usq.EDU.AU (ron house) writes:
>
>>How about "Ada - use it when you want to multiply a length by a length
>>and get a length, or when you want to be prevented from dividing an area
>>by a length because they have different types."
>
>total_area := area(length_one) * area(length_two);
>
>yawn....

The fact remains that Ada's type model disallows a sensible operation
(Total_Area := Length_One * Length_Two), and allows nonsense
(Some_Length := Length_One * Length_Two).

- Bob
-- 
Bob Duff                                bobduff@inmet.com
Oak Tree Software, Inc.
Ada 9X Mapping/Revision Team (Intermetrics, Inc.)

Re: Why don't large companies use Ada?

[Jules]

In article <D07JDD.H25@inmet.camb.inmet.com>,
	bobduff@dsd.camb.inmet.com (Bob Duff) writes:
>In article <3bjf2q$i69@goanna.cs.rmit.oz.au>,
>Michael Coburn <s9409498@yallara.cs.rmit.OZ.AU> wrote:
>>house@helios.usq.EDU.AU (ron house) writes:
>>
>>>How about "Ada - use it when you want to multiply a length by a length
>>>and get a length, or when you want to be prevented from dividing an area
>>>by a length because they have different types."
>>
>>total_area := area(length_one) * area(length_two);
>>
>>yawn....
>
>The fact remains that Ada's type model disallows a sensible operation
>(Total_Area := Length_One * Length_Two), and allows nonsense
>(Some_Length := Length_One * Length_Two).

Only if you want to go completely over the top and define new types for lengths
and areas. I mean, if you did the same in C++ you would get the same results.
So why criticise Ada, for something which almost all commonly used languages
today have a problem with, if you use them in this (unusual) manner?
-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Bob Duff]

In article <3bvadv$66p@holly.csv.warwick.ac.uk>,
Jules <csusb@csv.warwick.ac.uk> wrote:
>In article <D07JDD.H25@inmet.camb.inmet.com>,
>	bobduff@dsd.camb.inmet.com (Bob Duff) writes:
>>In article <3bjf2q$i69@goanna.cs.rmit.oz.au>,
>>Michael Coburn <s9409498@yallara.cs.rmit.OZ.AU> wrote:
>>>house@helios.usq.EDU.AU (ron house) writes:
>>>
>>>>How about "Ada - use it when you want to multiply a length by a length
>>>>and get a length, or when you want to be prevented from dividing an area
>>>>by a length because they have different types."
>>>
>>>total_area := area(length_one) * area(length_two);
>>>
>>>yawn....
>>
>>The fact remains that Ada's type model disallows a sensible operation
>>(Total_Area := Length_One * Length_Two), and allows nonsense
>>(Some_Length := Length_One * Length_Two).
>
>Only if you want to go completely over the top and define new types for lengths
>and areas.

But if you *don't* go over the top in that manner, then Ada allows
*both* of the above (one sensible, the other nonsense).  Neither usage
of the type system gives "perfect" results -- i.e. allow all sensible
operations while disallowing all nonsensical operations.  That's what
type checking is *for*, IMHO, and here's a case where it doesn't quite
work right.  There's no point in pretending that Ada's type system is
perfect -- probably none is at this early point in the history of
Programming Language Design.

> ...I mean, if you did the same in C++ you would get the same results.

Umm, I'm not sure how you do "the same" in C++, since one doesn't
generally define new integer types in C++.  But I agree that C++ is no
better than Ada in this respect.

>So why criticise Ada, for something which almost all commonly used languages
>today have a problem with, if you use them in this (unusual) manner?

I don't think honestly admitting to a minor flaw in Ada's type system is
such a bad thing.  I'm a language designer, and I like to think about
how one might do better.

- Bob
-- 
Bob Duff                                bobduff@inmet.com
Oak Tree Software, Inc.
Ada 9X Mapping/Revision Team (Intermetrics, Inc.)

Physical Types (was Why don't large companies use Ada?)

[Mike Chapman]

Jules (csusb@csv.warwick.ac.uk) wrote:
: In article <D07JDD.H25@inmet.camb.inmet.com>,
: 	bobduff@dsd.camb.inmet.com (Bob Duff) writes:
: >In article <3bjf2q$i69@goanna.cs.rmit.oz.au>,
: >Michael Coburn <s9409498@yallara.cs.rmit.OZ.AU> wrote:
: >>house@helios.usq.EDU.AU (ron house) writes:
: >>
: >>>How about "Ada - use it when you want to multiply a length by a length
: >>>and get a length, or when you want to be prevented from dividing an area
: >>>by a length because they have different types."
: >>
: >>total_area := area(length_one) * area(length_two);
: >>
: >>yawn....
: >
: >The fact remains that Ada's type model disallows a sensible operation
: >(Total_Area := Length_One * Length_Two), and allows nonsense
: >(Some_Length := Length_One * Length_Two).

: Only if you want to go completely over the top and define new types for lengths
: and areas. I mean, if you did the same in C++ you would get the same results.
: So why criticise Ada, for something which almost all commonly used languages
: today have a problem with, if you use them in this (unusual) manner?

There is a language called VHDL which was developed for
modelling integrated circuits. It is based on Ada. You can
actually write code which would compile and run with Ada as
well as compile and simulate (you don't run VHDL code!) in 
VHDL.

In this language you can define 'physical units'
For instance,

    type distance is range -1E18 to 1E18
       units
          -- Base unit
          A;            -- angstrom
          nm = 10 A;    -- nanometer
          um = 1000 nm; -- micron
          mm = 1000 um; -- millimeter
          cm = 10 mm;   -- centimeter
          m  = 1000 mm; -- meter
          km = 1000 m;  -- kilometer
       end units;

     variable x: distance; 
     variable z: integer;

     x := 5 A + 13 m - 25 um;
     x := 4 * x;

     z := x / mm;
     

However  x * x is undefined.
You can of course overload it if you have defined a
physical type area in the obvious way....

     function "*" (l,r: in distance) return area is
     begin
         return (l/A)*(r/A)*A2;
     end "*";

where A2 are square angstroms defined in a physical area
type.

Similiarly

     function "/" (l: in area; r: in distance) return distance is
     begin
         return ((l/A2)/(r/A))*A;
     end "/";

     
I am surprised something like this has not been added into
the Ada 9x proposals.

-- 
Mike Chapman <Mike.Chapman@muc.de>
    As far as the laws of mathematics refer to reality, they are 
    not certain; and as far as they are certain, they do not refer 
    to reality.  -- Albert Einstein --

Re: Physical Types (was Why don't large companies use Ada?)

[John Barton]

In article <1994Dec6.201030.366@isis.muc.de>, Mike.Chapman@muc.de (Mike Chapman) writes:
|> Jules (csusb@csv.warwick.ac.uk) wrote:
|> : In article <D07JDD.H25@inmet.camb.inmet.com>,
|> : 	bobduff@dsd.camb.inmet.com (Bob Duff) writes:
|> : >In article <3bjf2q$i69@goanna.cs.rmit.oz.au>,
|> : >Michael Coburn <s9409498@yallara.cs.rmit.OZ.AU> wrote:
|> : >>house@helios.usq.EDU.AU (ron house) writes:
|> : >>
|> : >>>How about "Ada - use it when you want to multiply a length by a length
|> : >>>and get a length, or when you want to be prevented from dividing an area
|> : >>>by a length because they have different types."
|> : >>
|> : >>total_area := area(length_one) * area(length_two);
|> : >>
|> : >>yawn....
|> : >
|> : >The fact remains that Ada's type model disallows a sensible operation
|> : >(Total_Area := Length_One * Length_Two), and allows nonsense
|> : >(Some_Length := Length_One * Length_Two).
|> 
|> : Only if you want to go completely over the top and define new types for lengths
|> : and areas. I mean, if you did the same in C++ you would get the same results.
|> : So why criticise Ada, for something which almost all commonly used languages
|> : today have a problem with, if you use them in this (unusual) manner?
|> 
|> There is a language called VHDL which was developed for
|> modelling integrated circuits. It is based on Ada....
[summary of VHDL version omitted]

Two references to C++ solutions to this problem are
  Zerksis D. Umrigar "Fully Static Dimensional Analysis with C++" ACM SIGPLAN
    Notices 29 (9) Sep. 1994 pg 135.
  John J. Barton and Lee R. Nackman, "Scientific and Engineering C++: Dimensional
    Analysis", C++ Report, next issue.  
-----
John.

John J. Barton        jjb@watson.ibm.com            (914)784-6645
H1-C13 IBM Watson Research Center P.O. Box 704 Hawthorne NY 10598

Re: Why don't large companies use Ada?

[Robert Dewar]

Ron, you don't need zillions of declarations. if you want to make
an operation unavailable, then just make it abstract.

basically there are only two possibilities for this sort of thing. specificlly
define all the operators you need, or by default define all of them, and then
specifically remove the ones you don't want. Ada 9X allows both approaches.

There are various approaches to more automated approaches to the units
problem, which have been discussed on the net in great detail, but many
of the contributors to this discussion (including me) felt that it was
overkill, and that in practice the basic mechanisms of defining the operators
you need (using one of the above two approaches) is sufficient.

Really the only step you could go further is to specifically add units
support to the language, but for most people that is feature overkill.

IN practice, I find the strong type protection of different integer types
to be of minimal value, strong typing is most important at a higher level
than that, but that's a matter of personal taste.

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3bldlu$84r@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:

>Really the only step you could go further is to specifically add units
>support to the language, but for most people that is feature overkill.

Yeah, we lost the battle on that one.

>IN practice, I find the strong type protection of different integer types
>to be of minimal value, strong typing is most important at a higher level
>than that, but that's a matter of personal taste.

In general I agree, and find that subtypes give the needed range checks
without getting the programmer into type-compatibility stew, but there
are these persistent "urban myths" about Fortran codes whose mangled
dimension analysis sent space probes off the wrong way, etc.

Mike Feldman

Re: Strong numeric type checking

[Tucker Taft]

In article <3bo402$23a@felix.seas.gwu.edu>,
Michael Feldman <mfeldman@seas.gwu.edu> wrote:

>In article <3bldlu$84r@gnat.cs.nyu.edu>, 
>Robert Dewar <dewar@cs.nyu.edu> wrote:
>
>>Really the only step you could go further is to specifically add units
>>support to the language, but for most people that is feature overkill.
>
>Yeah, we lost the battle on that one.

See the note by do-while jones on this one.

Actually one possibility we did consider seriously was allowing
any two floating point types to be multiplied together without requiring
explicit conversions.  This would have eliminated strong type checking 
for multiplying operators, while retaining it for adding and subtracting.  
That might have been a reasonable compromise, but there were some 
thorny ambiguity and upward-compatibility problems.  Note that this 
route was taken for fixed-point, to simplify use of the new "decimal" 
fixed-point types.

>>IN practice, I find the strong type protection of different integer types
>>to be of minimal value, strong typing is most important at a higher level
>>than that, but that's a matter of personal taste.
>
>In general I agree, and find that subtypes give the needed range checks
>without getting the programmer into type-compatibility stew, but there
>are these persistent "urban myths" about Fortran codes whose mangled
>dimension analysis sent space probes off the wrong way, etc.

I rarely disagree with the Professor, but if you are primarily
using the integer types as indices into various arrays, then strong
type checking is quite useful to help make sure you don't use an apple
number to index into a crate of oranges.  For the average systems
programmer, integers are probably used much more for counting, adding,
and indexing, than they are used for multiplying.

If you are primarily doing arithmetic (beyond just adding and 
subtracting), and you don't feel the urge to define the appropriate 
cross-product operators, then using subtypes rather than distinct 
types makes the most sense.

Note that in Ada 9X, you can effectively "disable" undesired multiplying
operators by overriding them with an abstract function definition.
E.g.:
    function "*"(Left, Right : Meters) return Meters is abstract;

This eliminates the need to declare the types as private just to
get rid of unwanted operators.

>Mike Feldman

-Tucker Taft   stt@inmet.com

P.S. For those who have not already seen similar offerings,
here is a sampling of some generics that could make it easier 
to define interesting dimensioned types, and associated cross-product 
operators, if the strong type checking is felt to be worth the effort
in some given application.

   generic
       type Left_Type is digits <>;
       type Right_Type is digits <>;
       type Result_Type is digits <>;
   function Product(Left : Left_Type; Right : Right_Type) return Result_Type;
   pragma Inline(Product);

   function Product(Left : Left_Type; Right : Right_Type) return Result_Type is
   begin
        return Result_Type(Left) * Result_Type(Right);
   end Product;

We could also define a package with a dimensionless floating point
type (which would have all the multiplying operations) plus
a root type with the multiplying operators made abstract, but
with ones with a dimensionless operand one added back, such as:

   with Product, Quotient;
   generic
      type Impl_Type is digits <>;  
        -- Impl Type should be Float/Long_Float/etc.
   package Generic_Units is
      type Dimensionless is new Impl_Type;
      type Dimensioned is new Impl_Type;

      /* First "disable" the undesirable operations */
      function "*"(Left, Right : Dimensioned) return Dimensioned is abstract;
      function "/"(Left, Right : Dimensioned) return Dimensioned is abstract;
      function "**"(Left : Dimensioned; Right : Integer) 
        return Dimensioned is abstract;

      /* Now define some additional ones */
      function "*" is new Product(Dimensioned, Dimensionless, Dimensioned);
      function "*" is new Product(Dimensionless, Dimensioned, Dimensioned);

      function "/" is new Quotient(Dimensioned, Dimensionless, Dimensioned);
      function "/" is new Quotient(Dimensioned, Dimensioned, Dimensionless);
   end Generic_Units;

After instantiating this with the appropriate implementation type,
subtypes of Dimensionless could be used for dimensionless types,
and derivatives of Dimensioned could be used for each dimensioned type,
followed by any additional multiplying operators needed for the application.
E.g.:
    with Generic_Units;
    package Units is new Generic_Units(Float);

    with Generic_Units;
    package Long_Units is new Generic_Units(Long_Float);
    
    with Long_Units; use Long_Units;
    with Product, Quotient;
    package CGS is
	subtype Ratio is Dimensionless;
        type Cm is new Dimensioned;
        type Gm is new Dimensioned;
        type Sec is new Dimensioned;
	type Cm_Per_Sec is new Dimensioned;
	function "/" is new Quotient(Cm, Sec, Cm_Per_Sec);
	function "*" is new Product(Cm_Per_Sec, Sec, Cm);
	function "*" is new Product(Sec, Cm_Per_Sec, Cm);
	...
    end CGS;

-TT

Re: Why don't large companies use Ada?

[David Weller]

Newsgroup trimmed to comp.lang.ada only.  Seems appropriate.

In article <3bo402$23a@felix.seas.gwu.edu>,
Michael Feldman <mfeldman@seas.gwu.edu> wrote:
>In article <3bldlu$84r@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>
>>IN practice, I find the strong type protection of different integer types
>>to be of minimal value, strong typing is most important at a higher level
>>than that, but that's a matter of personal taste.
>
>In general I agree, and find that subtypes give the needed range checks
>without getting the programmer into type-compatibility stew, but there
>are these persistent "urban myths" about Fortran codes whose mangled
>dimension analysis sent space probes off the wrong way, etc.
>

In general, my company uses a stepwise sense of typing.  The "higher"
up in the abstraction level, the more you're expected to use strong
typing.  Indeed, many algorithms we have at the very low level are
littered with Float32 and Int32.  Why?  Because in many cases they're
lifted from some modeling algorithm that is known to work (typically
specified in FORTRAN).  This isn't to say we apply this in a
haphazard way, it's just that we become less stringent on typing as
the "interface" level drops to 0 users.


-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
	|"Quitting C++ isn't so difficult, provided you show as much |
	|	persistence stopping as you did starting." dweller   |

Re: Why don't large companies use Ada?

[Norman H. Cohen]

In article <3bo402$23a@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu
(Michael Feldman) writes: 

|> In general I agree, and find that subtypes give the needed range checks
|> without getting the programmer into type-compatibility stew, but there
|> are these persistent "urban myths" about Fortran codes whose mangled
|> dimension analysis sent space probes off the wrong way, etc.

The urban myth about the space probe is not concerned with strong typing
of data, but with what we could call strong typing of lexical elements!

(The story is that the statement "DO 10 I = 1, 1000" got mistyped with a
period in place of the comma--an error easily overlooked if your keypunch
ribbon needs replacement--and got tokenized as "DO10I = 1.1000", a
perfectly valid assignment to an implicitly declared variable DO10I.  Two
distinct language-design blunders--insignficant spaces and implicit
declaration--conspired to hide this typo.)

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[M. J. Saltzman]

ncohen@watson.ibm.com (Norman H. Cohen) writes:

>In article <3bo402$23a@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu
>(Michael Feldman) writes: 

>|> In general I agree, and find that subtypes give the needed range checks
>|> without getting the programmer into type-compatibility stew, but there
>|> are these persistent "urban myths" about Fortran codes whose mangled
>|> dimension analysis sent space probes off the wrong way, etc.

>The urban myth about the space probe is not concerned with strong typing
>of data, but with what we could call strong typing of lexical elements!

Right enough.  On the other hand, in the late 80's there was a "Star
Wars" test involving a mirror on the Space Shuttle and a laser on the
ground.  At test time, it turned out that the Shuttle was oriented
with the mirror pointing *away* from the earth.  I believe that the
error was indeed traced to faulty dimensional analysis (a distance in
feet passed to a module that expected miles, or something similar).
-- 
		Matthew Saltzman
		Clemson University Math Sciences
		mjs@clemson.edu

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <mjs.786643878@hubcap>,
M. J. Saltzman <mjs@hubcap.clemson.edu> wrote:

>Right enough.  On the other hand, in the late 80's there was a "Star
>Wars" test involving a mirror on the Space Shuttle and a laser on the
>ground.  At test time, it turned out that the Shuttle was oriented
>with the mirror pointing *away* from the earth.  I believe that the
>error was indeed traced to faulty dimensional analysis (a distance in
>feet passed to a module that expected miles, or something similar).

Thanks! I knew I had seen it somewhere. :-)

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[M. J. Saltzman]

mfeldman@seas.gwu.edu (Michael Feldman) writes:

>In article <mjs.786643878@hubcap>,
>M. J. Saltzman <mjs@hubcap.clemson.edu> wrote:

>>Right enough.  On the other hand, in the late 80's there was a "Star
>>Wars" test involving a mirror on the Space Shuttle and a laser on the
>>ground.  At test time, it turned out that the Shuttle was oriented
>>with the mirror pointing *away* from the earth.  I believe that the
>>error was indeed traced to faulty dimensional analysis (a distance in
>>feet passed to a module that expected miles, or something similar).

>Thanks! I knew I had seen it somewhere. :-)

>Mike Feldman

OK, now that I've related the story, I have realized that I don't have
a bona fide citation to support it (although I have a quite vivid
memory of it).  So, can anyone supply a reference to supporting
documentation?

Thanks in advance.
-- 
		Matthew Saltzman
		Clemson University Math Sciences
		mjs@clemson.edu

Re: Why don't large companies use Ada?

[Arne Dehli Halvorsen]

In article <mjs.786717180@hubcap> mjs@hubcap.clemson.edu (M. J. Saltzman) writes:
>mfeldman@seas.gwu.edu (Michael Feldman) writes:
>>In article <mjs.786643878@hubcap>,
>>M. J. Saltzman <mjs@hubcap.clemson.edu> wrote:

>>>Right enough.  On the other hand, in the late 80's there was a "Star
>>>Wars" test involving a mirror on the Space Shuttle and a laser on the
>>>ground.  At test time, it turned out that the Shuttle was oriented
>>>with the mirror pointing *away* from the earth.  I believe that the
>>>error was indeed traced to faulty dimensional analysis (a distance in
>>>feet passed to a module that expected miles, or something similar).

>>Thanks! I knew I had seen it somewhere. :-)

>>Mike Feldman

>OK, now that I've related the story, I have realized that I don't have
>a bona fide citation to support it (although I have a quite vivid
>memory of it).  So, can anyone supply a reference to supporting
>documentation?
Karla Jennings: The devouring fungus (Norton ISBN 0-393-30732-8) gives
a few more details: Space shuttle Discovery above Hawaii, bouncing a laser 
down to Mauna Kea, Maui, 10023 feet of height. Reads figure as nautical miles,
flips whole shuttle and mirrors laser to the stars...

ADH 
 >Thanks in advance.
>-- 
>                Matthew Saltzman
>                Clemson University Math Sciences
>                mjs@clemson.edu

Re: Why don't large companies use Ada?

[Bob Duff]

In article <3bvc4k$1lla@watnews1.watson.ibm.com>,
Norman H. Cohen <ncohen@watson.ibm.com> wrote:
>The urban myth about the space probe is not concerned with strong typing
>of data, but with what we could call strong typing of lexical elements!
>...

There's another story, I think about the space shuttle, which involved
distances being measured in miles, whereas one piece of code assumed
feet (or vice versa), causing the space shuttle to point a certain
mirror in the wrong direction, thus ruining whatever experiment they
were doing.

- Bob
-- 
Bob Duff                                bobduff@inmet.com
Oak Tree Software, Inc.
Ada 9X Mapping/Revision Team (Intermetrics, Inc.)

Re: Why don't large companies use Ada?

[Robert Dewar]

Regarding the famous DO loop bug (the Venus probe I think if I remember right)
there is a nice addition to this story.

IBM issued an APAR against Fortran IV under IBSYS on the 7094 that said
something like:

   DO 10 I = 1.5

is misclassified and treated as an assignment statement. This will be
corrected in a fututure release.

Needsless to say, this was NOT corrected in a future release, and I would
guess that the APAR is still open :-)

Re: Why don't large companies use Ada?

[Mark S. Hathaway]

> In article <3c1oj4$h3s@gnat.cs.nyu.edu>,
> dewar@cs.nyu.edu (Robert Dewar) writes:

> Regarding the famous DO loop bug (the Venus probe I think if I remember right)
> there is a nice addition to this story.
> 
> IBM issued an APAR against Fortran IV under IBSYS on the 7094 that said
> something like:
> 
>    DO 10 I = 1.5
> 
> is misclassified and treated as an assignment statement. This will be
> corrected in a fututure release.
> 
> Needsless to say, this was NOT corrected in a future release, and I would
> guess that the APAR is still open :-)

In the world of music it is said that the silence between the notes is an
integral part of the music.  Apparently the language parser should recognize
the blanks between "key words" as relevant.

 
Mark S. Hathaway       <hathawa2@marshall.edu>

Spaces in Fortran (was Re: Why don't large companies use Ada?)

[Michael D Shapiro]

In article <1994Dec6.131404@hobbit>,
Mark S. Hathaway <hathawa2@marshall.edu> wrote:
>> In article <3c1oj4$h3s@gnat.cs.nyu.edu>,
>> dewar@cs.nyu.edu (Robert Dewar) writes:
>
>>   { statements describing the infamous Fortran assignment statement: }
>> 
>>    DO 10 I = 1.5
>
>In the world of music it is said that the silence between the notes is an
>integral part of the music.  Apparently the language parser should recognize
>the blanks between "key words" as relevant.

The current Fortran standard allows two forms for statements, "free"
and "fixed."  In free source form, a blank must be used to separate
names, constants, or labels from adjacent keywords, names constants, or
labels.  Also, the blank characters must not appear within lexical
tokens.  In fixed source form, blanks are insignificant except in a
character context.  I believe we would currently refer to this form as
a "legacy" form (something handed down from an ancestor or predecessor)
except that some of us that built and used FORTRAN compilers are still
around.
--
Michael D. Shapiro, Ph.D.         Internet: mshapiro@nosc.mil
Code 4123, NCCOSC RDT&E Division (NRaD)    San Diego CA 92152
Voice: (619) 553-4080    FAX: (619) 553-4808    DSN: 553-4080

Re: Why don't large companies use Ada?

[Norman H. Cohen]

In article <3bldlu$84r@gnat.cs.nyu.edu>, dewar@cs.nyu.edu (Robert Dewar)
writes: 

|> IN practice, I find the strong type protection of different integer types
|> to be of minimal value, strong typing is most important at a higher level
|> than that, but that's a matter of personal taste.

For computationally intensive uses of numeric types (the subject of
Robert's posting) I certainly agree.

However, I have found a compelling use of strong typing to be
the declaration of distinct array-index types: 

   type Dictionary_Index_Type is range Min_Index .. Max_Index;
   type Dictionary_Type is
      array (Dictionary_Index_Type) of Dictionary_Entry_Type;

   type Node_Name_Type is range Min_Node .. Max_Node;
   type Graph_Type is array (Node_Name_Type) of Node_Info_Type;

This kind of distinction rarely forces one into a type conversion and it
avoids a kind of error that really does occur in programs where the same
type is used to index different kinds of arrays.

Most importantly, as types like Dictionary_Index_Type and Node_Name_Type
propagate into subprogram interfaces, they provide a valuable aid to
program understanding, making it crystal clear what kind of information
is being passed to and from subprograms.

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[Thomas M. Breuel]

In article <3bvbaa$1lla@watnews1.watson.ibm.com> ncohen@watson.ibm.com (Norman H. Cohen) writes:
|In article <3bldlu$84r@gnat.cs.nyu.edu>, dewar@cs.nyu.edu (Robert Dewar)
|writes: 
|
||> IN practice, I find the strong type protection of different integer types
||> to be of minimal value, strong typing is most important at a higher level
||> than that, but that's a matter of personal taste.
|
|For computationally intensive uses of numeric types (the subject of
|Robert's posting) I certainly agree.
|
|However, I have found a compelling use of strong typing to be
|the declaration of distinct array-index types: [...]

Fortunately, you can define distinct array index types in C++ as well
as in Ada.

				Thomas.

Re: Why don't large companies use Ada?

[Sten Drescher]

In article <house.786178243@helios>, house@helios.usq.EDU.AU (ron house) writes:

 rh> How about "Ada - use it when you want to multiply a length by a
 rh> length and get a length, or when you want to be prevented from
 rh> dividing an area by a length because they have different types."

 rh> (PS: yes, you can fix that by defining zillions of additional *,/,-
 rh> and + functions, but good grief...)

	If you're silly enough to define types for length and area, then
yes, you should have to define the type conversions.  It's the
programmer who declared that there's a distinction, not the compiler.
--

Re: Why don't large companies use Ada?

[Robert Dewar]

You can add to Mike's list of "military
"

Re: Why don't large companies use Ada?

[Robert Dewar]

Add to Mike's list of "military" inventions: Internet itself, which presumably
most participants in this newsgroup find useful :-)

Re: Why don't large companies use Ada?

[Robert I. Eachus]

In article <3bcntp$dgj@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:

 > Add to Mike's list of "military" inventions: Internet itself, which
 > presumably most participants in this newsgroup find useful :-)

   About six months ago I had to research part of the history of the
first DoD developed portable high-level language.  It is called COBOL.
;-)

--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <EACHUS.94Nov28142716@spectre.mitre.org>,
Robert I. Eachus <eachus@spectre.mitre.org> wrote:

>   About six months ago I had to research part of the history of the
>first DoD developed portable high-level language.  It is called COBOL.
>;-)

Jean Sammet and others who were on the original CODASYL committee swear
that COBOL was _not_ "DoD-developed", rather that DoD was one of many 
players involved. Obviously DoD had an interest in COBOL, but can not
be said to have "commissioned" or funded COBOL in the sense that they
did with Ada. 

Can anyone else fill in gaps here? I'm adding my two cents only to get
the facts straight. The other stuff on thge list we've been accumulating
here really was Dod funded (X, Berkeley Unix, TCPIP, Internet, etc.).

Indeed, many people credit AT&T with popularizing Unix; my recollection
is that AT&T Unix was something of a research curiosity till Berkeley
"enhanced" it and - a few years later - Sun Microsystems (which can
be thought of as originally a Berkeley "spinoff") popularized it as
a commercial system. Berkeley's funding came (mostly) from ARPA.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[IanMaclure]

mfeldman@seas.gwu.edu (Michael Feldman) writes:

>In article <EACHUS.94Nov28142716@spectre.mitre.org>,
>Robert I. Eachus <eachus@spectre.mitre.org> wrote:

>>   About six months ago I had to research part of the history of the
>>first DoD developed portable high-level language.  It is called COBOL.
>>;-)

>Jean Sammet and others who were on the original CODASYL committee swear
>that COBOL was _not_ "DoD-developed", rather that DoD was one of many 
>players involved. Obviously DoD had an interest in COBOL, but can not
>be said to have "commissioned" or funded COBOL in the sense that they
>did with Ada. 

I'm trying to remember what Admiral Grace Hopper said about COBOL but all
I can remember is that she felt a certain amount of proprietary pride in it.
She did specifically mention that she had some of the ratings ( thats enlisted
types ) in her organization produce a compiler at one point.
COBOL may have originated with the Navy perhaps with her office?
And yes COBOL cannot be said to have evolved like Ada, of course if the
space program had "evolved" like Ada there'd still be flame wars over whether
or not to put windows in the capsules and an actual launch would be years
away.

IBM
-- 
################ No Times Like The Maritimes, Eh! ######################
# IBM   aka      #    Ian_Maclure@QMGATE.arc.nasa.gov           (desk) #
# Ian B MacLure  #    maclure@(remulak/eos).arc.nasa.gov   (currently) #
########## Opinions expressed here are mine, mine, mine. ###############

Re: Why don't large companies use Ada?

[David Weller]

In article <1994Nov29.182445.6678@eos.arc.nasa.gov>,
IanMaclure <maclure@eos.arc.nasa.gov> wrote:
>And yes COBOL cannot be said to have evolved like Ada, of course if the
>space program had "evolved" like Ada there'd still be flame wars over whether
>or not to put windows in the capsules and an actual launch would be years
>away.
>

Please clarify this or retract it.

-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
	|"Quitting C++ isn't so difficult, provided you show as much |
	|	persistence stopping as you did starting." dweller   |

Re: Why don't large companies use Ada?

[Robert Dewar]

David, regarding your demand to Ian that he clarify or retract his
silly statement, it's probably a waste of time. People who make statements
like that obviously don't know what they are talking about. It is
particularly interesting that in the programming language field, people
feel free to make technical pronouncements about languages they know
NOTHING about (how many Ada folks have done that to COBOL :-)

COBOL origin (was Re: Why don't large companies use Ada?)

[Michael D Shapiro]

In article <1994Nov29.182445.6678@eos.arc.nasa.gov>,
IanMaclure <maclure@eos.arc.nasa.gov> wrote:
>mfeldman@seas.gwu.edu (Michael Feldman) writes:
>
>>In article <EACHUS.94Nov28142716@spectre.mitre.org>,
>>Robert I. Eachus <eachus@spectre.mitre.org> wrote:
>
>>>   About six months ago I had to research part of the history of the
>>>first DoD developed portable high-level language.  It is called COBOL.
>>>;-)
>
>>Jean Sammet and others who were on the original CODASYL committee swear
>>that COBOL was _not_ "DoD-developed", rather that DoD was one of many 
>>players involved. Obviously DoD had an interest in COBOL, but can not
>>be said to have "commissioned" or funded COBOL in the sense that they
>>did with Ada. 
>
>I'm trying to remember what Admiral Grace Hopper said about COBOL but all
>I can remember is that she felt a certain amount of proprietary pride in it.
>She did specifically mention that she had some of the ratings ( thats enlisted
>types ) in her organization produce a compiler at one point.
>COBOL may have originated with the Navy perhaps with her office?
>And yes COBOL cannot be said to have evolved like Ada, of course if the
>space program had "evolved" like Ada there'd still be flame wars over whether
>or not to put windows in the capsules and an actual launch would be years
>away.

The story, as I recall, is that the computer makers, each of which had
a commercial programming language, could not get together for fear of
antitrust suits.  To get around the hurdle and come up with a common
business-oriented language, some "they" decided that the U.S.
government could convene a conference on data systems languages to
discuss the problem.  The task was assigned to the Navy, the most
scientifically advanced government agency at the time (late 1950s) and
the job of calling the conference and coordinating the work went to the
group with Grace Hopper.  Later she managed the team that developed the
COBOL validation process, later transfered to NIST.

COBOL did evolve, of course.  The 1960 standard served as a starting
point and many companies built language processors based on it, with
their own extensions.  By the time of the 1974 standard, people were
really starting to think of actually standardizing and the Federal
Information Processing Standard (FIPS) specifying four levels of the
language helped get rid of many idiosyncrasies.  Companies continued
to "improve" it, adding newer programming constructs.  Another standard
appeared in 1985.

One problem many people have with Ada is that it was not allowed to
evolve in the same way most other languages did.  This lack of
incremental improvements has meant that new paradigms cannot be handled
easily, requiring instead a massive step that still won't turn out
right in the view of many.  This view of a language, frozen in time,
may be wonderful for the managers of huge programs, but it turns many
programmers off.
--
Michael D. Shapiro, Ph.D.         Internet: mshapiro@nosc.mil
Code 4123, NCCOSC RDT&E Division (NRaD)    San Diego CA 92152
Voice: (619) 553-4080    FAX: (619) 553-4808    DSN: 553-4080

Re: COBOL origin

[Wayne Dernoncourt]

In article <1994Nov30.221351.2917@nosc.mil>,
mshapiro@nosc.mil (Michael D Shapiro) wrote:
 > In article <1994Nov29.182445.6678@eos.arc.nasa.gov>,
 > IanMaclure <maclure@eos.arc.nasa.gov> wrote:
 > >mfeldman@seas.gwu.edu (Michael Feldman) writes:
 > >
 > >>In article <EACHUS.94Nov28142716@spectre.mitre.org>,
 > >>Robert I. Eachus <eachus@spectre.mitre.org> wrote:
 
 > >I'm trying to remember what Admiral Grace Hopper said about COBOL but all
 > >I can remember is that she felt a certain amount of proprietary pride in it.
 > >She did specifically mention that she had some of the ratings ( thats enlisted
 > >types ) in her organization produce a compiler at one point.
 > >COBOL may have originated with the Navy perhaps with her office?
 > >And yes COBOL cannot be said to have evolved like Ada, of course if the
 > >space program had "evolved" like Ada there'd still be flame wars over whether
 > >or not to put windows in the capsules and an actual launch would be years
 > >away.

 > The story, as I recall, is that the computer makers, each of
 > which had a commercial programming language, could not get
 > together for fear of antitrust suits.  To get around the hurdle
<snip>
 > calling the conference and coordinating the work went to the
 > group with Grace Hopper.  Later she managed the team that
 > developed the COBOL validation process, later transfered to
 > NIST.

Yeah but I think you're missing the question.  What did she say about
COBOL.  I had the honor of hearing her talk a couple of times and she said
that no language should be around forever.  She also threatened everyone
that if we ever said "But that's the way we've always done it", that whe
would materialize beside us and hit us.  I think she was proud of the work
that she had done with COBOL, but she was saddened that people couldn't
let go of an old language and embrace a replacement language (COBOL 74 &
COBOL 85 bear little resemblance to the original).

The Navy team developed a test suite to verify the compliance of a vendors
implementation, after the testing function was transferred to NIST, NIST
replaced the lines that said "test programs developed by the US Navy", with
ones that said "test programs developed by NIST" (or something like that,
may not have been developed), she raised hell and it was changed back to
US Navy.

The last time I heard her talk was either at a trade show held in
Washington, DC trying to sell stuff to the feds, or at a CASE workshop put
on by a Navy command (NAVSEA - I think) - in either case, she had retired
and was working for DEC [rumor was that Ken Olsen wanted someone working
at DEC that was older than he was].  She wanted to see a interface
standard develop that would let CASE tools work together.  Her analogy was
screwdrivers and screws.  There were different types of screws from a lot
of different vendors, but most times you could find a screwdriver from a
different set of vendors that would make it work.

Enough reminiscing.

-- 
Take care      |   This clown speaks for himself, his job doesn't
Wayne D.       |   pay for this, etc. (directly anyway)

Re: COBOL origin (was Re: Why don't large companies use Ada?)

[Robert I. Eachus]

In article <1994Nov30.221351.2917@nosc.mil> mshapiro@nosc.mil (Michael D Shapiro) writes:

  > COBOL did evolve, of course.  The 1960 standard served as a
  > starting point and many companies built language processors based
  > on it, with their own extensions.  By the time of the 1974
  > standard, people were really starting to think of actually
  > standardizing and the Federal Information Processing Standard
  > (FIPS) specifying four levels of the language helped get rid of
  > many idiosyncrasies.  Companies continued to "improve" it, adding
  > newer programming constructs.  Another standard appeared in 1985.

   Don't ignore COBOL 68, and wasn't there a COBOL 64?  In the early
years COBOL evolved fairly rapidly.

  > One problem many people have with Ada is that it was not allowed to
  > evolve in the same way most other languages did.  This lack of
  > incremental improvements has meant that new paradigms cannot be handled
  > easily, requiring instead a massive step that still won't turn out
  > right in the view of many.  This view of a language, frozen in time,
  > may be wonderful for the managers of huge programs, but it turns many
  > programmers off.

   Ada went from Green, to Preliminary Ada, to Ada 80, to Ada 83 in
about 5 years, so the step from Ada 83 to Ada 95 should be regarded as
appropriate for a mature language.  Those first revisions were fairly
large steps (mostly visibly in exceptions and text IO), and Ada 80 was
a military standard...

--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: COBOL origin (was Re: Why don't large companies use Ada?)

[Dr. Richard Botting]

Michael D Shapiro (mshapiro@nosc.mil) wrote:
: In article <1994Nov29.182445.6678@eos.arc.nasa.gov>,
: IanMaclure <maclure@eos.arc.nasa.gov> wrote:
: >mfeldman@seas.gwu.edu (Michael Feldman) writes:
: >
: >>In article <EACHUS.94Nov28142716@spectre.mitre.org>,
: >>Robert I. Eachus <eachus@spectre.mitre.org> wrote:
: >
: >>>   About six months ago I had to research part of the history of the
: >>>first DoD developed portable high-level language.  It is called COBOL.
[...snip]
: One problem many people have with Ada is that it was not allowed to
: evolve in the same way most other languages did.  
Ada, like COBOL, FORTRAN and other USA National Standards (like C)
have an 11 year update cycle (if all goes well... FORTRAN 88 was a little 
late...).  Ada 94 is going to be on time.


: This lack of
: incremental improvements has meant that new paradigms cannot be handled
: easily, requiring instead a massive step that still won't turn out
: right in the view of many.  This view of a language, frozen in time,
Are you sure you are talking about Ada, and not C, FORTRAN, COBOL,... ?

: but it turns many
: programmers off.
Especially those who like debugging new features or hacking round
ambiguities... 

--
EMail: dick@csci.csusb.edu=rbotting@wiley.csusb.edu.
ftp://ftp.csci.csusb.edu/dick
<a href="http://www.csci.csusb.edu/dick/">WWW</a>
Disclaimer::=`CSUSB may or may not agree with this message`.
Copyright(1994)::=`Copy as long as you include this copyright and signature`.

Re: Why don't large companies use Ada?

[Paul Johnson]

Michael Feldman (mfeldman@seas.gwu.edu) wrote:

> Indeed, many people credit AT&T with popularizing Unix; my recollection
> is that AT&T Unix was something of a research curiosity till Berkeley
> "enhanced" it and - a few years later - Sun Microsystems (which can
> be thought of as originally a Berkeley "spinoff") popularized it as
> a commercial system. Berkeley's funding came (mostly) from ARPA.

There was also DEC Ultrix on the VAX, another commercialised BSD 4.2.
I think that may have predated SunOS.  It was certainly my first Unix.

Paul.

-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <5963@gec-mrc.co.uk>, Paul Johnson <paj@gec-mrc.co.uk> wrote:
>Michael Feldman (mfeldman@seas.gwu.edu) wrote:
>
>> Indeed, many people credit AT&T with popularizing Unix; my recollection
>> is that AT&T Unix was something of a research curiosity till Berkeley
>> "enhanced" it and - a few years later - Sun Microsystems (which can
>> be thought of as originally a Berkeley "spinoff") popularized it as
>> a commercial system. Berkeley's funding came (mostly) from ARPA.
>
>There was also DEC Ultrix on the VAX, another commercialised BSD 4.2.
>I think that may have predated SunOS.  It was certainly my first Unix.

Could be, though as I recall, Sun's first machines came out in about 1981
(I ordered an early Sun-100 for a project; I think it was that vintage).
If I recall, that was pre-Ultrix, still in the days when folks were
running "straight" BSD on their Vax-en and DEC was pretending UNIX
didn't exist. At some point DEC bowed to the inevitable and brought
Ultrix out. It was certainly around the same time as SunOS; I seem
to think it was a tad later. 

Really it doesn't matter - the point was that (D)ARPA brought UNIX 
to real popularity by funding Berkeley. And (D)ARPA is part of DoD.

Mike Feldman

Re: Why don't large companies use Ada?

[Davidson Corry]

In article <3be9as$jrh@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) says:
>
>...Indeed, many people credit AT&T with popularizing Unix; my recollection
>is that AT&T Unix was something of a research curiosity till Berkeley
>"enhanced" it and - a few years later - Sun Microsystems (which can
>be thought of as originally a Berkeley "spinoff") popularized it as
>a commercial system.

I think AT&T's** decision to make Unix binaries and source code available
cheaply to educational institutions _did_ popularize Unix, by exposing an
entire generation of programmers to it during their formative years.  By
the time they graduated into industry, it was the tool they knew best
(and some, inevitably, mistook this for a proof that it was the best tool)
and so they carried it with them into their jobs.

Certainly the BSD variants helped to make Unix useful and popular.  I
wonder if BSD would even exist, if AT&T hadn't made the source cheap?

**(It's as silly to speak of "AT&T" making a decision as it is to say
   that "Microsoft thinks..."  I'm sure _somebody_ at AT&T realized
   what the effect would be of giving Unix away to colleges.  I've just
   never heard that it was the official policy of the Machiavell...
   excuse me, of the Marketing Department.)  <grin>

Re: Why don't large companies use Ada?

[Richard Riehle]

In article <3be9as$jrh@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
>In article <EACHUS.94Nov28142716@spectre.mitre.org>,
>Robert I. Eachus <eachus@spectre.mitre.org> wrote:
>
>>   About six months ago I had to research part of the history of the
>>first DoD developed portable high-level language.  It is called COBOL.
>>;-)
>
>Jean Sammet and others who were on the original CODASYL committee swear
>that COBOL was _not_ "DoD-developed", rather that DoD was one of many 
>players involved. Obviously DoD had an interest in COBOL, but can not
>be said to have "commissioned" or funded COBOL in the sense that they
>did with Ada. 
>
>Can anyone else fill in gaps here?
>Mike Feldman

-- ====================== Reply Separator =============================

Aha! The COBOL question.  When a young U.S. Navy officer named Grace
Hopper developed the precursor of COBOL, it was a UNIVAC-owned product.
Eventually it became know as COBOL, but all COBOL manuals included a
statement of permission from UNIVAC.  Of course, UNIVAC's most important
customer in those days was the Department of Defense.

In April 1964, alas I was there, IBM announced its System 360 and a new
programming language, PL/1, which would be the language to replace all
languages (Sound familiar?).  COBOL was pronounced dead.

I must skip a lot of details in the interest of economy, but here are a few
items. In those days, UNIVAC, Honeywell, CDC, and four other "dwarfs" were
still important players in the computer industry. With the exception of CDC,
none of them had a serious commitment to PL/1.  When the DoD published 
software RFP's for information systems applications the language specified
was nearly always COBOL, not PL/1, so all vendors could compete.  I am
especially familiar with a large USAF contract that was going to be in PL/1,
evoked protests from the non-IBMers, and got reversed to COBOL.  

This is one more case of a better technical solution going down the
sewer for non-technical reasons.  In this case, the DoD prevailed, but only
because it was pummeled so effectively by the seven dwarfs that Snow White
was forced to consume the bitter apple and like it.  Sory for the mixed
metaphor.  

The influence of the DoD persuaded business to proceed cautiously or not
at all with PL/1. In retrospect, if PL/1 had succeeded, it could have evolved
in a way that would have made Ada unnecessary.   Then where would we be?

Richard Riehle 

Re: Why don't large companies use Ada?

[Paige Vinall]

In <1994Nov30.151523.19742@sei.cmu.edu> Richard Riehle writes:
>The influence of the DoD persuaded business to proceed cautiously or not
>at all with PL/1. In retrospect, if PL/1 had succeeded, it could have evolved
>in a way that would have made Ada unnecessary.   Then where would we be?
>

Actually that isn't really the case.  PL/I brings in millions of dollars
for IBM every year.  The DoD does use PL/I but not to the extent of
COBOL or Ada.  PL/I is heavily used in the US, Europe and Asia.  PL/I
has been enhanced greatly to fit better with C/C++.

PL/I is succeeding!

Paige Vinall - PL/I Development Manager, IBM
Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Re: Why don't large companies use Ada?

[Robert Dewar]

Note that one significant reason that PL/1 failed was the very poor initial
implementations that IBM produced early on, leading for example to the
rather disastrous reversal of the Sears commitment and their reversion
to COBOL, which had nothing to do with the DoD or procurement as I remember!

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3bmb4r$9kc@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>Note that one significant reason that PL/1 failed was the very poor initial
>implementations that IBM produced early on, leading for example to the
>rather disastrous reversal of the Sears commitment and their reversion
>to COBOL, which had nothing to do with the DoD or procurement as I remember!
>

At the time, I recall speaking with engineers who walked away from 
PL/1 _simply_ because it mapped multidimensional arrays row-major, so that
interfacing to old Fortran codes required the programmer to tranpose
everything on the PL/1 side before passing it to the Fortran subroutine.

The Ada 9X Fortran interface - finally - fixes this. Nothing I can see
would've prevented an Ada 83 implementation from supporting column-major
arrays, as the LRM does not prescribe a mapping. As I've said many times,
I think it showed real lack of imagination for Ada vendors to ignore
the potentially huge market of new Ada code interfaced to all that 
Fortran numeric stuff.

If _one_ vendor would come forward and say "Mike, we did a serious study
of that domain and found that, even if we adapted Ada compilers to suit it,
we'd get very few customers," I'd get off this hobbyhorse.

Mike Feldman

Re: Why don't large companies use Ada?

[Keith Thompson]

In <3bo4il$3lb@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
> At the time, I recall speaking with engineers who walked away from 
> PL/1 _simply_ because it mapped multidimensional arrays row-major, so that
> interfacing to old Fortran codes required the programmer to tranpose
> everything on the PL/1 side before passing it to the Fortran subroutine.

I'll take your word for this, but I still find it difficult to understand.
If you're writing new PL/I or Ada code to interface to existing Fortran
libraries, all you should need to do is swap the indices; there should
be no need to physically transpose the arrays themselves.  This should
be only a small fraction of the effort required to interface to Fortran
(for Ada, matching the types and writing the Interface pragmas; for
PL/I, ???).  Could it be that the engineers were using this as an excuse
to stick with Fortran, or am I missing something?

-- 
Keith Thompson (The_Other_Keith)  kst@alsys.com
TeleSoft^H^H^H^H^H^H^H^H Alsys, Inc.
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2718
/user/kst/.signature: I/O error (core dumped)

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <D07zvq.L3p@alsys.com>, Keith Thompson <kst@alsys.com> wrote:
>In <3bo4il$3lb@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:

>I'll take your word for this, but I still find it difficult to understand.
>If you're writing new PL/I or Ada code to interface to existing Fortran
>libraries, all you should need to do is swap the indices; there should
>be no need to physically transpose the arrays themselves.  

Yes, that's true. The transposition is mental. 

Don't take this personally, but the "all you should..." attitude is, in 
my experience, quite typical of computer folks who don't understand, or
don't want to understand, what it takes to get a customer to buy.
They don't focus on making it easy for the customer, by _really_
understanding the customer's needs and mentality.

>This should
>be only a small fraction of the effort required to interface to Fortran
>(for Ada, matching the types and writing the Interface pragmas; for
>PL/I, ???).  Could it be that the engineers were using this as an excuse
>to stick with Fortran, or am I missing something?

You hit the nail right on the head. Of course they were using it as an
excuse. And IBM apparently did not understand the extent to which non-
technical considerations affect a choice of language. 

In my experience, interfacing between languages - certainly for the
kind of fairly simple vector/matrix interfaces the typical Fortranner
would use (the _subprograms_ aren't simple but the _interfaces_
tend to be) - is, to a very large extent,a matter of getting the
data structures to agree. I don't understand your "small fraction"
assertion. 

There are big Fortran libraries out there (e.g. IMSL)
whose interfaces consist of a couple of arrays and (maybe) some
bounds parameters (which in "pure" Ada would be handled by
unconstrained array types). An engineer trying to use these
from PL/1 would have to keep all his arrays in transposed form,
which would make his new PL/1 code look pretty unintuitive -
what would naturally be a row suddenly becomes a column.
Why bother, if nobody is forcing you to use PL/1?

The Ada community could have observed this phenomenon, and gone
after those engineers with an Ada 83 equivalent of Interfaces.Fortran
and Pragma Convention. Nothing in LRM83 prevented an implementer from
doing it. To my knowledge, not one vendor thought all those engineers
in (say) the Fortran-dependent DoE labs were worth supporting.

My offer is still open: if _one_ implementer steps up and says "Mike, we
_did_ take a serious look at that community and our market studies
showed that it wouldn't pan out", and gave a reason or two, I'll
promise to get off the hobbyhorse.

That Interfaces.Fortran and Pragma Convention exist in Ada 9X tells me
that _somebody_ is now taking this issue seriously. All the more support
for my assertion that we should have taken it seriously 10 years ago.

Mike Feldman

Re: Why don't large companies use Ada?

[Robert Dewar]

if you are writing PL/1, you use ISUB to swap indices of arrays automatically

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3c05t4$ceg@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>if you are writing PL/1, you use ISUB to swap indices of arrays automatically
>

But the IBM chap who posted a note about this pointed out that ISUB
in fact _transposes_ the array. That's expensive in both time and
space. In Ada, because the storage mapping is unspecified in the LRM,
an implementer can choose to go column-major if his target audience
is doing a bunch of Fortran interfacing. 

Indeed, an implementation-dependent pragma could be used to let the 
programmer inform the compiler that certain array types were to look 
"Fortran-ish". This would all be perfectly legal, and - unless I am 
missing something really fundamental - not terribly hard to implement.

Unhappily, no Ada implementer took advantage of this opportunity to
make Ada more Fortran-friendly. Happily, Ada 95 standardizes the
whole thing with Interfaces.Fortran.

Mike Feldman

Re: Why don't large companies use Ada?

[Robert Dewar]

Mike, I think you misunderstand ISUB, it does exactly what you want and
was largely intended for that purpose. Yes of course it transposes the
array, that's what swapping subscripts does (logically) to an array!

Re: Why don't large companies use Ada?

[Keith Thompson]

In <3btgfs$m8c@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
> Don't take this personally, but the "all you should..." attitude is, in 
> my experience, quite typical of computer folks who don't understand, or
> don't want to understand, what it takes to get a customer to buy.
> They don't focus on making it easy for the customer, by _really_
> understanding the customer's needs and mentality.

I don't (fully) understand, but I'd like to; that's why I'm posting.

> >This should
> >be only a small fraction of the effort required to interface to Fortran
> >(for Ada, matching the types and writing the Interface pragmas; for
> >PL/I, ???).  Could it be that the engineers were using this as an excuse
> >to stick with Fortran, or am I missing something?
> 
> You hit the nail right on the head. Of course they were using it as an
> excuse. And IBM apparently did not understand the extent to which non-
> technical considerations affect a choice of language. 

So an estimate of the actual impact should be based, not on the total
number of engineers who rejected PL/I because of the indexing problem,
but on the subset of that total who wouldn't have found some other excuse
to reject it.  (I don't suppose anyone has any figures on this.  8-)})

> In my experience, interfacing between languages - certainly for the
> kind of fairly simple vector/matrix interfaces the typical Fortranner
> would use (the _subprograms_ aren't simple but the _interfaces_
> tend to be) - is, to a very large extent,a matter of getting the
> data structures to agree. I don't understand your "small fraction"
> assertion. 

Interfacing is the easy part.  I'm assuming that the bulk of the effort
is simply learning the other language; interfacing PL/I to Fortran isn't
even an issue if you're unwilling to use PL/I in the first place.  For an
engineer who isn't primarily a programmer, but who has learned to use
Fortran as an engineering tool, learning another programming language
(PL/I, Ada, C, whatever) could be a significant obstacle, regardless
of how smoothly the two languages can be interfaced.  The problem
(I'm guessing) isn't just "it indexes arrays backwards", but "it
isn't Fortran".  (No offense to engineers; I'd rather run a Fortran
program written by an engineer that drive across a bridge designed by
a programmer 8-)}.)

Hmm.  How difficult (or possible) would it be to create an Ada "thick
binding" interface to a Fortran library that handles the index swapping?
I suppose it depends on the Fortran library.  Presumably a solution that
actually re-maps the arrays would be unacceptable due to performance
problems.  Has anyone ever looked into this possibility, though?

Disclaimers: I've never used Fortran, beyond a few tiny test programs.
I've heard about these big Fortran libraries, but I know very little about
them.  I've never used PL/I at all.  I seldom use multidimensional arrays.
I probably don't know what I'm talking about; if I did, I'd probably be
on the other side of this discussion.

> That Interfaces.Fortran and Pragma Convention exist in Ada 9X tells me
> that _somebody_ is now taking this issue seriously. All the more support
> for my assertion that we should have taken it seriously 10 years ago.

I agree (I think) that adding these to Ada 9X is A Good Thing.  Note that,
since Convention is a representation pragma, compilers will be required
to implement array re-mapping (transposition?) for type conversions:

    type Ada_Array is array(1 .. 100, 1 .. 100) of Float;
    type Fortran_Array is new Ada_Array;
    pragma Convention(Fortran, Fortran_Array);
    A: Ada_Array := ...;
    F: Fortran_Array := Fortran_Array(A);

Implementers take note.

-- 
Keith Thompson (The_Other_Keith)  kst@alsys.com
TeleSoft^H^H^H^H^H^H^H^H Alsys, Inc.
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2718
/user/kst/.signature: I/O error (core dumped)

Re: Why don't large companies use Ada?

[Robert Dewar]

Interesting that PL/1 creates an image of an ISUB'ed array passed to Fortran.
As far as I can see, this is not at all required by the language, it is just
a bit of bad implementation, isn't that right?

Re: Why don't large companies use Ada?

[Robert Dewar]

Keith, the array conversion case is really not a special one. In general
an array conversion may involve a massive and possibly complex copy, and
most compilers will just generate a loop to do the copy, which handles
all possible cases, including the swapping of indices.

Re: Why don't large companies use Ada?

[Keith Thompson]

In <3c6ofm$50v@gnat.cs.nyu.edu> dewar@cs.nyu.edu (Robert Dewar) writes:
> Keith, the array conversion case is really not a special one. In general
> an array conversion may involve a massive and possibly complex copy, and
> most compilers will just generate a loop to do the copy, which handles
> all possible cases, including the swapping of indices.

You're right, of course; converting between index-swapped array types
shouldn't be any more difficult than converting between packed and
unpacked array types (once you've implemented index-swapped array types
in the first place).

I knew that -- I just wanted to see who was paying attention.  Yeah,
there's the ticket.

-- 
Keith Thompson (The_Other_Keith)  kst@alsys.com
TeleSoft^H^H^H^H^H^H^H^H Alsys, Inc.
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2718
/user/kst/.signature: I/O error (core dumped)

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3c6ofm$50v@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>Keith, the array conversion case is really not a special one. In general
>an array conversion may involve a massive and possibly complex copy, and
>most compilers will just generate a loop to do the copy, which handles
>all possible cases, including the swapping of indices.
>

Interesting discussion here. Let me play devil's advocate. Under what 
circumstances will an engineer writing new Ada to interface with
Fortran want to do this sort of array conversion? It seems to me that
almost all arrays likely to be passed as parameters to Fortran would be of
one or a few types declared as Fortran-friendly using pragma Convention.

Indeed, which arrays would such engineers _not_ wish to make 
Fortran-friendly?

I agree that the array conversion might be complex, but

(1) not more so than manually generating a transpose, and
(2) not used much, by the argument above.

Am I on the right track?

Mike Feldman

Fortran Arrays was: Why don't large companies use Ada?

[Richard Riehle]

The problem of Fortran arrays seems to be of serious interest.  One of
the things I like about Ada is that I can write a program which 
explicitly deals with this problem.  Of course it is a little more work,
but the following:

         type Row is array (some index of my choice) of Real;
         type Col is array (some index of my choice) of Row;

is useful, and so:

         type Col is array (index) of Real;
         type Row is array (index) of Col;

is also sometimes useful.

In fact, I find it quite nice that I can send an entire column or row
as the actual paramter of a subprogram.  For example,

         procedure P(C : in out Col);

         function  F(C : in Col) return Col;

I realize this is trivial in Ada, but not so trivial in other 
languages.  One of the things I have always liked about Ada is its
wonderful posiblilities for array processing.  

Richard Riehle
AdaWorks Software Engineering
Palo Alto, CA
email:  adaworks@netcom.com                   

Re: Fortran Arrays was: Why don't large companies use Ada?

[Richard Riehle]

In article <1994Dec9.133050.17629@sei.cmu.edu> riehler@ajpo.sei.cmu.edu (Richard Riehle) writes:
>
    Flaming my own stupid error:

>
>         type Col is array (index) of Real;
>         type Row is array (index) of Col;

   Ooooops!  That's what I get for composing this message before breakfast.
   It doesn't matter what I call it, does it.   We are still row major
   unless I  deal with the problem algorithmically.

   Sorry about that!

   I wonder how many people will pounce on me for this one?  OUCH   :-)


   Still, this is a handy feature of the language.

>Richard Riehle
>AdaWorks Software Engineering
>Palo Alto, CA
>email:  adaworks@netcom.com                   
>
>
>

Re: Why don't large companies use Ada?

[Robert Dewar]

Michael, I think you are confused about ISUB, and barking up a wrong tree
to say that Ada 9X is superior to PL/1 here. ISUB is a much more powerful
feature than pragma Convention Fortran in Ada 9X, and subsumes it.

Re: Why don't large companies use Ada?

[Robert Dewar]

With respect to array transposition, Keith was just pointing out something
to compiler implementors, I agree it is relatively unlikely that one would
do conversion between fortran style arrays and normal row major arrays, but
an OBVIOUS situation is when you are passing data between existing fortran
libraries and existing Ada libraries, and it is nice to have a simple
notation for the necessary array conversion.

Re: Why don't large companies use Ada?

[Richard A. O'Keefe]

mfeldman@seas.gwu.edu (Michael Feldman) writes:
>Yes, that's true. The transposition is mental. 
>In my experience, interfacing between languages - certainly for the
>kind of fairly simple vector/matrix interfaces the typical Fortranner
>would use (the _subprograms_ aren't simple but the _interfaces_
>tend to be) - is, to a very large extent,a matter of getting the
>data structures to agree.

I haven't used many PL/I compilers, but they all had "iSUB defining".
I never used that feature myself, but you could describe it as exposing
the linear subscript mapping at the source level.  The result is that
you could have a direct and a transposed view of the _same_ array.
I would have thought that made it fairly easy to interface PL/I to Fortran.
PL/I had, back in the 70's when I met it, a great many data types that
would not have mapped onto Fortran, but every Fortran variable _could_ be
mapped into PL/I more or less directly (with the aid of iSUB defining for
>=2-dimensional arrays).

-- 
"The complex-type shall be a simple-type."  ISO 10206:1991 (Extended Pascal)
Richard A. O'Keefe; http://www.cs.rmit.edu.au/~ok; RMIT Comp.Sci.

Re: Why don't large companies use Ada?

[Robin Vowels]

	ok@goanna.cs.rmit.oz.au (Richard A. O'Keefe) writes:

>mfeldman@seas.gwu.edu (Michael Feldman) writes:
|>Yes, that's true. The transposition is mental. 
|>In my experience, interfacing between languages - certainly for the
|>kind of fairly simple vector/matrix interfaces the typical Fortranner
|>would use (the _subprograms_ aren't simple but the _interfaces_
|>tend to be) - is, to a very large extent,a matter of getting the
|>data structures to agree.

>I haven't used many PL/I compilers, but they all had "iSUB defining".
|I never used that feature myself, but you could describe it as exposing
|the linear subscript mapping at the source level.  The result is that
|you could have a direct and a transposed view of the _same_ array.
|I would have thought that made it fairly easy to interface PL/I to Fortran.
|PL/I had, back in the 70's when I met it, a great many data types that
|would not have mapped onto Fortran, but every Fortran variable _could_ be
|mapped into PL/I more or less directly (with the aid of iSUB defining for
|>=2-dimensional arrays).

>-- 
|"The complex-type shall be a simple-type."  ISO 10206:1991 (Extended Pascal)
|Richard A. O'Keefe; http://www.cs.rmit.edu.au/~ok; RMIT Comp.Sci.

Re: Why don't large companies use Ada?

[Robin Vowels]

	ok@goanna.cs.rmit.oz.au (Richard A. O'Keefe) writes:

|mfeldman@seas.gwu.edu (Michael Feldman) writes:
|>Yes, that's true. The transposition is mental. 
|>In my experience, interfacing between languages - certainly for the
|>kind of fairly simple vector/matrix interfaces the typical Fortranner
|>would use (the _subprograms_ aren't simple but the _interfaces_
|>tend to be) - is, to a very large extent,a matter of getting the
|>data structures to agree.

|I haven't used many PL/I compilers, but they all had "iSUB defining".
|I never used that feature myself, but you could describe it as exposing
|the linear subscript mapping at the source level.  The result is that
|you could have a direct and a transposed view of the _same_ array.
|I would have thought that made it fairly easy to interface PL/I to Fortran.
|PL/I had, back in the 70's when I met it, a great many data types that
|would not have mapped onto Fortran, but every Fortran variable _could_ be
|mapped into PL/I more or less directly (with the aid of iSUB defining for
|>=2-dimensional arrays).

|-- 
|"The complex-type shall be a simple-type."  ISO 10206:1991 (Extended Pascal)
|Richard A. O'Keefe; http://www.cs.rmit.edu.au/~ok; RMIT Comp.Sci.

Re: Why don't large companies use Ada?

[Matt Kennel]

Keith Thompson (kst@alsys.com) wrote:
: In <3bo4il$3lb@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
: > At the time, I recall speaking with engineers who walked away from 
: > PL/1 _simply_ because it mapped multidimensional arrays row-major, so that
: > interfacing to old Fortran codes required the programmer to tranpose
: > everything on the PL/1 side before passing it to the Fortran subroutine.

: I'll take your word for this, but I still find it difficult to understand.
: If you're writing new PL/I or Ada code to interface to existing Fortran
: libraries, all you should need to do is swap the indices; there should
: be no need to physically transpose the arrays themselves.  This should
: be only a small fraction of the effort required to interface to Fortran
: (for Ada, matching the types and writing the Interface pragmas; for
: PL/I, ???).  Could it be that the engineers were using this as an excuse
: to stick with Fortran, or am I missing something?

No you're not missing anything.  The practical problem is that
*only* difference is that the indices are swapped.

It's like using two pieces of software to do very similar things except
that they have slightly different sign conventions.  This is a trivial
difference, and thus a real nightmare.

Either your Fortran or your PL/I code will be using indices the
"wrong way".

You typically program an algorithm to follow an indexing convention
as defined in a textbook or other scientific document.  You really
need to get these completely right, so it's a major maintenance and
consistency problem that can cause very irritating bugs, if you
can't write the indices the same way everywhere.

Imagine a professor who changed his notation every other lecture, and
the confusion that causes. 

The point is that this is something that is 

* really utterly trivial for a compiler to do (allow matrices to go
whichever way), and

* an authentic pain in the ass to humans when you don't have it,

the fact that subsequent computer langauges and compilers have nonthetheless
given Fortran programmers the finger over and over for decades is really
galling.

I'm very glad to hear that Ada9x will finally stop this madness.

: -- 
: Keith Thompson (The_Other_Keith)  kst@alsys.com
: TeleSoft^H^H^H^H^H^H^H^H Alsys, Inc.
: 10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2718
: /user/kst/.signature: I/O error (core dumped)

--
-Matt Kennel  		mbk@inls1.ucsd.edu
-Institute for Nonlinear Science, University of California, San Diego
-*** AD: Archive for nonlinear dynamics papers & programs: FTP to
-***     lyapunov.ucsd.edu, username "anonymous".

Re: Why don't large companies use Ada?

[Robert Dewar]

Mike, have you REALLY met engineers who walked away from PL/1 because of
the array subscripting order "problem". If so their behavior is peculiar
given the fact that PL/1 allows the specification of subscripting regimes
in full generality using ISUB (swapping indices is a trivial application).

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <3c060k$cf7@gnat.cs.nyu.edu>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>Mike, have you REALLY met engineers who walked away from PL/1 because of
>the array subscripting order "problem". If so their behavior is peculiar
>given the fact that PL/1 allows the specification of subscripting regimes
>in full generality using ISUB (swapping indices is a trivial application).
>

Yes.

But unless I mis-read the post from the IBM chap the other day,
ISUB does more than logically swap the indices, it actually
transposes the array. I don't think I kept that post, but I do 
think that's what he said.

Has PL/1 supported ISUB all along? I do not remember it from my
PL/1 days 20 or so years ago.

Mike Feldman

Re: Why don't large companies use Ada?

[Kenneth Almquist]

Michael Feldman <mfeldman@seas.gwu.edu> wrote:
> But unless I mis-read the post from the IBM chap the other day,
> ISUB does more than logically swap the indices, it actually
> transposes the array.

ISUB lets you declare an array which is really a veiw of another
array.  For example:

	DCL A_FORTRAN(5, 7);
	DCL A(7, 5) DEFINED A_FORTRAN(2SUB, 1SUB);

declares the array A_FORTRAN, and then overlays the array A on top
of it with the subscripts in the opposite order.  This means that
A(3, 2) refers to the same element as A_FORTRAN(2, 3).

Now you can write:

	DCL FSUB ENTRY((5, 7) FLOAT) OPTIONS(FORTRAN NOMAP);
	FSUB(A_FORTRAN);

The FORTRAN option tells the PL/1 compiler that FSUB is a FORTRAN
routine (probably unnecessary, but a good precaution), and the NOMAP
option tells the compiler to pass the arguments directly even though
the routine is in FORTRAN.  Without the NOMAP option, the PL/1 compiler
would create a copy of the array with the subscripts swapped.

> Has PL/1 supported ISUB all along? I do not remember it from my
> PL/1 days 20 or so years ago.

The information in this posting comes from a 1974 manual.
				Kenneth Almquist

Re: Why don't large companies use Ada?

[Robert Dewar]

As far as I know ISUB was always in PL/1, it may not have been in the
G subset.

iSUB in PL/I (was: Re: Why don't large <you-know-whats> use <you-know what>?)

[Norman H. Cohen]

In article <3c610q$gm8@gnat.cs.nyu.edu>, dewar@cs.nyu.edu (Robert Dewar)
writes: 

|> As far as I know ISUB was always in PL/1, it may not have been in the
|> G subset.

In the museum section of my bookshelf (next to the July 1982 Ada RM that
Robert would have me burn :-) ), is a 1972 edition of the PL/I-F LRM, and
it includes iSUB.

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: iSUB in PL/I (was: Re: Why don't large <you-know-whats> use <you-know what>?)

[Robert Dewar]

Two ancient books in the same place!!!
My goodness we will have to send round the firemen :-)

Re: iSUB in PL/I (was: Re: Why don't large <you-know-whats> use <you-know what>?)

[Robin Vowels]

	ncohen@watson.ibm.com (Norman H. Cohen) writes:

	In article <3c610q$gm8@gnat.cs.nyu.edu>, dewar@cs.nyu.edu (Robert Dewar) writes: 

	|> As far as I know ISUB was always in PL/1, it may not have been in the
	|> G subset.

	In the museum section of my bookshelf (next to the July 1982 Ada RM that
	Robert would have me burn :-) ), is a 1972 edition of the PL/I-F LRM, and
	it includes iSUB.


---The first edition (1967) of the PL/I-F LRM had it in.
---The second edition (1968) of the PL/I-F LRM had it in.

Re: Why don't large companies use Ada?

[Paige Vinall]

In <3bmb4r$9kc@gnat.cs.nyu.edu> Robert Dewar writes:
>Note that one significant reason that PL/1 failed was the very poor initial

Please don't use the past tense with PL/I.  IBM makes millions of
dollars on PL/I.  Other companies are also starting to invest-
like Liant and Micro Focus.  I also know of a startup company
that was created this year to build tools for PL/I.

Paige Vinall - PL/I Development Manager, IBM
Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Re: Why don't large companies use Ada?

[Richard Riehle]

In article <19941203.130921.511@almaden.ibm.com> vinall@VNET.IBM.COM (Paige Vinall) writes:
>In <3bmb4r$9kc@gnat.cs.nyu.edu> Robert Dewar writes:
>>Note that one significant reason that PL/1 failed was the very poor initial
>
>Please don't use the past tense with PL/I.  IBM makes millions of
>dollars on PL/I.
>
>Paige Vinall - PL/I Development Manager, IBM
>Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Paige,

  Sorry that some of us (who have good experiences with PL/1 ) have 
  prematurely spoken of PL/1 in the past tense. 

  Can anyone out there confirm that Alsys' first Ada compiler was written
  in PL/1?

  Richard Riehle

Re: Why don't large companies use Ada?

[David Kehs]

In <1994Dec6.021635.6184@sei.cmu.edu> riehler@ajpo.sei.cmu.edu (Richard Riehle) writes:

>  Can anyone out there confirm that Alsys' first Ada compiler was written
>  in PL/1?

Well, not exactly.  In the early days, Alsys developed a tool that translated
from a subset of Ada to PL/1.  This internal tool was written in PL/1.  The
first Alsys full-Ada compiler was written in Ada (really the subset) and was
compiled with the Ada-to-PL/1 translation tool.  The resulting compiler was
eventually used to compile itself and the Ada-to-PL/1 tool was abandoned.

Dave Kehs
Alsys, Inc.

Re: Why don't large companies use Ada?

[Robert Dewar]

"Can someone out there confirm that Alsys
' first compiler was written in PL/1"

No, because it's not true, but there is a germ of truth. Alsys decided
to write in Ada from the very start (they believed in the advantages
of the language, unlike some other vendors who even today are writing
Ada compilers in other languages :-)

That meant they needed a bootstrap path, and the path they used was a
temporary partial subset translator that translated the Ada into PL/1,
so they were using PL/1 not as a source language, but as an intermediate
object language.

Re: Why don't large companies use Ada?

[Bob Dalgleish]

In article <3be9as$jrh@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
>Indeed, many people credit AT&T with popularizing Unix; my recollection
>is that AT&T Unix was something of a research curiosity till Berkeley
>"enhanced" it and - a few years later - Sun Microsystems (which can
>be thought of as originally a Berkeley "spinoff") popularized it as
>a commercial system. Berkeley's funding came (mostly) from ARPA.

AT&T provide Unix Edition 6 to many universities after the BSTJ was
published in 1977(?) -- Bell Systems Technical Journal.  Many thousands
of sites had Unix along with Berkeley.  Edition 7 and System III were
commercially popular, partly because AT&T introduced the binary
licensing scheme that allowed VARs, (Microsoft, and SCO come to mind) to
port the OS to their brand of hardware.  It might have been as late as
1983 before Berkeley versions started getting sold in binary only forms;
up until then it was available only in source form, which required a
source licence from AT&T which cost non-university sites an arm and a
leg ($43K or more).  The second Berkeley Unix that I saw (the first that
I used) was Ultrix in 1985.

Not that I am not gratefull for the Berkeley systems, but some big-name
players felt that Unix was more than a "research curiosity".
-- 
-- * * * CFV: net.short.signatures * * *--
Bob Dalgleish	 Bob_Dalgleish@Develcon.com     CompuServe: 70521,2011

Re: Why don't large companies use Ada?

[Andre Spiegel]

Michael Feldman writes:
[...]
> Berkeley Unix, X Windows, and a raft of other software things over the
                 ^^^^^^^^^
[...]

  I guess to someone who knows how to spell "Ada" correctly, it should
be pointed out that the official name is `the X window system' 
(a window system named `X'), but certainly not `X Windows'.
(The latter suggesting a relation to `MS Windows,' which couldn't be 
further from the truth of course ;-).

--
Andre Spiegel                     |  This life is a test. It is only a test.
                                  |  Had this been an actual life, you would
University of Stuttgart, Germany  |  have received further instructions as to
Computer Science                  |  where to go and what to do.
                                                            -- Author unknown

	   email: spiegel@bruessel.informatik.uni-stuttgart.de


 

Re: Why don't large companies use Ada?

[greg harvey]

Andre Spiegel (spiegel@bruessel.informatik.uni-stuttgart.de) wrote:
: Michael Feldman writes:
: [...]
: > Berkeley Unix, X Windows, and a raft of other software things over the
:                  ^^^^^^^^^
: [...]

:   I guess to someone who knows how to spell "Ada" correctly, it should
: be pointed out that the official name is `the X window system' 
: (a window system named `X'), but certainly not `X Windows'.
: (The latter suggesting a relation to `MS Windows,' which couldn't be 
: further from the truth of course ;-).

Ooops...you've committed an out-of-thread penalty and will receive 
a 10 yard penalty, a technical foul, and a yellow card.

Despite the desire to keep 'X' pure, the current marketing phrase is
indeed 'X Windows' among TOO many companies in the US.  You can't change
it...and it doesn't matter.  The marketers ARE in charge.

Greg Harvey (gharvey@rwi.com)

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <SPIEGEL.94Nov30105105@berlin.bruessel.informatik.uni-stuttgart.de>,
Andre Spiegel <spiegel@bruessel.informatik.uni-stuttgart.de> wrote:
>Michael Feldman writes:

>  I guess to someone who knows how to spell "Ada" correctly, it should
>be pointed out that the official name is `the X window system' 
>(a window system named `X'), but certainly not `X Windows'.

Yep, you're right. Sorry.

>(The latter suggesting a relation to `MS Windows,' which couldn't be 
>further from the truth of course ;-).

Indeed.

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

the X rule (was Re: Why don't large companies use Ada?)

[John Goodsen]

In article <3bj0lu$ruq@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:

   >  I guess to someone who knows how to spell "Ada" correctly, it should
   >be pointed out that the official name is `the X window system' 
   >(a window system named `X'), but certainly not `X Windows'.

   Yep, you're right. Sorry.

Actually, the "official" name in the ranks is just plain "X".
This is also known as the "rule of X".
Anyone who says "window" or "windows" within 2-3 words of "X"
in a sentence is automatically identified as a green X developer.
I'm trying to get the X community to associate some type of 
mandatory obligation to buy the first round of beers to anyone
who breaks the "rule of X".  Any votes?

followup to comp.windows.x

   >(The latter suggesting a relation to `MS Windows,' which couldn't be 
   >further from the truth of course ;-).

   Indeed.

indeed squared
-- 
--
John Goodsen                         Currently on-site at:
The Dalmatian Group                       JP Morgan 
User Interface Specialists                60 Wall St., New York City
jgoodsen@radsoft.com                      jgoodsen@jpmorgan.com

Re: Why don't large companies use Ada?

[Brian J. Zimbelman]

In article <Cz9H4I.5HK@ois.com> beckwb@ois.com (R. William Beckwith) writes:

[much discussion of ADA, why it is/isn't great deleted]

I think one major reason has been omitted from the discussion.  A major factor 
in what language I develop in is what language the company/department already 
has invested in.  Ada developers are rare.  C/C++ developers are all over the 
place.  Therefore, most of the time, my customers want the product in C or 
C++.  The price of software is not how much it costs to develop it, but how 
much it costs to maintain it.  If I can't find a developer who knows the 
language it was developed in, then I chose the wrong language no matter how 
cheap the initial development was.   Oh, and personally I'd choose smalltalk 
over all other languages if it was up to me, but the guy with the check in his 
hands chooses, and that ain't me!

Good Luck,

Brian


Brian J. Zimbelman
Innovative Solutions, Inc.
bjz@innsol.com

Re: Why don't large companies use Ada?

[David Emery]

>Ada developers are rare.  C/C++ developers are all over the place.  

It's a big mistake to equate C and C++ programmers.  There's a
substantial difference in approach between the two languages.  In my
observation, the number of C programmers >> the number of C++
programmers.  What's worse, the number of people claiming to
know C++ >> the number of people who actually understand the language.

Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
sprinkled over the code.

				dave
--
--The preceeding opinions do not necessarily reflect the opinions of
--The MITRE Corporation or its sponsors. 
-- "A good plan violently executed -NOW- is better than a perfect plan
--  next week"                                      George Patton
-- "Any damn fool can write a plan.  It's the execution that gets you
--  all screwed up"                              James Hollingsworth
-------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Martijn Dekker]

emery@goldfinger.mitre.org (David Emery) wrote: 

:>Ada developers are rare.  C/C++ developers are all over the place.  
:
:It's a big mistake to equate C and C++ programmers.  There's a
:substantial difference in approach between the two languages.  In my
:observation, the number of C programmers >> the number of C++
:programmers.  What's worse, the number of people claiming to
:know C++ >> the number of people who actually understand the language.
:
:Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
:sprinkled over the code.
:
:				dave

Couldn't have said it better! Most C++ programs I have seen, use C++ as
nothing more than a convenient preprocessor of C. The ideas of OOP are
not found in those programs.

But, of course, there are people who understand the concept of OOP and
seem to be able to program in an 'OOP-style' using C++.
I an not sure I understand what Object Oriented Programming is all about,
but I one very happy C programmer.

 

-- 
Martijn Dekker         | PFF - software: coding for Atari and X machines
University of Amsterdam|  Mail for info about Ishido  
Math department        | 
mail: mdekker@fwi.uva.nl, 2:280/706.14@fidonet.org, 90:500/104.14@nest.ftn

Stop posting Ada stuff to oracle group!

[Logicon RDA]

Do you guys mind keeping your Ada related stuff to Ada appropriate
groups...

Re: Why don't large companies use Ada?

[Doug Robertson]

In article <3aamuc$14l@matrix.fwi.uva.nl>
           mdekker@fwi.uva.nl "Martijn Dekker" writes:

> emery@goldfinger.mitre.org (David Emery) wrote: 
> 
> :>Ada developers are rare.  C/C++ developers are all over the place.  
> :
> :It's a big mistake to equate C and C++ programmers.  There's a
> :Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
> :sprinkled over the code.
> :
> :                               dave
> 
> Couldn't have said it better! Most C++ programs I have seen, use C++ as
> nothing more than a convenient preprocessor of C. The ideas of OOP are
> not found in those programs.
> 
> But, of course, there are people who understand the concept of OOP and
> seem to be able to program in an 'OOP-style' using C++.
> I an not sure I understand what Object Oriented Programming is all about,
> but I one very happy C programmer.
> 
> Martijn Dekker         | PFF - software: coding for Atari and X machines
> 
I'm not sure I'm clear on OOP - but I *THINK* I've now managed to
migrate from being a C programmer to at least a semi-C++ one.

I've read several OOP/OOD books - but I'm still not sure that I'm
clear on exactly what OO C++ *IS*.

Could some kindly soul direct me towards some GOOD examples of
					      ^^^^
OO C++ application source code ?

Object Orientation is one of those things where almost everyone seems
to be able to identify what isn't, but few people seem to be able to
identify what is.

-- 
Doug Robertson

Re: Why don't large companies use Ada?

[Scott McLoughlin]

Doug@rippleco.demon.co.uk (Doug Robertson) writes:

[ Much stuff on finding Ada and C++ programmers and _real_
C++ programmers deleted. ]

> 
> Could some kindly soul direct me towards some GOOD examples of
> 					      ^^^^
> OO C++ application source code ?
> 
> Object Orientation is one of those things where almost everyone seems
> to be able to identify what isn't, but few people seem to be able to
> identify what is.
> 
> -- 
> Doug Robertson

Howdy,
        Take a look at the C++ runtime library for 
"good OO" C++ code. If you don't have access to the
sources for your compiler, you might look at P.J.
Plaugher's text.  A big template based library written
at HP was just dragged into the standard, so you might
try to find those sources if they are available. They
presumably (perhaps, had better be) are "good C++ OO"
code.
        As for the vagueness of "OO" principles --
well, they are a pretty vague jumble of clearer
principles developed in various other programming
language research efforts. I'm not a big theorist,
but a brief overview of the "themes" might be:
(1) Variable scoping policy.
(2) Extensible type systems.
(3) Encapsulation.
(4) Late binding of names to executable code.

There's been all sorts of research on all of these
in many different "directions", as it were. I find
the common thread through all this that most closely
tracks "OOPiness" is theme (4) when linked to theme
(2) in the language design.

The _way_ these themes are played out differs
_greatly_ from language to language, often
rather arbitrarily in my opinion, e.g., there is
no obvious store of "progress" from on language
design to the next.

Some big differences in language design can be
simplified along a few axes:

(1) manifest (typed variables) vs. latent (typed
values) typing.
(2) explicit (delete operator) vs. implicit
(gc) memory management.
(3) distinguised receiver vs. full blown 
multimethod dispatch.

Various languages, all claiming to be inheritors
of OOP principles, have mixed and match the above
three ingredients in various combinations.  These
ingredients effect the way one effectively uses
the language _enourmously_ (and only the last 
ties in tightly to the OOP themes listed above).

Thus, I disagree that there is one true "OOP" way
or school into which one can be "indoctrinated". 
Elegant look pieces of C++/Smalltalk/CLOS/Eiffel
code are going to look, "feel" and be designed
very differently.

Whatever "OOP"'s true merits (I think it has
many, BTW) -- it's hyped up success in the 
marketplace, I think, is largely due to its
ability the deliver large chunks of "automatic"
functionality with specialization options, which
has been useful restling with the advent of
_enourmous_ operating system API's; GUI's are the
most visible, but only one example.  

It seems this hasn't proved "dumbed down" enough
for the hoi polloi, so "code generator" type tools
are gaining popularity once again, which could
have been done, of course, using a non-oop language
as the target source language.

Anyway, C++ is a nice systems programming language;
in many ways better than C for the tasks C was 
supposed to address. A problem seems to be the
sheer size of the language, which creates a 
problem for systems tools developers who want
standards compliance from language vendors. It's
alot easier to meet the C standard than to track
the "emerging" C++ standard.

I don't happen to think C++ is a great "general
purpose" programming language, mainly because it
is such a great low level language.  A "general
purpose" app language should include garbage
collection and better multidimensional arrays.
Maybe when the hype dies down, Eiffel, Sather,
Modula-3, Lisp, Smalltalk, etc. will have their
day.  (Not a criticism. I use C++ everyday, but
I wouldn't wish it on my dad's MIS crew!)

Sorry for the long post. I've been thinking 
about languages alot recently.

=============================================
Scott McLoughlin
Conscious Computing
=============================================

Re: Why don't large companies use Ada?

[Peter Hermann]

Scott McLoughlin (smcl@sytex.com) wrote:

[good stuff deleted]

: Maybe when the hype dies down, Eiffel, Sather,
: Modula-3, Lisp, Smalltalk, etc. will have their
: day.  (Not a criticism. I use C++ everyday, but
: I wouldn't wish it on my dad's MIS crew!)

are you omitting Ada94 by intention?

--
Peter Hermann  Tel:+49-711-685-3611 Fax:3758 ph@csv.ica.uni-stuttgart.de
Pfaffenwaldring 27, 70569 Stuttgart Uni Computeranwendungen
Team Ada: "C'mon people let the world begin" (Paul McCartney)

Re: Why don't large companies use Ada?

[Hugh Miles]

In article <785104767snz@rippleco.demon.co.uk>
           Doug@rippleco.demon.co.uk "Doug Robertson" writes:

> In article <3aamuc$14l@matrix.fwi.uva.nl>
>            mdekker@fwi.uva.nl "Martijn Dekker" writes:
> 
> > emery@goldfinger.mitre.org (David Emery) wrote: 
> > 
> > :>Ada developers are rare.  C/C++ developers are all over the place.  
> > :
> > :It's a big mistake to equate C and C++ programmers.  There's a
> > :Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
> > :sprinkled over the code.
> > :
> > :                               dave
> > 
> > Couldn't have said it better! Most C++ programs I have seen, use C++ as
> > nothing more than a convenient preprocessor of C. The ideas of OOP are
> > not found in those programs.
> > 
> > But, of course, there are people who understand the concept of OOP and
> > seem to be able to program in an 'OOP-style' using C++.
> > I an not sure I understand what Object Oriented Programming is all about,
> > but I one very happy C programmer.
> > 
> > Martijn Dekker         | PFF - software: coding for Atari and X machines
> > 
> I'm not sure I'm clear on OOP - but I *THINK* I've now managed to
> migrate from being a C programmer to at least a semi-C++ one.
> 
> I've read several OOP/OOD books - but I'm still not sure that I'm
> clear on exactly what OO C++ *IS*.
> 
> Could some kindly soul direct me towards some GOOD examples of
>                                               ^^^^
> OO C++ application source code ?
> 
> Object Orientation is one of those things where almost everyone seems
> to be able to identify what isn't, but few people seem to be able to
> identify what is.
> 
> -- 
> Doug Robertson
> 

Just my own thoughts, but OO-mysticism can be seen as an attempt to
centralise the act with the context of the act. How often have you passed
the wrong variable to a C function, or had to write some ghastly defensive
code to check for an exception case? Software engineering and information
engineering disciplines separated data and function so that the intersection 
would form a useful cross-check, and because the programming languages of the
day forced that separation; not because they are independent of one another.


-- 
Hugh Miles   hugh@huge.demon.co.uk   "O for a Muse of fire, that would ascend
                                      The brightest heaven of invention"

Re: Why don't large companies use Ada?

[Paul Johnson]

In article <785104767snz@rippleco.demon.co.uk>
           Doug@rippleco.demon.co.uk "Doug Robertson" writes:

> I'm not sure I'm clear on OOP - but I *THINK* I've now managed to
> migrate from being a C programmer to at least a semi-C++ one.
> 
> I've read several OOP/OOD books - but I'm still not sure that I'm
> clear on exactly what OO C++ *IS*.
> 
> Could some kindly soul direct me towards some GOOD examples of
> OO C++ application source code ?


This is a common problem.  C++ is an OO language, but it is also a
non-OO language (thanks to its C legacy).  Many programmers who learn
C++ find it difficult to figure out what this OO stuff is, but most
prefer not to admit it.


First, take a look in Grady Booch's "Object-oriented Design with
Applications".  You will find several example apps, with code.  One of
these is in C++.

Then read "Object-Oriented Software Construction" by Bertrand Meyer,
and *learn Eiffel*.  This last point is important.

You will want to know how learning Eiffel is going to help you learn
C++.  It won't.  What it will do is help you learn OO design.  There
are a number of things that C++ does very badly (exception handling,
multiple inheritance and templates spring to mind).  As a result these
things are difficult to understand and all C++ books treat them as
advanced topics.  These are all things that Eiffel does well, so if
you learn Eiffel you will learn how these things *should* be done.

You can get "Personal Eiffel for Windows" from Roger Browne
<rogerb@eiffel.demon.co.uk>.  Email him for a price list.

Paul.

-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Object-Oriented Concepts (was: Why don't large companies use Ada?)

[Jules]

In article <785191405snz@huge.demon.co.uk>,
	hugh@huge.demon.co.uk writes:

>Just my own thoughts, but OO-mysticism can be seen as an attempt to
>centralise the act with the context of the act. How often have you passed
>the wrong variable to a C function, or had to write some ghastly defensive
>code to check for an exception case? Software engineering and information
>engineering disciplines separated data and function so that the intersection 
>would form a useful cross-check, and because the programming languages of the
>day forced that separation; not because they are independent of one another.

I would agree that this is one of the original points that started the concept
we now call Object-Oriented Programming, but I would disagree that this forms
a good statement of what it is about.

OOP has evolved beyond this simple combination of data and code. In my opinion,
the most important points of it at it's state of the art today are:

Polymorphism - arising directly from the combination of code and data is the
               ability to work with different types of data, and treat them
               all in the same way - you no longer need to know what type of
               data to process, as it processes itself.
Abstraction -  Abstraction is an import concept, necessary for polymorphism
               to work - a layer of seperation between objects to prevent
               data-dependencies from arising is crucial to being able to
               manipulate new data-types without extensive rewriting.

Admittedly, none of this would be possible without the original combination
of code and data, but this is not I believe the crucial point of OOP.

Or would somebody like to disagree with me?

-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Object-Oriented Concepts (was: Why don't large companies use Ada?)

[R. Kerr]

Jules (csusb@csv.warwick.ac.uk) wrote:

> OOP has evolved beyond this simple combination of data and code. In my opinion,
> the most important points of it at it's state of the art today are:

> Polymorphism - arising directly from the combination of code and data is the
>                ability to work with different types of data, and treat them
>                all in the same way - you no longer need to know what type of
>                data to process, as it processes itself.
> Abstraction -  Abstraction is an import concept, necessary for polymorphism
>                to work - a layer of seperation between objects to prevent
>                data-dependencies from arising is crucial to being able to
>                manipulate new data-types without extensive rewriting.

> Admittedly, none of this would be possible without the original combination
> of code and data, but this is not I believe the crucial point of OOP.

OOP has not evolved this way. These notions have ALWAYS been at its heart 
but are lost to many people who can only think mechanistically.

Cheers....Ron

------------------------------------------------------------------------
  Ron Kerr, Computing Service, Newcastle University, NE1 7RU, England.
 Tel. +44 191 222 8187       Fax. +44 191 222 8765  (NOTE new area code)
------------------------------------------------------------------------

Re: What is OO (Was Why don't large companies use Ada?)

[Don Vick]

In article <785104767snz@rippleco.demon.co.uk>,
Doug Robertson <Doug@rippleco.demon.co.uk> wrote:
>> 
>
>Object Orientation is one of those things where almost everyone seems
>to be able to identify what isn't, but few people seem to be able to
>identify what is.
>

I'm sure lots of people will have answers for this one.  Here's a
historical view (as best I can remember it).

In the mid to late '70s, people who developed large complex software
for a living were trying to find a good way to divide a large system
into smaller pieces to make it more manageable and (theoretically)
allow the coding to be done by teams instead of individuals.  Lots of
people had theories on ways to divide, or decompose a system.
Ken Orr said, "Make the program structure like the output structure."
Michael Jackson (the computer scientist, not the singer) said, "Make the
program structure like the input data structure."
Glen Myers said, "Make the program structure like the functional structure
of the application."
Others (I've forgotten who said what) looked at data flow or process
structure in different ways to get clues on decomposition criteria.
And someone said, "Think of the objects that the application models;
identify the data (what we know about the objects) and the operations
(what we do to the objects), and make *that* your program structure.

This turned out to be a pretty good way to do things - programs could become
more maintainable and understandable, and it was easy to hand out objects
to different programmers for implementation, as long as they implemented
the agreed-upon operations for their object.

Now this object oriented decomposition could be done in any language.
I have done it or seen it done in FORTRAN, COBOL, BASIC, C, assembler,
Unix shell scripts -- you name it.  But as you use this technique, you
start to wish for features in your programming language.  "I wish COBOL
had functions and allowed for private data."  "I wish FORTRAN allowed
longer variable names."  "I wish C made it easier to encapsulate an object."
etc, etc.  So people started making languages that had these features.
Then they wanted to store the data objects they could describe in these
languages, so they started making databases that could do *that*.

In other words, "object oriented" was first of all a decomposition strategy,
and only incidentally a kind of language, or database, or design methodology.

Anyway, that's the way I remember it.
--------------------------------------------------------
Donald E. Vick  (dvick@lanier.com, dvick@crl.com)
Voice: (404) 493-2194    Fax: (404) 493-2399

Re: What is OO (Was Why don't large companies use Ada?)

[R. Kerr]

Don Vick (dvick@lanier.com) wrote:

> I'm sure lots of people will have answers for this one.  Here's a
> historical view (as best I can remember it).

> In the mid to late '70s, people who developed large complex software
> for a living were trying to find a good way to divide a large system
> into smaller pieces to make it more manageable and (theoretically)
> allow the coding to be done by teams instead of individuals.  Lots of
> people had theories on ways to divide, or decompose a system.
> Ken Orr said, "Make the program structure like the output structure."
> Michael Jackson (the computer scientist, not the singer) said, "Make the
> program structure like the input data structure."
> Glen Myers said, "Make the program structure like the functional structure
> of the application."
> Others (I've forgotten who said what) looked at data flow or process
> structure in different ways to get clues on decomposition criteria.
> And someone said, "Think of the objects that the application models;
> identify the data (what we know about the objects) and the operations
> (what we do to the objects), and make *that* your program structure.

> This turned out to be a pretty good way to do things - programs could become
> more maintainable and understandable, and it was easy to hand out objects
> to different programmers for implementation, as long as they implemented
> the agreed-upon operations for their object.

> .....................

> In other words, "object oriented" was first of all a decomposition strategy,
> and only incidentally a kind of language, or database, or design methodology.

> Anyway, that's the way I remember it.

That's not quite the way it was. The fundamental concepts were developed 
in the discrete event simulation context for the purpose of mastering the 
inherent complexity in modelling dynamic systems. The physical and 
conceptual decomposition techniques supported by these concepts were 
recognised to be equally relevant in the software architecture scene 
generally. 

This all happened in the 1960s but it took rather a long time for the rest 
of the world to catch on and, in many cases, they still haven't. The term 
"object-oriented" had not been coined at that time but those of us 
involved in this work did use the word "object" in exactly the way it is 
used now.

Detailed accounts of these developments can be read in:

      "The Development of the SIMULA Languages", K. Nygaard & O-J-Dahl,
      Academic Press, New York, 1981.

and 

      "Compiling  SIMULA: A Historical Study of Technological Genesis",
      Jan Rune Holmevik, IEEE  Annals  of  the  History  of  Computing,
      vol.16, no. 4, 1994.

These papers give an all-round account of the conception, gestation and
birth pangs of the OOP paradigm. (For those who may be unaware of this, 
everything object-oriented is derived from SIMULA in some sense. Even 
now, those wishing to get a sound grasp of OOP principles could do a lot 
worse than have a look at SIMULA which still has a few notions not yet 
adopted generally.)

Cheers....Ron

------------------------------------------------------------------------
  Ron Kerr, Computing Service, Newcastle University, NE1 7RU, England.
 Tel. +44 191 222 8187       Fax. +44 191 222 8765  (NOTE new area code)
------------------------------------------------------------------------

Re: What is OO (Was Why don't large companies use Ada?)

[Sergio R. Sigrist]

R. Kerr (R.Kerr@ncl.ac.uk) wrote:
: Don Vick (dvick@lanier.com) wrote:

: > I'm sure lots of people will have answers for this one.  Here's a
: > historical view (as best I can remember it).

: > In the mid to late '70s, people who developed large complex software
: > for a living were trying to find a good way to divide a large system
: > into smaller pieces to make it more manageable and (theoretically)
: > allow the coding to be done by teams instead of individuals.  Lots of
: > people had theories on ways to divide, or decompose a system.
: > Ken Orr said, "Make the program structure like the output structure."
: > Michael Jackson (the computer scientist, not the singer) said, "Make the
: > program structure like the input data structure."
: > Glen Myers said, "Make the program structure like the functional structure
: > of the application."
: > Others (I've forgotten who said what) looked at data flow or process
: > structure in different ways to get clues on decomposition criteria.
: > And someone said, "Think of the objects that the application models;
: > identify the data (what we know about the objects) and the operations
: > (what we do to the objects), and make *that* your program structure.

: > This turned out to be a pretty good way to do things - programs could become
: > more maintainable and understandable, and it was easy to hand out objects
: > to different programmers for implementation, as long as they implemented
: > the agreed-upon operations for their object.

: > .....................

: > In other words, "object oriented" was first of all a decomposition strategy,
: > and only incidentally a kind of language, or database, or design methodology.

: > Anyway, that's the way I remember it.

: That's not quite the way it was. The fundamental concepts were developed 
: in the discrete event simulation context for the purpose of mastering the 
: inherent complexity in modelling dynamic systems. The physical and 
: conceptual decomposition techniques supported by these concepts were 
: recognised to be equally relevant in the software architecture scene 
: generally. 

: This all happened in the 1960s but it took rather a long time for the rest 
: of the world to catch on and, in many cases, they still haven't. The term 
: "object-oriented" had not been coined at that time but those of us 
: involved in this work did use the word "object" in exactly the way it is 
: used now.

: Detailed accounts of these developments can be read in:

:       "The Development of the SIMULA Languages", K. Nygaard & O-J-Dahl,
:       Academic Press, New York, 1981.

: and 

:       "Compiling  SIMULA: A Historical Study of Technological Genesis",
:       Jan Rune Holmevik, IEEE  Annals  of  the  History  of  Computing,
:       vol.16, no. 4, 1994.

: These papers give an all-round account of the conception, gestation and
: birth pangs of the OOP paradigm. (For those who may be unaware of this, 
: everything object-oriented is derived from SIMULA in some sense. Even 
: now, those wishing to get a sound grasp of OOP principles could do a lot 
: worse than have a look at SIMULA which still has a few notions not yet 
: adopted generally.)

: Cheers....Ron

: ------------------------------------------------------------------------
:   Ron Kerr, Computing Service, Newcastle University, NE1 7RU, England.
:  Tel. +44 191 222 8187       Fax. +44 191 222 8765  (NOTE new area code)
: ------------------------------------------------------------------------

Re: What is OO (Was Why don't large companies use Ada?)

[Paul Johnson]

Don Vick (dvick@lanier.com) wrote:
> [...]

> Others (I've forgotten who said what) looked at data flow or process
> structure in different ways to get clues on decomposition criteria.
> And someone said, "Think of the objects that the application models;
> identify the data (what we know about the objects) and the operations
> (what we do to the objects), and make *that* your program structure.

> This turned out to be a pretty good way to do things [...]

> Now this object oriented decomposition could be done in any language.
> [...]

Yes, but you missed out inheritance.  What you have described is the
object *based* techniques.  Object *oriented* techniques allow the
programmer to describe new classes of objects in terms of their
similarity to existing classes.  This is impossible in some
traditional languages (e.g Ada) and very difficult in others (e.g C).
That was the main push behind the development of OO languages.

Paul.
-- 
Paul Johnson            | GEC-Marconi Ltd is not responsible for my opinions. |
+44 245 473331 ext 3245 +-----------+-----------------------------------------+
Work: <paj@gec-mrc.co.uk>           | You are lost in a twisty maze of little
Home: <Paul@treetop.demon.co.uk>    | standards, all different.

Re: What is OO (Was Enough crap about Ada already!)

[David Weller]

In article <5918@gec-mrc.co.uk>, Paul Johnson <paj@gec-mrc.co.uk> wrote:
>Don Vick (dvick@lanier.com) wrote:
>> [...]
>
>> Others (I've forgotten who said what) looked at data flow or process
>> structure in different ways to get clues on decomposition criteria.
>> And someone said, "Think of the objects that the application models;
>> identify the data (what we know about the objects) and the operations
>> (what we do to the objects), and make *that* your program structure.
>
>> This turned out to be a pretty good way to do things [...]
>
>> Now this object oriented decomposition could be done in any language.
>> [...]
>
>Yes, but you missed out inheritance.  What you have described is the
>object *based* techniques.  Object *oriented* techniques allow the
>programmer to describe new classes of objects in terms of their
>similarity to existing classes.  This is impossible in some
>traditional languages (e.g Ada) and very difficult in others (e.g C).
>That was the main push behind the development of OO languages.
>

Careful, Paul.  Ada is OO now with Inheritance, Run-time dispatching,
and lightweight threads.  It ain't a "traditional" language any more.
In fact, it's now the first ISO standardized OO language.  finger my
account for more details.

Having said all that, you're correct that it is possible to implement
OOA/D with a "non-OO" language.  Many, many Ada projects since 1983
have been doing that.  Of course, Ada was already pretty close to OO
with abstraction and genericity features.  Trying it with something
like C is a little harder (been there, done that).  Trying it with
assembler or COBOL is just plain nuts (but it has been done!).

Oh, I've modified the Followup-To over to comp.object, where this
discussion is more appropriate.  I hope others will do the same.

-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
	|"Quitting C++ isn't so difficult, provided you show as much |
	|	persistence stopping as you did starting." dweller   |

Re: What is OO (Was Why don't large companies use Ada?)

[Andrew Dunstan]

In article <5918@gec-mrc.co.uk>, paj@gec-mrc.co.uk (Paul Johnson) writes:

|> Yes, but you missed out inheritance.  What you have described is the
|> object *based* techniques.  Object *oriented* techniques allow the
|> programmer to describe new classes of objects in terms of their
|> similarity to existing classes.  This is impossible in some
|> traditional languages (e.g Ada) and very difficult in others (e.g C).
|> That was the main push behind the development of OO languages.

Your statement implies that this is more possible in C than in Ada83.
Would you care to justify that? I've written a lot in both, and AFAIK
there is nothing at all more you can do in this line in C than in Ada83.
Rather the reverse.

cheers

andrew

Re: What is OO (Was Why don't large companies use Ada?)

[Igor Chudov]

Andrew Dunstan (andrew@bugalugs.apana.org.au) wrote:
: Your statement implies that this is more possible in C than in Ada83.
: Would you care to justify that? I've written a lot in both, and AFAIK
: there is nothing at all more you can do in this line in C than in Ada83.
: Rather the reverse.

Since you can implement Turing machine in both there can be nothing new under 
the moon any more.
----------------------------------------------------------------------------
Igor Chudov,            Resource Solutions Intl         office (918)588-2309
Systems Engineer,       for WilTel.                     home   (918)585-5862
                                          E-mail:     igor_chudov@wiltel.com
                                          http://m-net.arbornet.org/~ichudov
                                     1819 South Jackson #32-P Tulsa OK 74107

                 f u cn rd ths, u cn gt a gd jb n cmptr prgrmmng.

Re: What is OO (Was Why don't large companies use Ada?)

[David J Hopwood]

In article <3bfbjr$306@gateway.wiltel.com>,
Igor Chudov <igor_chudov@wiltel.com> wrote:
>Andrew Dunstan (andrew@bugalugs.apana.org.au) wrote:
>: Your statement implies that this is more possible in C than in Ada83.
>: Would you care to justify that? I've written a lot in both, and AFAIK
>: there is nothing at all more you can do in this line in C than in Ada83.
>: Rather the reverse.
>
>Since you can implement Turing machine in both there can be nothing new under 
>the moon any more.

I do wish people would stop quoting this. It really gets on my nerves.
Turing completeness has little or no relevance to whether one language is
better than another for writing useful and understandable programs.

David Hopwood
david.hopwood@lmh.oxford.ac.uk

Re: What is OO (Was Why don't large companies use Ada?)

[Shawn Willden]

Andrew Dunstan (andrew@bugalugs.apana.org.au) wrote:
: In article <5918@gec-mrc.co.uk>, paj@gec-mrc.co.uk (Paul Johnson) writes:

: |> Yes, but you missed out inheritance.  What you have described is the
: |> object *based* techniques.  Object *oriented* techniques allow the
: |> programmer to describe new classes of objects in terms of their
: |> similarity to existing classes.  This is impossible in some
: |> traditional languages (e.g Ada) and very difficult in others (e.g C).
: |> That was the main push behind the development of OO languages.

: Your statement implies that this is more possible in C than in Ada83.
: Would you care to justify that? I've written a lot in both, and AFAIK
: there is nothing at all more you can do in this line in C than in Ada83.
: Rather the reverse.

It's definitely *easier* to do in C, however.  For example:

typedef struct abstractBaseClassStruct
    {
    int (*foo)(struct abstractBaseClassStruct*);
    } AbstraceBaseClass;

typedef struct
    {
    int (*foo)(AbstractBaseClass*);
    int x;
    } DerivedClass;
int derivedFoo(AbstractBaseClass* a) { return ((DerivedClass*)a)->x; }

void initBase(struct AbstractBaseClass* a) { a.foo = 0; }
void initDerived(DerivedClass* a)   { a.foo = derivedFoo; }

int main()
    {
    DerivedClass d;
    AbstractBaseClass* a;

    initDerived(&d);
    a = (AbstractBaseClass*)&d;   /* Okay since structures are same */

    a->foo(a);                    /* calls derivedFoo */
    }

Doing this in C for a non-trivial program is hard.  Doing it in Ada
would be even more difficult, involving some sort of selectors that
would be passed to a global function to emulate function pointers,
and I'm not sure it could even be guaranteed to work (are Ada records
guaranteed to always have the same layout?).

--
Shawn Willden
swillden@icarus.weber.edu

Question about class destructors...Easy???

[Roy J. Davis]

Hi,

	I have a question regarding class destructors in c++. Can a
	destructor receive any arguments?? And if one can, how would
	it be done and in what instance would it be used?

	Thanks to any and all who reply.

Roy D.

Re: Question about class destructors...Easy???

[zhebu]

No, you can't pass arguments to a destructor.

If you mean the operator delete, the arguments are a pointer to the 
memory-block and the size of the block to be released.

--
NAME   Urs Eberle
EMAIL  urs.eberle@zhflur.ubs.ubs.ch
PHONE  ++41-1-236-58-08

Re: Question about class destructors...Easy???

[Hartmut Kocher US/ESA 60/1L/2? #5629]

In article <3bfjul$kfh@lab2.cs.purdue.edu>, davisrj@cs.purdue.edu (Roy J. Davis) writes:
> 
> 	I have a question regarding class destructors in c++. Can a
> 	destructor receive any arguments?? And if one can, how would
> 	it be done and in what instance would it be used?

Destructors don't take any argument. The only way to pass 
some information to a destructor would be via data, e.g.,
data members that are read by the destructor, static or global
variables.

I don't see a reason why a destructor would need a parameter.
Whereas it makes sense to have different constructors to create
an object from different contexts, once it is constructed, it
should be in a consistent state and the destructor shouldn't
care how it was constructed in the first place. 

Also, as I said above, you always have the possibility to set 
a special data member during construction of the object (or
dynamically at runtime) to tell the destructor how it should
behave.


-- 
+==============================|==============================+
| Dr. Hartmut Kocher           |                              |
| Technical Consultant         | All opinions expressed here  |
| Rational GmbH                | are my own.                  |
| Rosenstrasse 7               |                              |
| 82049 Pullach im Isartal     | I know you guessed it,       |
| Germany                      | but it keeps my lawyer happy.|
| Email: kocher@lts.sel.alcatel.de                            |
+==============================|==============================+

Re: Question about class destructors...Easy???

[Chamundi Sabanathan]

Roy J. Davis (davisrj@cs.purdue.edu) wrote:

: Hi,

: 	I have a question regarding class destructors in c++. Can a
: 	destructor receive any arguments?? And if one can, how would
: 	it be done and in what instance would it be used?

: 	Thanks to any and all who reply.

: Roy D.

Interesting question.  Perhaps if the destructor is to act on an element in
a linked list or other repetitive structure, one might wish to tell it to
(or not to) act likewise on any other elements in the structure which share a
particular characteristic... or on the next/previous n elements, or on every
nth element...

As my c++ is minimal (and rusty to boot), I'm not sure just how this would be
implemented; I await further postings with interest.

Chamundi Sabanathan
Santa Rosa Junior College
csabanat@nermal.santarosa.edu

Re: Question about class destructors...Easy???

[\x01]

  Roy -
 
    A destructor can have arguments although its extremely rare for one to
do so and the only time it would be used is when it was called explicitly.
If you've got some kind of class with special memory requirements (say, in
a shared memory segment under OS/2 or UNIX) you could likely have a need
to explicitly call the destructor or a class and, depending on how you
handle these needs, might want a parameter in that destructor.
  later,

   Ben Scherrey
   Proteus Technologies, Inc.

Re: Question about class destructors...Easy???

[zhebu]

 (bscherrey@bix.com) wrote:
:   Roy -
:  
:     A destructor can have arguments although its extremely rare for one to
..
In C++ destructors may not have arguments. The mentioned "special memory
requirements" can be handled in better ways (i.e. a semaphore for Unix
shared memory). This has nothing to do with the original question.

--
NAME   Urs Eberle
EMAIL  urs.eberle@zhflur.ubs.ubs.ch
PHONE  ++41-1-236-58-08

Re: Question about class destructors...Easy???

[Steve Clamage]

bscherrey@bix.com (\x01) writes:

> 
>    A destructor can have arguments although its extremely rare for one to
>do so ...

It should be very rare indeed, because it is not allowed by the
language defintion.

--
Steve Clamage, stephen.clamage@eng.sun.com

Re: What is OO (Was Why don't large companies use Ada?)

[Robert Dewar]

the notion that inheritance is impossible in Ada 83 is bogus, obviously
you can program inheritance using nested components, it's not as pretty,
and you have to mess to get dispatching, but it's certainly doable.

Re: Why don't large companies use Ada?

[Gerrit Thomson]

In article <785104767snz@rippleco.demon.co.uk> Doug@rippleco.demon.co.uk (Doug Robertson) writes:
>From: Doug@rippleco.demon.co.uk (Doug Robertson)
>Subject: Re: Why don't large companies use Ada?
>Date: Thu, 17 Nov 1994 20:39:27 +0000

>In article <3aamuc$14l@matrix.fwi.uva.nl>
>           mdekker@fwi.uva.nl "Martijn Dekker" writes:

>> emery@goldfinger.mitre.org (David Emery) wrote: 
>> 
>> :>Ada developers are rare.  C/C++ developers are all over the place.  
>> :
>> :It's a big mistake to equate C and C++ programmers.  There's a
>> :Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
>> :sprinkled over the code.
>> :
>> :                               dave
>> 
>> Couldn't have said it better! Most C++ programs I have seen, use C++ as
>> nothing more than a convenient preprocessor of C. The ideas of OOP are
>> not found in those programs.
>> 
>> But, of course, there are people who understand the concept of OOP and
>> seem to be able to program in an 'OOP-style' using C++.
>> I an not sure I understand what Object Oriented Programming is all about,
>> but I one very happy C programmer.
>> 
>> Martijn Dekker         | PFF - software: coding for Atari and X machines
>> 
>I'm not sure I'm clear on OOP - but I *THINK* I've now managed to
>migrate from being a C programmer to at least a semi-C++ one.

>I've read several OOP/OOD books - but I'm still not sure that I'm
>clear on exactly what OO C++ *IS*.

>Could some kindly soul direct me towards some GOOD examples of
>                                              ^^^^
>OO C++ application source code ?

>Object Orientation is one of those things where almost everyone seems
>to be able to identify what isn't, but few people seem to be able to
>identify what is.

>-- 
>Doug Robertson
For object oriented understanding perhaps you should refer to a software 
engineering text of recent publication. Object oriented design understanding 
is fundamental to understanding object oriented implementation.
    I personally predict the end of "c" & "c++" as presure from 
non-scienctific users means we software developers have to look to a form of 
safety critical quality control. I am scared of the number of layers of 
software I am currently using that were writtne in "c" ( msdos, windows, 
winsock, novell networking ). I put up with many crashes per day which I 
understand but as a user am very frustrated with.
    Ada by design is not exactly object oriented ( the new a9x apparently 
implements object in ada). Ada does promote software engineering strategies 
which lead to less buggy code. Software engineering will change our basic 
language requirements or the software industry will never gain public trust.   
my 2c worth,
    Cheers,
       Gerrit Thomson

Re: Why don't large companies use Ada?

[Todd Dunnavant]

>>Could some kindly soul direct me towards some GOOD examples of
>>                                              ^^^^
>>OO C++ application source code ?

The Rogue Wave class libraries are good.  Borland's OWL seems to be reasonably 
designed, as well.  OWL, being a framework, has a lot of interaction between 
the various classes.  In you choose to be ambitious and look at OWL or some 
other highly object oriented framework, I recommend that you use a tool like 
Rational ROSE or Paradigm Plus to reverse engineer the source code in order to 
look at the class relations.  Reverse engineering really provides a jump start 
toward understanding what it is that you are looking at.



Regards,

Todd Dunnavant   Houston, Texas   tdunnava@neosoft.com

Re: Why don't large companies use Ada?

[Jules]

In article <tdunnava.38.0001651F@Sam.Neosoft.Com>,
	tdunnava@Sam.Neosoft.Com (Todd Dunnavant) writes:
>
>>>Could some kindly soul direct me towards some GOOD examples of
>>>                                              ^^^^
>>>OO C++ application source code ?
>
>The Rogue Wave class libraries are good. Borland's OWL seems to be reasonably 
>designed, as well.  OWL, being a framework, has a lot of interaction between 
>the various classes.  

I would agree with this, Once I understood the initial principles of OOP,
I started using OWL. Even just using the framework forces you into using
good OOP practice, creating a different object for each type of window you
use, and creating specialisations of predefined types.

Reading through the source for these libraries is even better - they are
programmed in a really fluent, and easy to understand style, that makes you
realise what each object is for and how it interacts with the others.

However, I would urge you NOT to consider using Borland's TurboVision
framework. Although similar in style to OWL, it is nowhere near as well
written, and is much more complicated to understand. There are many more
classes in this library and the interactions are much more complex. Attempt
to understand it only when you have completely mastered OWL.



-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[John Goodsen]

In article <3b7ou2$f27@holly.csv.warwick.ac.uk> csusb@csv.warwick.ac.uk (Jules) writes:

   In article <tdunnava.38.0001651F@Sam.Neosoft.Com>,
	   tdunnava@Sam.Neosoft.Com (Todd Dunnavant) writes:
   >
   >>>Could some kindly soul direct me towards some GOOD examples of
   >>>                                              ^^^^
   >>>OO C++ application source code ?

Tune into the Design Patterns work: A starting point is the ftp site

      ftp://st.cs.uiuc.edu/pub/patterns

Use these papers as pointers into patterns of various C++ frameworks,
including ET++ and Interviews.  GoPATH, a structured C++ graphics framework,
is also describable by many of the patterns documented at that site.
All are real good examples of clean C++ framework source code.
-- 
--
John Goodsen                         Currently on-site at:
The Dalmatian Group                       JP Morgan 
User Interface Specialists                60 Wall St., New York City
jgoodsen@radsoft.com                      jgoodsen@jpmorgan.com

Re: Why don't we take this thread to Ada and c only?

[Michael R. Fronheiser]

Please.

Thanks,

Mike

Re: Why don't large companies use Ada?

[Jules]

In article <EMERY.94Nov15094017@goldfinger.mitre.org>,
	emery@goldfinger.mitre.org (David Emery) writes
> ..., the number of C programmers >> the number of C++ ...

Do you mean "Shift the number of C programmers right by the number of C++
programmers" (an operation that wil almost certainly result in an
overflow), or retrieve the number of C++ programmers from the stream
"The number of C programmers" (a very sensible thing to do!). You
should have used hungarian - I'd know exactly what you're talking
about then ;-)

>Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
>sprinkled over the code.
>
>				dave

Yes, it probably is, but is this a bad thing? It can't be worse than
just plain old fashioned C. I mean, at least these people are STARTING
to learn OOP, in some way or another, and in a few years it may just hit
them what it's all about, if they stick at it. THEN they'll be writing
REAL C++.

-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Gary McKee]

In Article <3aqrfl$vo@tripe.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk
(Jules) wrote:
>In article <EMERY.94Nov15094017@goldfinger.mitre.org>,
>        emery@goldfinger.mitre.org (David Emery) writes
>> ..., the number of C programmers >> the number of C++ ...
>
>Do you mean "Shift the number of C programmers right by the number of C++
>programmers" (an operation that wil almost certainly result in an
>overflow), or retrieve the number of C++ programmers from the stream
>"The number of C programmers" (a very sensible thing to do!). You
>should have used hungarian - I'd know exactly what you're talking
>about then ;-)
================================================================
Actually, "C programmers >> the number of C++ " 
will cause an underflow, NOT an overflow as claimed.

Where did you learn how to read C/C++ code anyway, in an Ada class?  :-)

Gary McKee
--------------------------------------------------------------------
McKee Consulting                        (303) 795-7287
P. O. Box 3009                          gmckee@cloudnine.com
Littleton, CO 80161-3009                
--------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Jules]

In article <gmckee.1135999149A@news-2.csn.net>,
	gmckee@cloudnine.com (Gary McKee) writes:
>Actually, "C programmers >> the number of C++ " 
>will cause an underflow, NOT an overflow as claimed.
>
>Where did you learn how to read C/C++ code anyway, in an Ada class?  :-)
>
No - I'm doing a Pascal class, and want to do Ada (because Pascal's naff),
and I am having to resort to C/C++ to do all the things that Pascal won't
(ie Random Access Files, with names that you can actually change).


-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[gjong]

Dear David:

Thought you might know of some people who might be qualified and interested 
in several Oracle related jobs we have available at Xircom.  I've attached 
job descriptions for the available jobs.  Would appreciate it if you would 
pass them along to potential leads/candidates, or refer leads to me for 
follow up.  All 3 positions offer extraordinary opportunities within a 
company experiencing geometric growth.  

Regards, 

G. Jong
Director of Recruitment
Xircom


ORACLE Implementation Project Leader:

Will have overall responsibility for implementing and upgrading Oracle 
Application Systems world-wide (US, Europe, Asia) including: 

*  Developing  project schedules and organize projects to achieve 
   major project goals

*  Managing Oracle implementation projects in Europe and Asia

*  Managing Oracle migration projects to new releases world-wide

*  Communicating project status to project steering committee

*  Conducting  regular project meetings with project team members

*  Tracking  project/system issues and insure resources are available 
   to resolve/complete work in a timely manner

Requirements:

* Bachelor's degree in a related field

* Five years experience in systems analysis within a manufacturing 
  environment, with a minimum of two years programming experience

* Previous experience implementing and upgrading  application 
   systems, ideally, Oracle on a multi-national level

* Capability and willingness to travel abroad; specifically, Europe and  
Asia
 
Salary: $75k - 90K with stock participation.
*************************************************************************
Systems Analyst:

* A BS degree in Computer Science/Information Systems

* A minimum of 2 years programming experience (ORACLE Applications in a 

  Windows/Novell/Unix environment is highly desirable) 

* Three years experience in systems analysis  

* Previous experience within a manufacturing environment.

The position requires a team player who is customer-service and quality 
oriented with the ability to solve problems and achieve results in a 
timely fashion.  

The selected individual will have overall responsibility for the 
implementation and support of the application modules for one of three 
areas of users (Operations, Finance, or Sales/Marketing).  

Salary:  $50-65k with stock participation
**********************************************************************

Network Systems Engineer :

A BS degree and a minimum of 4 years experience in a 200+ user network, a 
minimum of one year of experience inNetWare administration, and CNE 
certification.  Familiarity with Uix, Novell, Windows (NT), OS/2, Internet 
and Synoptics/Cisco hardware is highly desirable.

The selected individual will have the opportunity to architect a "Premier 
Multinational Network" which will be robust (eg. 100% availability with 
sub-second response time) and feature-rich (eg. FDDI or ATM backbone, 
Ethernet/FDDI/ATM switches, routers, "super servers", server for HD-to-HD 
backup, tape drives, Unix, Netware, NT, OS/2, Windows, MacIntosh, email). 

Salary:  $65-75k with stock participation

Interested/qualified candidates are welcome to email resumes to: 
jobs@xircom.com; fax resumes to 805-376-9030; or mail resumes to: Xircom, 
2300 Corporate Center Drive, Thousand Oaks, CA 91320.
  

In article <EMERY.94Nov15094017@goldfinger.mitre.org>, 
<emery@goldfinger.mitre.org> writes:
> Path: 
news.cerf.net!io.rdc.puc-rio.br!lakesis.fapesp.br!fnnews.fnal.gov!usenet.ee
l.ufl.edu!news.mathworks.com!news.duke.edu!eff!blanket.mitre.org!linus.mitr
e.org!linus!mbunix!emery
> From: emery@goldfinger.mitre.org (David Emery)
> Newsgroups: 
comp.lang.ada,comp.lang.c,comp.programming,comp.lang.c++,comp.object,comp.d
atabases.sybase,comp.databases.oracle,comp.client-server
> Subject: Re: Why don't large companies use Ada?
> Date: 15 Nov 94 09:40:17
> Organization: The Mitre Corp., Bedford, MA.
> Lines: 20
> Message-ID: <EMERY.94Nov15094017@goldfinger.mitre.org>
> References: <3a6oc5$dkh@nntp1.u.washington.edu> <Cz9H4I.5HK@ois.com>
> 	<bjz.153.00154405@innsol.com>
> NNTP-Posting-Host: goldfinger.mitre.org
> In-reply-to: bjz@innsol.com's message of Tue, 15 Nov 1994 04:15:48 GMT
> Xref: news.cerf.net comp.lang.ada:19682 comp.lang.c:86055 
comp.programming:12557 comp.lang.c++:90684 comp.object:21370 
comp.databases.sybase:12826 comp.databases.oracle:22250 
comp.client-server:6800
> 
> >Ada developers are rare.  C/C++ developers are all over the place.  
> 
> It's a big mistake to equate C and C++ programmers.  There's a
> substantial difference in approach between the two languages.  In my
> observation, the number of C programmers >> the number of C++
> programmers.  What's worse, the number of people claiming to
> know C++ >> the number of people who actually understand the language.
> 
> Thus a lot of alleged C++ is nothing more than bad C with 'OOP-sugar'
> sprinkled over the code.
> 
> 				dave
> --
> --The preceeding opinions do not necessarily reflect the opinions of
> --The MITRE Corporation or its sponsors. 
> -- "A good plan violently executed -NOW- is better than a perfect plan
> --  next week"                                      George Patton
> -- "Any damn fool can write a plan.  It's the execution that gets you
> --  all screwed up"                              James Hollingsworth
> -------------------------------------------------------------------------
> 

Re: Why don't large companies use Ada?

[David Weller]

In article <bjz.153.00154405@innsol.com>,
Brian J. Zimbelman <bjz@innsol.com> wrote:
>In article <Cz9H4I.5HK@ois.com> beckwb@ois.com (R. William Beckwith) writes:
>
>[much discussion of ADA, why it is/isn't great deleted]
>
>I think one major reason has been omitted from the discussion.  A major factor 
>in what language I develop in is what language the company/department already 
>has invested in.  

<Am I the only person that thinks the newsgroup list is excessively
cross-posted?>

By itself, this is a compelling argument regardless of whether it's
Fortran, Assembler, C++, or any other language.  This argument is the
"Appeal to Momentum".  It's a cool argument for something other than
Ada 9X, but with the ability of Ada to interface to C/Fortran/Cobol
(and with SGI's cool demonstration of inheriting from C++ classes),
this becomes a less significant issue.  I claim the argument is less
compelling when considering Ada (but certainly not enough to make a
manager change their mind...yet).

>Ada developers are rare.  C/C++ developers are all over the 
>place.  

I disagree here.  For two reasons:  1) Ada developers aren't "rare",
just less visible.  Unfortunately, I know LOTS of ex-Ada people stuck
in C++ jobs that would jump to Ada again if they could.  2) It's "C"
developers that are "all over the place".  There is, from my personal
observation, a dearth of C++ developers.  Many companies are hiring C
programmers now and training them in C++.  I agree with Bill Beckwith
here -- it takes less time to create a productive Ada developer than
it does to create a productive C++ developer, even if they only have
a C background.

>Therefore, most of the time, my customers want the product in C or 
>C++.  The price of software is not how much it costs to develop it, but how 
>much it costs to maintain it.  If I can't find a developer who knows the 
>language it was developed in, then I chose the wrong language no matter how 
>cheap the initial development was.   Oh, and personally I'd choose smalltalk 
>over all other languages if it was up to me, but the guy with the check in his 
>hands chooses, and that ain't me!
>

Boy, I"m confused.  YOu mean to tell me that it's LESS expensive to
use a langauge that has been empirically proven to be less reliable,
because it's cheaper to find/train a developer in that language?  Why
not get a language that provides more maintainable code, thus
requiring less people (which is the REAL cost factor).



-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
   ObNitPick: Spelling Ada as ADA is like spelling C++ as CPLUSPLUS. :-) 

Re: Why don't large companies use Ada?

[Steven Whatley]

I have done a lot of Ada development.  Personally, I liked ada except for
the package Text_Io.  I hated Text_Io!  Ada has the reputation of expencive
complilers, huge executables, hogs system resources, and "its only for the
government to use."  (i.e. NASA, DoD, etc.)

Also, Ada requires a extensive development environment to take advantage of
all of its features.  (i.e. CASE tool, a language specific editor, a library
manager, a configuration management/version control system, etc.)

Some people associate Ada with the Rational Ada development system.  It is a 
very nice and integrated environment.  But, it cost over a $100,100 for a 
system that can only be used by twelve people.  Six to eight simultaneous
users was optimal.  And don't forget the keyboard!  I have never seen so
many keys plus the number of ESC+control+shift+alt+meta+keypress
combinations!  Personally, I did not mind that but, it is a very scary
sight.  The apollo Rational emulator software was very nice.

A lot of people did not like the strict type checking.  They felt like it
was very combersome and too restrictive.

I have not used Ada 9x so, it may have relieved a lot the problems with 
Ada 83.  I would like to see Ada make a comeback but its reputation as
a dinosuar is probably to strong to allow it.

Just my $0.02 (plus tax) worth,
Steven Whatley

--
 _   _  _ _  __ : Steven Whatley              \ The opinions expressed are my
|_  |_|  |  |   : System Analyst/Programmer    \ own.  You may freely copy,
 _| | | _|_ |__ : saichelp@starbase.neosoft.com \_use,_or_ignore_them._______
Scientific Applications International Corporation (An employee-owned company)

Re: Why don't large companies use Ada?

[David Weller]

In article <3aagtj$h5g@uuneo.neosoft.com>,
Steven Whatley <swhatle@starbase.neosoft.com> wrote:
>I have done a lot of Ada development.  Personally, I liked ada except for
>the package Text_Io.  I hated Text_Io!  Ada has the reputation of expencive
>complilers, huge executables, hogs system resources, and "its only for the
>government to use."  (i.e. NASA, DoD, etc.)
>
That was then, this is now.  Times have changed.  If you have such a
compiler, call your vendor.  Better yet, look around at other
vendors.  They're out there.

>Also, Ada requires a extensive development environment to take advantage of
>all of its features.  (i.e. CASE tool, a language specific editor, a library
>manager, a configuration management/version control system, etc.)
>
Yeah.  Right.  And "serious" development in other languages only need
vi.

Not!

FYI, I do lots of "Small" Ada work at home using only vi and RCS.  I
don't need no steenkin' CASE tools or LSE. :-)

>Some people associate Ada with the Rational Ada development system.  It is a 
>very nice and integrated environment.  But, it cost over a $100,100 for a 
>system that can only be used by twelve people.  Six to eight simultaneous
>users was optimal.  And don't forget the keyboard!  I have never seen so
>many keys plus the number of ESC+control+shift+alt+meta+keypress
>combinations!  Personally, I did not mind that but, it is a very scary
>sight.  The apollo Rational emulator software was very nice.
>
Yup, and IBM sells these 8088 computers with monochrome monitors, and
Apple sells this newfangled computer called the Apple IIc.  Get with
it, Rational won't even sell you an R-1000 anymore.  Their new Ada
development environment (and rumor has it soon to be available for
C/C++), called Apex, is priced somewhere around the same cost as
Centerline's product.  It's available for a variety of platforms:
Sun, HP, IBM RS6000, etc.  It's a very slick and "open" environment.
Extremely "Unix-friendly".

>A lot of people did not like the strict type checking.  They felt like it
>was very combersome and too restrictive.
>
Well, most of that is diminished with the new version of Ada.  The
real issue is that, if you're writing software that must be reliable,
Ada is the best OO language to choose (of course, this open's up the
whole language-flamefest again).

>I have not used Ada 9x so, it may have relieved a lot the problems with 
>Ada 83.  I would like to see Ada make a comeback but its reputation as
>a dinosuar is probably to strong to allow it.
>
Only by posting such mythical rumors does it perpetuate.  Free your
mind -- our lady Ada has finally matured.  It's definitely worth
another look.

"Try it again for the first time", as I'm so fond of saying.

The free GNU gcc-based Ada 94 compiler is available at cs.nyu.edu.
Look under /pub/gnat.  It's available for just about every OS that
gcc is available on, including a version that works under DOS.

-- 
Proud (and vocal) member of Team Ada! (and Team OS/2)        ||This is not your
   	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
For all sorts of interesting Ada tidbits, run the command:   ||________________
"finger dweller@starbase.neosoft.com | more" (or e-mail with "finger" as subj.)
   ObNitPick: Spelling Ada as ADA is like spelling C++ as CPLUSPLUS. :-) 

Re: Why don't large companies use Ada?

[Tony Langdon]

In article erk@Starbase.NeoSoft.COM, dweller@Starbase.NeoSoft.COM (David Weller) writes:
>Yeah.  Right.  And "serious" development in other languages only need
>vi.
>
>Not!
>
Well, if you throw in sccs and a C compiler 

Yes!

---
T.Langdon                              | Email : langdont@jeeves.fi.gs.com
 Goldman Sachs International Limited   | Phone : +44 71 774 5420
 Peterborough Court, 133 Fleet Street, |
 London. EC4B 2AA                      | ...and the Reds go crawling on...

Re: Why don't large companies use Ada?

[Carlos Perez]

<This follow-up is going only to comp.lang.ada>

Steven Whatley (swhatle@starbase.neosoft.com) wrote:
: ...I hated Text_Io! ... 

Text_Io is not part of the Ada language, but rather, a very simple standard 
package for doing very simple I/O.  The designers of the language 
wanted to provide the building blocks and if you wanted something better
than Text_Io, you went out and bought a "reusable" package
from your favorite vendor.  This secondary market never developed and so
I suppose people were forced to make-do with Text_Io.

I think Ada missed an important trend here-- people demand feature
rich systems and so what we get are bloated, do-everything-for-everyone
suites.  Anyone old enough to remember when operating systems provided
an interface to your system instead of zillions of applets, network functions,
and application suites?  Lean is out!  McDouble Meat Bacon Cheesburger is
in! 

: Also, Ada requires a extensive development environment to take advantage of
: all of its features.  (i.e. CASE tool, a language specific editor, a library
: manager, a configuration management/version control system, etc.)

News to me.  The language never required an extensive CASE environment,
DOS and EDLIN are all you need, really.  However, most organizations
who adopted Ada believed in strong software engineering processes in general,
and were more likely to adopt what I call the "Ada-philosophy" which
had little to do with the language itself.  Hence, the use of "cumbersome"
CASE tools and the linkage with Ada.  I think this is called guilt by
association.

: many keys plus the number of ESC+control+shift+alt+meta+keypress
: combinations!  Personally, I did not mind that but, it is a very scary
: sight.  The apollo Rational emulator software was very nice.

Having used a Rational from a 101 keyboard, I admit there was a steep  
learning curve, but once you got use to the "object oriented" keymap, an old
"functional procedural" type like myself could navigate the keyboard
with ease.  I now see why DOS WordPerfect users stick to powerful
macro and function keys.

Anyways, Rational exited the hardware business and now has Apex, so
you can use the mouse to manipulate pull-down menus rather than
twisting your fingers into a pretzel.

: A lot of people did not like the strict type checking.  They felt like it
: was very combersome and too restrictive.

As a veteran Ada programmer, I curse and swear at the strong typing
all the time--- up until I realize that I have a logic error and the
Ada compiler was struggling hard to prevent me from getting into real trouble.
(Like assigning floats into integers, I should think about whether I want
to trunc, round, etc. rather than accept some unknown default operation).

Types in Ada are like garlic, too much and you ruin your dinner (and your
program).   The challenge is to *think* about your data structures
and use subtypes generally and types sparingly.

-- Carlos

Re: Why don't large companies use Ada?

[Brian J. Zimbelman]

In article <3aa7jo$7j@Starbase.NeoSoft.COM> dweller@Starbase.NeoSoft.COM (David Weller) writes:

>In article <bjz.153.00154405@innsol.com>,
>Brian J. Zimbelman <bjz@innsol.com> wrote:
>>In article <Cz9H4I.5HK@ois.com> beckwb@ois.com (R. William Beckwith) writes:
>>
>>[much discussion of Ada, why it is/isn't great deleted]
>>
>>I think one major reason has been omitted from the discussion.  A major factor 
>>in what language I develop in is what language the company/department already 
>>has invested in.  

><Am I the only person that thinks the newsgroup list is excessively
>cross-posted?>
It is very excessively cross-posted.  I've reduced the list, feel free to do 
so again if you follow this message.

>By itself, this is a compelling argument regardless of whether it's
>Fortran, Assembler, C++, or any other language.  This argument is the
>"Appeal to Momentum".  It's a cool argument for something other than
>Ada 9X, but with the ability of Ada to interface to C/Fortran/Cobol
>(and with SGI's cool demonstration of inheriting from C++ classes),
>this becomes a less significant issue.  I claim the argument is less
>compelling when considering Ada (but certainly not enough to make a
>manager change their mind...yet).

>>Ada developers are rare.  C/C++ developers are all over the 
>>place.  

>I disagree here.  For two reasons:  1) Ada developers aren't "rare",
>just less visible.  Unfortunately, I know LOTS of ex-Ada people stuck
>in C++ jobs that would jump to Ada again if they could.  2) It's "C"
>developers that are "all over the place".  There is, from my personal
>observation, a dearth of C++ developers.  Many companies are hiring C
>programmers now and training them in C++.  I agree with Bill Beckwith
>here -- it takes less time to create a productive Ada developer than
>it does to create a productive C++ developer, even if they only have
>a C background.

Well in all the shops I have consulted to, no one has worked on a Ada project 
outside of school, and I can't remember three people who have done that.  On 
the other hand over 200 developers have worked with C, and more than 50 have 
worked with C++ in production environments.

>>Therefore, most of the time, my customers want the product in C or 
>>C++.  The price of software is not how much it costs to develop it, but how 
>>much it costs to maintain it.  If I can't find a developer who knows the 
>>language it was developed in, then I chose the wrong language no matter how 
>>cheap the initial development was.   Oh, and personally I'd choose smalltalk 
>>over all other languages if it was up to me, but the guy with the check in his 
>>hands chooses, and that ain't me!
>>

>Boy, I"m confused.  YOu mean to tell me that it's LESS expensive to
>use a langauge that has been empirically proven to be less reliable,
>because it's cheaper to find/train a developer in that language?  Why
>not get a language that provides more maintainable code, thus
>requiring less people (which is the REAL cost factor).

First, if we alway choose the "Best" tools/environments, NextStep would be 
very popular, NFS would have been run out of town by NCS, Microsoft & Novell 
would be out of business, etc.  Second, when you factor in the costs of 
retraining your existing staff, code developed by "new Ada" programmers vs. 
code developed by "experienced C/C++" programmers, and finding Ada programmers 
when you need to maintain the existing applications, then yes it's far LESS 
expensive to use C/C++ than Ada, Smalltalk, or whatever is the language of the 
month.  

Personally, I've always been a smalltalk fan.  I've suggested it often, but 
none of my clients have had enough smalltalk experience to be convinced that 
they should change.  I did not like Ada8x, and did not keep up to date with 
the language.  I'll have to pick up the GNU Ada system and try it.

>   ObNitPick: Spelling Ada as ADA is like spelling C++ as CPLUSPLUS. :-) 
Doesn't Ada stand for "Assemblers with disabilities Act?" :-)

David, Unless we have more signal to noise, we might want to make this email 
only soon!  Of course, I'm only saying this becouse I'll get the last word in 
that way!

Have Fun,
Brian

Brian J. Zimbelman
Innovative Solutions, Inc.
bjz@innsol.com

Re: Why don't large companies use Ada?

[Dan Thies]

David Weller (dweller@Starbase.NeoSoft.COM) wrote:
: <Am I the only person that thinks the newsgroup list is excessively
: cross-posted?>
I don't know which of the newsgroups you wrote this from, but the answer 
is definitely YES.  This article did not merit such a broad 
cross-posting.  In fact, the only subject large enough to merit that 
will probably be the day that the world government announces that all 
programming languages except SNOBOL are to be banned forever.

: I disagree here.  For two reasons:  1) Ada developers aren't "rare",
: just less visible.  Unfortunately, I know LOTS of ex-Ada people stuck
: in C++ jobs that would jump to Ada again if they could.  2) It's "C"
: developers that are "all over the place".  There is, from my personal
: observation, a dearth of C++ developers.  Many companies are hiring C
: programmers now and training them in C++.  I agree with Bill Beckwith
: here -- it takes less time to create a productive Ada developer than
: it does to create a productive C++ developer, even if they only have
: a C background.
There are not a lot of Ada developers.  There are not a lot of developers 
working with OO languages, period.  It will take time for the educational 
systems of the world to catch up.  In the meantime I think it is 
absolutely true that it takes less time to "create" a productive Ada 
developer, simply because of Ada's high level of abstraction.  Even when 
I am working in C++, I use Ada as pseudo-code because it's easier to 
write it that way.

: >Therefore, most of the time, my customers want the product in C or 
: >C++.  The price of software is not how much it costs to develop it, but how 
: >much it costs to maintain it.  If I can't find a developer who knows the 
: >language it was developed in, then I chose the wrong language no matter how 
: >cheap the initial development was.   Oh, and personally I'd choose smalltalk 
This is rich.  Customers want the product in an executable form, and 
don't have the slightest idea with C or C++ even looks like.  Customers 
also want maintainability, even if they don't know it when the project 
starts off.

: Boy, I"m confused.  You mean to tell me that it's LESS expensive to
: use a langauge that has been empirically proven to be less reliable,
: because it's cheaper to find/train a developer in that language?  Why
: not get a language that provides more maintainable code, thus
: requiring less people (which is the REAL cost factor).
The company I work for is going through major hassles (have been for a 
long time) over all of the software systems we are using.  The main issue 
is that no update ever fixes all the old bugs, and every update adds new 
bugs.  A lot of the software we use has to function in real time, such as 
cash registers.  I think that the inherent confirmability of Ada would 
have eliminated a lot of the recurring and new bugs.  I am certain that a 
more competent C developer would not have so many problems, but it would 
be easier to deliver the goods using Ada.

:    	      Ada -- Very Cool.  Doesn't Suck.               ||  father's Ada 
And all those years I was stuck with C, it was right there under my nose.

Dan
rtfm@cyberspace.com
-- 
rtfm@cyberspace.com

Re: Why don't large companies use Ada?

[Pete Gontier]

In article <3ac3i2$48i@case.cyberspace.com>,
rtfm@cyberspace.com (Dan Thies) wrote:

> : > ...Most of the time, my customers want the product in C or C++.  
> Customers want the product in an executable form, and 
> don't have the slightest idea with C or C++ even looks like.

You are failing to draw a distinction between (1) what the customer would
want if she had a clue, (2) what the customer wants, and (3) what the
customer says she wants. Often she *says* she wants a system written in
C/C++ because it's a buzzword with which she is familiar. This is of
course a sub-optimal methodology for requirements specification. :-)

Of course, there's always the possibility that the customer actually has a
reason for favoring C/C++ over Ada. Off the top of my head, it might be
the case that the customer expects in-house people to do the maintenance,
and the in-house people don't know Ada. Pick any of a number of other
reasons cited already in this thread.

I agree that customers are often completely clue-free when it comes to
determining and/or asking for what they want. However, the safest thing to
give them is almost always what they ask for. If what they ask for is
unreasonable, then you can try gently pushing them in another direction.
But other than that, or if that fails, second-guessing them just gets you
in trouble later.

-- 

 Views expressed here do not necessarily reflect those of my employer.

Re: Why don't large companies use Ada?

[Dirk Broer]

In article <pgontier-1611941121020001@avail.wc.novell.com>, pgontier@novell.com (Pete Gontier) writes...
> 
>Of course, there's always the possibility that the customer actually has a
>reason for favoring C/C++ over Ada. Off the top of my head, it might be
>the case that the customer expects in-house people to do the maintenance,
>and the in-house people don't know Ada. Pick any of a number of other
>reasons cited already in this thread.

Around here (NASA at Goddard Spaceflight Center) the adversion is in two parts:
Risk:  mentality is anything 'new' entails risk - and from what they've (upper 
managment) heard Ada and C++ are risky.  2nd - there is the in-house support 
aspect.  Although programmers might be available much of the algorithms are 
written by engineers who have trouble understanding the difference between 
passing a pointer and passing a value.  C++ scares them - let alone Ada.

I'm now in the final debuging stage of a large C program (DOS).  It uses 14 
databases based on linked list - each slightly different.  End result is me 
cutting and pasting 14 sets of linked-list code...  Although the program is 
modular and easily translatable into C++ I was told no because the other 
engineers would not be able to debug it.  To this day, they have yet to look at 
the code...  Due to the overall size and complexity I doubt they ever will.

>I agree that customers are often completely clue-free when it comes to
>determining and/or asking for what they want. However, the safest thing to
>give them is almost always what they ask for. If what they ask for is
>unreasonable, then you can try gently pushing them in another direction.
>But other than that, or if that fails, second-guessing them just gets you
>in trouble later.

What your alluding to is the difference between requirements and specifications
  The customer should state what the requirements are.  The service provider 
should specify how it will meet those requirements.  We all know this isn't how 
it works.  You must first gain the customers' confidence in your abilities.  
Then you gently nudge the requirements - and have the customer leave the 
specifications up to you...  Carry the big stick of money savings and you 
basically convince the customer.

Dirk

Re: Why don't large companies use Ada?

[Erland Sommarskog]

Brian J. Zimbelman (bjz@innsol.com) writes:
>I think one major reason has been omitted from the discussion.  A major factor
>in what language I develop in is what language the company/department already
>has invested in.  Ada developers are rare.  C/C++ developers are all over the
>place.  Therefore, most of the time, my customers want the product in C or
>C++.  The price of software is not how much it costs to develop it, but how
>much it costs to maintain it.  If I can't find a developer who knows the
>language it was developed in, then I chose the wrong language no matter how
>cheap the initial development was.

I'd like to contest this. Not the conclusion in the latter part of
the text, but the presumption in the first part. Sure, C programmers
come 13 by the dozen, but good C++ programmers? OK, I only have the
perspective the Swedish market, so I might be very wrong, but here
the situation is that if you know C++ well, you have lots of jobs 
to choose from. Or can make a good living as a consultant.

My point is that just because you know C, doesn't mean that you have
less to learn to become a good C++ programmer, than someone whose
never seen C before. In fact, if you have experience in Eiffel, Simula,
Smalltalk or any other O-O language, you are probably better fitted
than the guy who has been hacking C for 10 years, but cannot spell
to imheritance, eh I mean inheritance.

From this follows that if you have a bunch of C programmers (or Cobol
programmers, an even more common phenonemom), you wish to re-educate, 
Ada or C++ is not much difference. It is a considerable investment 
in both cases.

Of course, in real life managers think it is much easier to upgrade
the C programmer to a C++ programmer than an Ada one, and goes on 
shooting himself in the foot.

Then again, I have a feeling that everyone who say that they are using
C++ are really doing it. They might use it as a C compiler with strict
type checking, which undeniably is a great leap forward. But sometimes
the C++ compiler is too slow, so they don't even use it...

No matter what, another poster was optimistic and hoped that Ada 9x
will become popular. How much I like Ada, I fear that it is too late.
Ada 83 has a few weaknesses, not O-O for instance, but still superiour
to C++ if you ask me. But such considerations has never had much im-
portance in the software industry. Fortran, Cobol, C and now C++,
that just how the story goes. (Although I must hasten to add that in
comparison with the first three, C++ is an enormous improvment.)
-- 
Erland Sommarskog, sommar@enea.se, Stockholm
Pour qui est-ce qui vous croyez que je parle?

Re: Why don't large companies use Ada?

[James A. Krzyzanowski]

Erland Sommarskog (sommar@enea.se) wrote:
: My point is that just because you know C, doesn't mean that you have
: less to learn to become a good C++ programmer, than someone whose
: never seen C before. In fact, if you have experience in Eiffel, Simula,
: <snip>
: From this follows that if you have a bunch of C programmers (or Cobol
: programmers, an even more common phenonemom), you wish to re-educate, 
: Ada or C++ is not much difference. It is a considerable investment 
: in both cases.

: Of course, in real life managers think it is much easier to upgrade
: the C programmer to a C++ programmer than an Ada one, and goes on 
: shooting himself in the foot.

Does this mean the official name for Ada-9X should be C-Ada ??? ;^}
-- 
---------------------------------------------------------------------------
James A. Krzyzanowski - Senior Software Engineer - AFATDS
Magnavox Electronic Systems Company * Fort Wayne, IN 46808 * (219) 429-6446
E-Mail : jakrzy@most.magec.com

Re: Why don't large companies use Ada?

[Weiqi Gao]

The only purpose of this post is to redirecte follow-ups to this thread 
to comp.lang.ada.

\/\/eiqi Gao
weiqigao@crl.com

Re: Why don't large companies use Ada?

[Jules]

In article <3ae36c$6l2@crl9.crl.com>,
	weiqigao@crl.com (Weiqi Gao) writes:
>The only purpose of this post is to redirecte follow-ups to this thread 
>to comp.lang.ada.
>
You're not going to get very far - I bet I'm the only person who's followed
this part of the thread up. Everyone else is going to have followed up
from articles that actually have something in them. Nice try Weiqi, but
I bet you aren't even interested in Ada and you're not reading this.
Ha ha ha I've had the last laugh because everyone else is still following up
to all the groups. Ha hha ! Why not just put the word 'Ada' in your kill file!

-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Jules]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=US-ASCII, Size: 4640 bytes --]

In article <Cz9H4I.5HK@ois.com>,
	beckwb@ois.com (R. William Beckwith) writes:
>Jamie Jamison (niteowl@u.washington.edu) wrote:
>: 	I am taking an introductory computer science class at the UW and 
>: we are learning the Ada language. So far it's pretty nice.  The language
>: seems to lend itself to easy readability, the fact that the language is 
>: case insensitive is also a nice factor and the language is incredibly 
>: powerful. We're in the second quarter of this class and we're already 
>: writing ADT's and generic functions. Ada seems to be pretty neat, strong 
>: typing, bounds checking and generic functions are all nice language 
>: features, and from what I've seen Ada is a lot better than C.
>
>Yes, Ada is a wonderful language.
>
>: So why is 
>: it that the real world, such�as it is, programs in C. Sure, C compilers 
>: are cheaper, but it seems to me that the labor that you put into writing 
>: and maintaining the programs is the real cost, and I know that, at least 
>: for me, writing something in C takes a lot longer than writing something 
>: in Ada. So why hasn't Ada caught on? Why aren't people developing 
>: applications for it?

[deletions]

>IMHO, the structural factors include (in order):
>
>    1.	operating systems and windowing software were written in C
>
>	 MS-Windows and X/Xt/Motif are written in C.  Thus, it was
>	 easier to write software in C the environment.  In addition,
>	 Ada 83's built in multi-threading wreaked havoc on the
>	 non-reentrant O/S and windowing software.

I have not personally programmed Ada, although I know a little about it.
I would however strongly disagree that it is easier to write MS-Windows
software (I don't know about X as I haven't started programming it yet, only
using it so far) with C. I personally use Turbo Pascal (a language which
I believe has many features in common with Ada) to write Windows programs,
and find it much easier than using C for the following reasons:

  1: Pascal's strong typing is very useful when dealing with the multitude
     of structures and other data-types associated with a windows program.
  2: Pascal's handling of pointers is much easier to understand and more
     natural to use than C's. This is the main place where Pascal wins out
     over C, by exchanging the flexibility of C's pointer arithmetic for
     ease of use.
  3: Turbo Pascal's object oriented extensions are much easier to use than
     C++ - due to the lack of flexibility there is much less to think about
     when creating objects - no access rights to worry about, no multiple
     inheritance to confuse the issue, and so on. It really is much easier.
     This makes using object libraries such as Borland's ObjectWindows much
     easier to use than equivalent libraries in C++, where the extra
     flexibility of C++'s classes is not needed.

>	 The O/S's and windowing systems are finally catching up
>	 to Ada's multi-threading capabilities. In addition,
>	 Ada 9X's improved non-Ada interfacing features and the
>	 advent of Fresco for Ada 9X Win NT and X11R6 windowing
>	 software and CORBA for Ada 9X should remove these problems.
>
>    2.	high price of Ada compilers compared to C/C++/BASIC products
>
>	 I agree with your assessment of the real cost of software,
>	 but most companies must realize their return on investment
>	 in the first year or two to get project funding.  Thus,
>	 they won't pay four times the price of a C++ compiler for
>	 an Ada compiler.
>	 
>	 Fortunately, Ada compiler vendors are pricing their new
>	 Ada 9X products competitively with C++ products.  In
>	 addition, the free GNAT compiler allows one to learn the
>	 language without any up front cost.
>
>    3.	lack of flexibility in the first version of Ada (Ada 83)
>
>	 The lack of program pointers, polymorphism, and inheritance
>	 precluded the introduction of extensible tools in Ada 83.
>	 Ada 83 was designed for ultimate reliability and
>	 maintainability only.
>	 
>	 Ada 9X adds incredible flexibility.  Thus, Ada 9X addresses
>	 these requirements and much, much more. 
>
>None of these reasons exist anymore.  Let's see how large
>corporations react in the next couple of years.

May I add that I am eagerly anticipating the coming of Ada into my life.
It must be better than the Pascal they're making us poor first years use,
I thought this language was supposed to be easy to learn. WITHOUT ANY STRING
OPERATIONS BUILT IN? Why?

-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Alistair James Robert Young]

In article <3alh02$rgo@holly.csv.warwick.ac.uk>,
Jules <csusb@csv.warwick.ac.uk> wrote:
>I have not personally programmed Ada, although I know a little about it.
>I would however strongly disagree that it is easier to write MS-Windows
>software (I don't know about X as I haven't started programming it yet, only
>using it so far) with C. I personally use Turbo Pascal (a language which
>I believe has many features in common with Ada) to write Windows programs,
>and find it much easier than using C for the following reasons:
>
>May I add that I am eagerly anticipating the coming of Ada into my life.
>It must be better than the Pascal they're making us poor first years use,
>I thought this language was supposed to be easy to learn. WITHOUT ANY STRING
>OPERATIONS BUILT IN? Why?
>

Hang on... Turbo Pascal does have string operations built in. Unless you're
using another one for your course, of course.

However, Turbo Pascal's strings have one major disadvantage. They are structurd
as [length][string], rather than null-terminated strings, which you need to
pass to Windows functions (as well as OS/2, unix, and just about everything
else). Having to use other functions to convert back and forth is, IMO, a
major pain in the unspecified body part, as well as imposing an overhead on
calling any OS function requiring or returning a string. That's one of the
reasons I moved to C from Pascal in the first place.

PS, if you think your Pascal course is bad, pity them at St. Andrews:
s-algol, developed there, only used there, completely useless for any practical
purpose!

------------------------------------------------------------------------
Alistair Young -- Arkane Systems Software Development & PC Consultancy
The opinions above are my company's, because I OWN it!
[Development for OS/2 only!] Contact: ajry@st-and.ac.uk
"Time flies like an arrow, fruit flies like a banana" - Groucho Marx
GAT/CS/C/S d++ H s++:- g+ !p+ !au a- w+++ v+++ C++++ O++++$ P+ E+++
N+++ W--- M-- !V -po+(++) Y+ t+ R++ tv b+++ D++ e+ u** h--- f+ r++ n+ y?
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Jules]

In article <3alk8j$ok7@calvin.st-and.ac.uk>,
	ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:

>Hang on... Turbo Pascal does have string operations built in. Unless you're
>using another one for your course, of course.

Yes - I forgot to mention that, didn't I. I love Turbo Pascal. I hate Pascal
in general.

>However,Turbo Pascal's strings have one major disadvantage.They are structurd
>as [length][string], rather than null-terminated strings, which you need to
>pass to Windows functions (as well as OS/2, unix, and just about everything
>else). Having to use other functions to convert back and forth is, IMO, a
>major pain in the unspecified body part, as well as imposing an overhead on
>calling any OS function requiring or returning a string. That's one of the
>reasons I moved to C from Pascal in the first place.

Using Turbo Pascal 7, there is a unit which allows you to use null terminated
strings all the time if you so desire - but the problem is you can't use
any of Pascal's nice operators on them, you must use C style function calls
to manipulate them. But hey - you lose nothing from working in C.

>PS, if you think your Pascal course is bad, pity them at St. Andrews:
>s-algol,developed there,only used there, completely useless for any practical
>purpose!

Is (Standard) Pascal useful for any practical purpose?
A language where every time you wish to read a string you must write your own
procedure to do it? And all strings are fixed length?
A language which does not allow random access to files?
I could go on....



-- 
/* Julian R Hall				csusb@csv.warwick.ac.uk
   
   Flames should be redirected to /dev/null - I don't know what
   I'm saying myself so don't expect it to make sense all the time!         */

Re: Why don't large companies use Ada?

[Alistair James Robert Young]

In article <3avnpi$1rf@sage.csv.warwick.ac.uk>,
Jules <csusb@csv.warwick.ac.uk> wrote:
>In article <3alk8j$ok7@calvin.st-and.ac.uk>,
>	ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
>
>Using Turbo Pascal 7, there is a unit which allows you to use null terminated
>strings all the time if you so desire - but the problem is you can't use
>any of Pascal's nice operators on them, you must use C style function calls
>to manipulate them. But hey - you lose nothing from working in C.
>
Nice to know. The last version I used was Turbo Pascal for Windows 1.5, and
using null-terminated strings in that was a major pain.

>>PS, if you think your Pascal course is bad, pity them at St. Andrews:
>>s-algol,developed there,only used there, completely useless for any practical
>>purpose!
>
>Is (Standard) Pascal useful for any practical purpose?
>A language where every time you wish to read a string you must write your own
>procedure to do it? And all strings are fixed length?
>A language which does not allow random access to files?
>I could go on....
>

Well, at least you pick up some things that might be useful later on with
Turbo Pascal or C. Does anyone still use algol out there? For anything? And
when this dialect tries to implement arrays (called 'vectors' for some
bizarre reason), or pointers - well, does the word 'bastardized' mean
anything? :-)

Alistair

------------------------------------------------------------------------
Alistair Young -- Arkane Systems Software Development & PC Consultancy
The opinions above are my company's, because I OWN it!
[Development for OS/2 only!] Contact: ajry@st-and.ac.uk
"Time flies like an arrow, fruit flies like a banana" - Groucho Marx
GAT/CS/C/S d++ H s++:- g+ !p+ !au a- w+++ v+++ C++++ O++++$ P+ E+++
N+++ W--- M-- !V -po+(++) Y+ t+ R++ tv b+++ D++ e+ u** h--- f+ r++ n+ y?
------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Brian Duff]

In article <3b2bfn$487@calvin.st-and.ac.uk>,
Alistair James Robert Young <ajry@st-andrews.ac.uk> wrote:
>In article <3avnpi$1rf@sage.csv.warwick.ac.uk>,
>Jules <csusb@csv.warwick.ac.uk> wrote:
>>In article <3alk8j$ok7@calvin.st-and.ac.uk>,
>>	ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
>>
>>Using Turbo Pascal 7, there is a unit which allows you to use null terminated

<SNIP>

>>Is (Standard) Pascal useful for any practical purpose?
>>A language where every time you wish to read a string you must write your own
>>procedure to do it? And all strings are fixed length?
>>A language which does not allow random access to files?
>>I could go on....
>>
>
>Well, at least you pick up some things that might be useful later on with
>Turbo Pascal or C. Does anyone still use algol out there? For anything? And
>when this dialect tries to implement arrays (called 'vectors' for some
>bizarre reason), or pointers - well, does the word 'bastardized' mean
>anything? :-)
>

Hi Alistair. Just though I`d complain a little about S-Algol. It's
possibly one of the worst languages I've seen. I suppose the idea is to
teach us to program in a structured way etc, etc, spraff, spraff, but other
universities manage to do this quite effectively using C.
That has the added advantage that C is used in corporate circles (certainly
more than any kind of Algol, particularly S-Algol).

Oh well. Mustn't grumble, guess we're stuck with it....


Brian

>
>------------------------------------------------------------------------
>Alistair Young -- Arkane Systems Software Development & PC Consultancy
>The opinions above are my company's, because I OWN it!
>[Development for OS/2 only!] Contact: ajry@st-and.ac.uk
>"Time flies like an arrow, fruit flies like a banana" - Groucho Marx
>GAT/CS/C/S d++ H s++:- g+ !p+ !au a- w+++ v+++ C++++ O++++$ P+ E+++
>N+++ W--- M-- !V -po+(++) Y+ t+ R++ tv b+++ D++ e+ u** h--- f+ r++ n+ y?
>------------------------------------------------------------------------
>


------------------------------------------
|> |\           Brian Duff           |> |\
|> |/     University St. Andrews     |> |/  Sensible SIG v1.0..
------------------------------------------

Why don't large companies use Paradox?

[Nathan Hand]

After all, it has objects.

<bwahahaha>

Re: Why don't large companies use Paradox?

[Tres Seaver]

In article <3b4775$e1c@manuel.anu.edu.au> Nathan Hand <h9304891@student.anu.edu.au> writes:
>From: Nathan Hand <h9304891@student.anu.edu.au>
>Subject: Why don't large companies use Paradox?
>Date: 25 Nov 1994 08:28:21 GMT

>After all, it has objects.

><bwahahaha>

Lots of large companies _do_ use it -- I have several Fortune 500 clients here 
in Houston for whom I do Paradox consulting.  It all depends on what you need 
a language/tool for.

Tres Seaver,
MACRO Enterprises, Inc.,
Houston, Texas.

Re: Why don't large companies use Paradox?

[Jesse C. Wang]

In article <3b4775$e1c@manuel.anu.edu.au> Nathan Hand <h9304891@student.anu.edu.au> writes:
>From: Nathan Hand <h9304891@student.anu.edu.au>
>Subject: Why don't large companies use Paradox?
>Date: 25 Nov 1994 08:28:21 GMT

>After all, it has objects.

><bwahahaha>

So, can you describe why should large companies use Paradox ? No flame ! Just 
want to ask the reason why ?

		Thanks for your time !!

-Jesse

Re: Why don't large companies use Ada?

[Andrew Dunstan]

In article <3b2bfn$487@calvin.st-and.ac.uk>, ajry@st-andrews.ac.uk 
(Alistair James Robert Young) writes:
|> In article <3avnpi$1rf@sage.csv.warwick.ac.uk>,
|> Jules <csusb@csv.warwick.ac.uk> wrote:
|> >In article <3alk8j$ok7@calvin.st-and.ac.uk>,
|> >	ajry@st-andrews.ac.uk (Alistair James Robert Young) writes:
|> >>PS, if you think your Pascal course is bad, pity them at St. Andrews:
|> >>s-algol,developed there,only used there, completely useless for any practical
|> >>purpose!
|> >
|> >Is (Standard) Pascal useful for any practical purpose?
|> >A language where every time you wish to read a string you must write your own
|> >procedure to do it? And all strings are fixed length?
|> >A language which does not allow random access to files?
|> >I could go on....
|> >
|> 
|> Well, at least you pick up some things that might be useful later on with
|> Turbo Pascal or C. Does anyone still use algol out there? For anything? And
|> when this dialect tries to implement arrays (called 'vectors' for some
|> bizarre reason), or pointers - well, does the word 'bastardized' mean
|> anything? :-)
|> 

I'm probably one of the few people who read these groups who has had the
misfortune to have to do with s-algol. I even have a few manuals stashed away
somewhere, and still have a copy lurking on my mac.

I don't like the language, for a number of reasons which I won't bother to
go into here.

But your attack is wide of the mark. Languages should be used in teaching for
their pedagogic values, not for their application in industrial use. I happen
to think that Ada is well-suited to both purposes, but let us at least make
the distinction clear. If students can be taught good programming practise
with one tool, then picking up another language should be a snap. I don't even
remember how many computer languages I have written in - I lost count somewhere
in the 'teens.

cheers

andrew

Re: Teaching (was: Why don't large companies use Ada?)

[Michael Feldman]

In article <3b9s34$m58@cleese.apana.org.au>,
Andrew Dunstan <andrew@bugalugs.apana.org.au> wrote:

>But your attack is wide of the mark. Languages should be used in teaching for
>their pedagogic values, not for their application in industrial use. I happen
>to think that Ada is well-suited to both purposes, but let us at least make
>the distinction clear. If students can be taught good programming practise
>with one tool, then picking up another language should be a snap. I don't even
>remember how many computer languages I have written in - I lost count somewhere
>in the 'teens.

Bingo.

We are tending more and more to blur the distinction Andrew points to,
and this is a real problem. I happen to agree that Ada is a good choice
for both purposes, and I think most teachers who have given Ada an honest
try in recent years have tended to agree. Many of the students do too.

In any event, majors in computing disciplines must be taught to keep
their minds open and ready for change, because whatever they are learning
now (especially in coding-level languages) is guaranteed to be pretty
out-of-date in ten or twenty years when our freshmen will be at the
height of their careers in industry. Any computing curriculum that
lets majors get away without strength in at least two languages and
some exposure to a few more, is cheating its students. 

It is very difficult to compare an apple; students need a frame of 
reference. They get this only by seeing several apples, or even some apples 
and some oranges. 

What I find most unsettling about the movement to teach C (or C++) in
the first year is _precisely_ that C/C++ is "needed in later years".
Wait a minute: if this is so, where will the students get the other
apples and oranges? Two weeks of a language in a comparative-languages
course is _not_ enough.

Most of us who are doing Ada as an _intro_ language are interested in 
starting our students off with what we think is the "right" mindset,
namely the development of robust and portable software.

We justify Ada as an optimal choice for this purpose, with the explicit
intention that students will learn other languages as they advance.
We hear anecdotal reports that students who start with Ada make
very good C and C++ programmers later on; there are a few stories
"out there" suggesting that industry recruiters are starting to agree.

Even if Ada were more industrially popular, I know of nobody that
would push it as the _only_ language their majors should learn well.
(And I know most of the Ada-in-first-year teachers...)

The strange and baffling part of the C-in-first-year movement
is its justification _not_ as the best first-year language but as the
language the students will need later. In fact, I've met a lot of
teachers who _explicitly_ say "well, C (or C++) is _not_ the best
intro language but the students need it later, so we'll do it as a
first language." In seeking a single-language model for their
curricula, these teachers are IMHO doing serious damage to their
students' ability to cope in industry in later years when C and C++ 
are old hat.

I'd be much more comfortable with the C-family-in-first-year movement
if a number of teachers rushed forward to tell me how their students
were learning several languages well, starting from the strong first-year
pedagogical basis of C and C++. I have not yet observed such a rush.
Perhaps I will at some point, but I'm not holding my breath.

Cheers -

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Illegitimi non carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Teaching (was: Why don't large companies use Ada?)

[Jack Beidler]

In article <3bfn7t$75j@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:
|> In article <3b9s34$m58@cleese.apana.org.au>,
|> Andrew Dunstan <andrew@bugalugs.apana.org.au> wrote:
|> 
|> >But your attack is wide of the mark. Languages should be used in teaching for
|> >their pedagogic values, not for their application in industrial use. I happen
|> >to think that Ada is well-suited to both purposes, but let us at least make
|> >the distinction clear. If students can be taught good programming practise
|> >with one tool, then picking up another language should be a snap. I don't even
|> >remember how many computer languages I have written in - I lost count somewhere
|> >in the 'teens.
|> 
|> Bingo.
|> 
|> We are tending more and more to blur the distinction Andrew points to,
|> and this is a real problem. I happen to agree that Ada is a good choice
|> for both purposes, and I think most teachers who have given Ada an honest
|> try in recent years have tended to agree. Many of the students do too.
|> 
   Let me relate an incident that occurred last summer --  Through our
local Chamber of Commerce, I met with high level executives (CEOs,
CI<information>Os, etc.) of 38 firms in our area that have 
significant software development staffs.  This includes several firms
whose local staffs develop software that is used nationally and
internationally.
   As the meeting was coming to an end I explained to them why we
use Ada as the foundation language for our CS, IS, and SE degree
programs (namely, we wish to emphasize concepts -- not only does
Ada not get in the way, but it actually helps with this goal).
   A number of participants insisted that we should offer courses
in C++ for their employees.  I stated that concepts are the issue.
Their typical C programmers do not appreciate the concepts, hence
need to understand concepts before they will make good use of C++.
The typical resonse from them was that they did not need the 
concepts, they just wanted to use C++ because that's were the
industry was going!!

   I am reminded of a comment made to me once (I believe it was
made by Rich Pattis).  "Just as today's good C programmers were
educated using Pascal, tomorrow's good C++ programmers will be
educated using Ada".

   I see this point made every year when students who have been 
using Ada in CS 2 and our Advanced Data Structures and Algorithms
Course take our Operating Systems Course.  The faculty member who
teaches that course requires student programs be written in C
(or C++).  The students learn the new language on their own.
His first response to student programs was that the student
programs don't look like typical C programs -- they are modularize,
you can read them, they use meaning variable names, etc.

   As for learning C or C++ in the first course, I am reminded of
an anti drinking and driving billboard "Drinking and driving is stoopid"
(spelled that way).  

               TEACHING C OR C++ IN THE FIRST COURSE IS STOOPID

     Teacher who teach such a course deserve the grade of C or C++
as an educator.  They certainly don't deserve tenure.
+----------------------------------------+------------------------+
|John (Jack) Beidler                    ++   beidler@cs.uofs.edu  |
|  Professor, Computing Sciences Dept. ++                         |
|  University of Scranton             ++  (717) 941-7446 (voice)  |
| Scranton, PA 18510                 ++     (717) 941-4250 (FAX)  |
+------------------------------------+----------------------------+

Re: Teaching (was: Why don't large companies use Ada?)

[Andrew Koenig]

In article <3c21e1$nj@gopher.cs.uofs.edu> beidler@guinness.cs.uofs.edu (Jack Beidler) writes:

>    I see this point made every year when students who have been 
> using Ada in CS 2 and our Advanced Data Structures and Algorithms
> Course take our Operating Systems Course.  The faculty member who
> teaches that course requires student programs be written in C
> (or C++).  The students learn the new language on their own.
> His first response to student programs was that the student
> programs don't look like typical C programs -- they are modularize,
> you can read them, they use meaning variable names, etc.

>    As for learning C or C++ in the first course, I am reminded of
> an anti drinking and driving billboard "Drinking and driving is stoopid"
> (spelled that way).  

>                TEACHING C OR C++ IN THE FIRST COURSE IS STOOPID

Hmm.  Sounds like the experience doesn't support the conclusion.

I do agree with one thing: it is not nearly as important to
teach a langauge as it is to teach programming.  If you understand
what you are doing, acquiring the language skills is easy; if
not, it is impossible.

Once upon a time I taught a data structures course in Fortran.
Pointers at all.  I did this in part to prove that the concepts
I was trying to teach were language-independent.

I also told students they could use any language they liked
for their homework; if they could learn to write it, I could
learn to read it.  Most of them wound up using Fortran.  Sigh.
-- 
				--Andrew Koenig
				  ark@research.att.com

Re: Teaching (was: Why don't large companies use Ada?)

[mat]

In article <3c21e1$nj@gopher.cs.uofs.edu>, beidler@guinness.cs.uofs.edu (Jack Beidler) writes:
> In article <3bfn7t$75j@felix.seas.gwu.edu>, mfeldman@seas.gwu.edu (Michael Feldman) writes:

 ...
>    I am reminded of a comment made to me once (I believe it was
> made by Rich Pattis).  "Just as today's good C programmers were
> educated using Pascal, tomorrow's good C++ programmers will be
> educated using Ada".

Actually, some of the _worst_ C programmers I met were educated using
Pascal.  Those who got better had to unlearn a lot of the garbage that
Pascal forces on you.
 
>    I see this point made every year when students who have been 
> using Ada in CS 2 and our Advanced Data Structures and Algorithms
> Course take our Operating Systems Course.  The faculty member who
> teaches that course requires student programs be written in C
> (or C++).  The students learn the new language on their own.
> His first response to student programs was that the student
> programs don't look like typical C programs -- they are modularize,
> you can read them, they use meaning variable names, etc.

It's interesting to read, should you ever have the chance, the sources
of UNIX `v6' or of Dennis Ritchie's C compiler.  Both were highly modular
programs; both were a revelataion to me, then a young programmer in
school.

What we need are not highly pedantic programming languages but good
examples of programming.

Unfortunately, the _worst_ examples are those produced by `fundamental
algorithms' courses, for the simple reason that those tricky algorithms
are not and cannot be representative of good, clear programming.  With
the exception of linked lists, they involve algorithms and data structures
whose relationship to the information they store is neither direct nor clear
and cannot be made so by _any_ exposition.  The best that good exposition
can do is make clear the _implementation_ of these representations; the
representations themselves are esoteric and `accidental' rather than
`essential.'

This is not to say that these representations should not be used; they
are too valuable to ignore, to valuable not to teach.  But the programmer
should first be taught to construct simple, clear representations of
simple, clear problems.

>    As for learning C or C++ in the first course, I am reminded of
> an anti drinking and driving billboard "Drinking and driving is stoopid"
> (spelled that way).  
> 
>                TEACHING C OR C++ IN THE FIRST COURSE IS STOOPID
> 
>      Teacher who teach such a course deserve the grade of C or C++
> as an educator.  They certainly don't deserve tenure.

Teaching fundamental algorithms before a second course in programming,
using a `real' language, whether C, C++, or Ada is a crime against a
programmer's career.  Teaching Pascal with its crippled flow structures
compounded by its crippled conditional expression semantics ought to
be punished by banishment to a COBOL shop.
-- 
 (This man's opinions are his own.)
 From mole-end				Mark Terribile
 mat@mole-end.matawan.nj.us, Somewhere in Matawan, NJ
	(Training and consulting in C, C++, UNIX, etc.)

s-algol

[Alistair James Robert Young]

In article <3b9s34$m58@cleese.apana.org.au>,
Andrew Dunstan <andrew@bugalugs.apana.org.au> wrote:
>
>I'm probably one of the few people who read these groups who has had the
>misfortune to have to do with s-algol. I even have a few manuals stashed away
>somewhere, and still have a copy lurking on my mac.
>
>I don't like the language, for a number of reasons which I won't bother to
>go into here.
>
>But your attack is wide of the mark. Languages should be used in teaching for
>their pedagogic values, not for their application in industrial use. I happen
>to think that Ada is well-suited to both purposes, but let us at least make
>the distinction clear. If students can be taught good programming practise
>with one tool, then picking up another language should be a snap. I don't even
>remember how many computer languages I have written in - I lost count somewhere
>in the 'teens.
>

I was 'taught good programming practice` in s-algol after having learnt
C, and so am familiar with both. In my opinion, the non-standard
implementations of such things as arrays and pointers compared to most
languages in wide use actually make it more difficult for the student to
translate examples to another language. Also IMO, languages used in
teaching should be chosen for their value for that purpose, but the
languages used in industry should also be taken into account. Choosing
a language that is widely different from any in common use makes it
more difficult to adjust on entering the industrial environment.

BTW: Can anyone reading this suggest a more appropriate group?

Alistair

------------------------------------------------------------------------
Alistair Young -- Arkane Systems Software Development & PC Consultancy
The opinions above are my company's, because I OWN it!
[Development for OS/2 only!] Contact: ajry@st-and.ac.uk
"Time flies like an arrow, fruit flies like a banana" - Groucho Marx
GAT/CS/C/S d++ H s++:- g+ !p+ !au a- w+++ v+++ C++++ O++++$ P+ E+++
N+++ W--- M-- !V -po+(++) Y+ t+ R++ tv b+++ D++ e+ u** h--- f+ r++ n+ y?
------------------------------------------------------------------------

Re: s-algol

[Robert Dewar]

anyone who has trouble adjusting from one language to another and finds that
an impediment to entering the commercial marketplace shouldn't (i.e 
should not enter that market place), since they are clearly unqualified.

Re: Why don't large companies use Paradox?

[wesley]

In article <cwang.118.0037EF1E@mailbox.syr.edu>, cwang@mailbox.syr.edu (Jesse C. Wang) says:
>
>In article <3b4775$e1c@manuel.anu.edu.au> Nathan Hand <h9304891@student.anu.edu.au> writes:
>>From: Nathan Hand <h9304891@student.anu.edu.au>
>>Subject: Why don't large companies use Paradox?
>>Date: 25 Nov 1994 08:28:21 GMT
>
>>After all, it has objects.
>
>><bwahahaha>
>
>So, can you describe why should large companies use Paradox ? No flame ! Just 
>want to ask the reason why ?
>
>                Thanks for your time !!
>
>-Jesse

Ada/Pascal (was Re: Why don't large companies use Ada?)

[Davidson Corry]

In article <3alh02$rgo@holly.csv.warwick.ac.uk>, csusb@csv.warwick.ac.uk (Jules) says:
>
>May I add that I am eagerly anticipating the coming of Ada into my life.
>It must be better than the Pascal they're making us poor first years use,
>I thought this language was supposed to be easy to learn. WITHOUT ANY STRING
>OPERATIONS BUILT IN? Why?

Because Niklaus Wirth designed Pascal as a tool for teaching the fundamentals
of programming, and he did not consider "string" a fundamental data type.
A string is something you build up, _using_ the char data type and the concept
of array.

Of course, the omission of strings (and other things like random file I/O) made
it difficult to use Pascal for "real world" programming... for which purpose
Wirth had not intended it anyway.

For what it's worth (no pun intended), C and C++ don't have a fundamental
string type, either.  What they have is two bits of syntactic convenience --

    1) the name of an array is equivalent to a pointer to
        its initial element

    2) "a string in quotes" is compiled as an array of char,
        with a null added to the end of the array

-- that make it easy to use [null-terminated array of char] as if it _were_
a built-in "string type".  A raft of functions that accreted over the years
and were eventually standardized into the run-time library make it even
easier to use it that way... and harder to use any other implementation 
of "string" in a C program.  And of course the String class that's part of
the standard C++ library makes it effectively part of the language, even
though in the strictest sense it's not.

Re: Ada/Pascal (was Re: Why don't large companies use Ada?)

[Stefan Tilkov]

>In article <3atp2k$job@eve.atm.com> davidsco@atm.com (Davidson Corry) writes:

>   From: davidsco@atm.com (Davidson Corry)
>   Newsgroups: comp.lang.ada,comp.lang.c++
>   Date: 22 Nov 1994 21:50:12 GMT
>   Organization: Attachmate Corporation

	[snip]

>   Of course, the omission of strings (and other things like random file I/O) made
>   it difficult to use Pascal for "real world" programming... for which purpose
>   Wirth had not intended it anyway.

>   For what it's worth (no pun intended), C and C++ don't have a fundamental

It might interest you that the correct (German) pronunciation of "Wirth" sounds
more like "weird" than "worth".

No pun intended, of course ;-)

	[snip]


Stefan Tilkov
MLC Ratingen
-- 
__________________________________________________________________
Stefan Tilkov                                MLC Ratingen, Germany
                                                st@mlc-ratingen.de
                                        Phone +49 (0) 2102 8506 20
                                        Fax   +49 (0) 2102 8506 30

Re: Ada/Pascal (was Re: Why don't large companies use Ada?)

[Michael Feldman]

In article <ST.94Nov23204352@hobbes.mlc-ratingen.de>,
Stefan Tilkov <st@hobbes.mlc-ratingen.de> wrote:

>>   Wirth had not intended it anyway.
>
>>   For what it's worth (no pun intended), C and C++ don't have a fundamental
>
>It might interest you that the correct (German) pronunciation of "Wirth" sounds
>more like "weird" than "worth".

Actually, Niklaus Wirth has always given English-speakers permission to
call him by name ("veert") or call him by value ("worth").  :-)

(I'm not sure how to call him by reference...)

Mike Feldman
------------------------------------------------------------------------
Michael B. Feldman -  chair, SIGAda Education Working Group
Professor, Dept. of Electrical Engineering and Computer Science
The George Washington University -  Washington, DC 20052 USA
202-994-5919 (voice) - 202-994-0227 (fax) - mfeldman@seas.gwu.edu (Internet)
------------------------------------------------------------------------
         Ada on the World-Wide Web: http://lglwww.epfl.ch/Ada/
------------------------------------------------------------------------
"Non illegitimi carborundum." (Don't let the bastards grind you down.)
------------------------------------------------------------------------

Re: Ada/Pascal (was Re: Why don't large companies use Ada?)

[Keith Thompson]

In <3b0ims$f36@felix.seas.gwu.edu> mfeldman@seas.gwu.edu (Michael Feldman) writes:
> Actually, Niklaus Wirth has always given English-speakers permission to
> call him by name ("veert") or call him by value ("worth").  :-)
> 
> (I'm not sure how to call him by reference...)

"The guy who wrote _Algorithms + Data Structures = Programs_ -- his name
is on the cover."

-- 
Keith Thompson (The_Other_Keith)  kst@alsys.com
TeleSoft^H^H^H^H^H^H^H^H Alsys, Inc.
10251 Vista Sorrento Parkway, Suite 300, San Diego, CA, USA, 92121-2718
/user/kst/.signature: I/O error (core dumped)

Re: Ada/Pascal (was Re: Why don't large companies use Ada?)

[John English]

Stefan Tilkov (st@hobbes.mlc-ratingen.de) wrote:
: It might interest you that the correct (German) pronunciation of "Wirth"
: sounds more like "weird" than "worth".
: No pun intended, of course ;-)

Aha, "call by name" vs. "call by value" (= "worth")     ;-)

-- 
-------------------------------------------------------------------------------
John English                    | Thoughts for the day:
Dept. of Computing              | - People who live in windowed environments
University of Brighton          |   shouldn't cast pointers
E-mail: je@brighton.ac.uk       | - In C++ only your friends can access your
Fax:    0273 642405             |   private parts
-------------------------------------------------------------------------------

Re: Why don't large companies use Ada?

[Dan Thies]

Jamie Jamison (niteowl@u.washington.edu) wrote:
<a bunch of stuff about how great Ada is>

<which I agree with>

This article has been cross-posted to just about every comp.lang 
newsgroup in existence.

Whoever decided to do this, please get over to news.answers and get some 
of the basic usenet and netiquette FAQs.  Nobody is going to be swayed if 
posts from Team Ada are appearing in comp.databases.oracle.  At the most, 
I would say comp.object and comp.programming have some remote possibility 
of having some interest in this.

Dan
rtfm@cyberspace.com
-- 
rtfm@cyberspace.com

Re: Why don't large companies use Ada?

[Peter Coffee AC6EN]

I think the biggest remaining criticism of C++ is the high incidence
of programs that successfully compile but don't do what was intended.
Ada still makes it a good deal harder to make an error that still results
in compilable code, except of course for coding the wrong algorithm or
something that's equally likely to be recognized by an expert in the
problem domain -- except that Ada's higher readability also improves
the likelihood of catching that class of error as well.

Re: Why don't large companies use Ada?

[Jeff Gray]

In article <D03AL2.5xw@aston.ac.uk>, Gareth Bromley writes:


>Why should they bother??
>
>Given that most students coming out of college/University will have some
>experience in programming some type of language Pascal, Modula-2/3, C
>and possibly BASIC plus any others and also most colleges/Universities
>provide courses in C, it is totally logical, and more important cost
>saving to send/train employees on C++ courses, and thus program systems
>in C++.

I have not seen a recent survey of language use (in terms of CS-1 and CS-2) in
U.S. universities. The last one that I am aware of, however, placed Ada second 
behind Pascal. Many students are coming out with Ada experience. Why not
utilize that experience?

>Why?
>
>i) Cost. If the user has a basic knowledge of C or other programming language
>then converting to C++ is relatively easy.

Relatively easy? Quickly converting C programmers to C++ may give them an
understanding of the syntactic differences between the two languages but
they will be sorely unprepared for understanding the proper way of applying
the object-oriented paradigm. The result will resemble the AdaTRAN programmers
of the early '80s.

>ii) Ada's complexity. Most new users will quickly come across all Ada's *nice*
>type checking, or special constructs which require a different syntax to normal.
>In other languages these are not required.

Nor are they always available. I have often found the type checking in Ada to
be a blessing rather than a hindrance.

>iii) Support. Virtually every computer system in the world has a C++ compiler,
>and it is being updated continually. Because that is where all the money is.
>Not every computer system in the world has Ada on it, plus most will not be
>powerful enough to run it.                                 ^^^^^^^^^^^^^^^^^ 
^^^^^^^^^^^^^^^^^^^^^^^^^^

Aside from a Commodore 64 or Vic-20 (or even Timex Sinclair), could you give
me an example of some present day machines that can not support an Ada
compiler?


>iv) Libraries, Code already written and available experts. C++ has access straight
>away to all the C code libraries written. Not just a few nice ones, but all
>C library code ever written, one reason the famous COBOL is still around!
>Lots and lots of code has already been written in C++, and also any company can
>go out and get any number of C++ programmers, or even C ones and train them,
>whereas there are fewer Ada programmers.

Lots and Lots of code is also available from some of the repositories like
CARDS, ASSET, AdaNet (is it still called this anymore) and other suppliers.

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <1994Dec1.043246.13829@news.vanderbilt.edu>,
Jeff Gray <jgray@vuse.vanderbilt.edu> wrote:

[snip]

>I have not seen a recent survey of language use (in terms of CS-1 and CS-2) in
>U.S. universities. The last one that I am aware of, however, placed Ada second 
>behind Pascal. Many students are coming out with Ada experience. Why not
>utilize that experience?

The latest Reid Report on CS-1 languages shows Ada in second place,
'way behind Pascal, which is still the champ by far. The Reid Report
data may be a bit misleading, though, because I do not know how
representative it is. Reid's survey consists of listings from schools
who voluntarily report by e-mail. He lists some 400 or so, but there
are several thousand worldwide. 

I can tell you that his Ada data is reaonably complete. There are a few
places he missed, but not too many, as far as I can tell. I know this
because I think I've tracked down most if not all of the Ada-in-Cs1/CS2
institutions. My list is in the next message; note that it shows _only_ 
CS1/CS2 uses of Ada, NOT "Ada courses" in general. BTW - the CS1 number has
increased _fourfold_ in the last three years, the CS2 number has
doubled. These data are pretty verifiable.

In comparing Reid's data, it is very dangerous, and IMHO not intellectually
honest, to look at raw percentages, because whereas his language A
listings might be nearly exhaustive (because language A has champions
working hard to ensure that language A teachers report to him), language
C or P or M or L might be only a sample, and who knows whether that
sample can be extrapolated across all schools? So we have no idea
what the Ada percentages are. I don't care, because "market share"
is a less important criterion for success than a continuing upward
trend, which is what we've got.

Reid posts updates to comp.edu twice a year or so. I'm hesitant to
post it here because I've seen the data misinterpreted too many times.

Later -

Mike Feldman

Re: Why don't large companies use Ada?

[Bennett, Chip (KTR) ~U]

ron house <house@HELIOS.USQ.EDU.AU> wrote:

> How about "Ada - use it when you want to multiply a length by a length
> and get a length, or when you want to be prevented from dividing an area
> by a length because they have different types."

> (PS: yes, you can fix that by defining zillions of additional *,/,- and +
> functions, but good grief...)

It seems to me that how many functions you have to define depends on how you
approach the problem.  Let's start with the premise that there are many
triples of numbers that exhibit similar relationships.  For example:

     1) length, 2 (squared), area
     2) length, 3 (cubed), volume
     3) length, N, hyperspace  :-)

Or perhaps a different relationship:

     1) speed, time, distance
     2) principle, rate, interest
     3) mass, acceleration, force
     4) current, resistance, voltage
     5) mass, (light speed) squared, energy
     6) large unit of measure, constant, small unit of measure

Now, we ought to be able to create a generic package that declares a set of
allowable functions for a particular relationship (A := B * C, C := A / B,
etc.).  Then we could create a package that declares a set of private types
(i.e. speed, time, and distance) and instantiates the generic package of
math functions using the private types.  Since the types are private they
are limited to the functions in the instantiated generic.  Area := Length **
2 is allowed, but Length := Length ** 2 is not allowed.

In Ada 83 there is a limitation in that we can't extend the set of allowable
functions for some triple that has some _extra_ relationship that the others
don't have, but child packages in Ada 9X would allow for adding new
functions to the private type.

Given that there is a fairly limited set of relationship groups for triples
that make sense in the real world, I doubt you would have to write very many
additional math functions.

All that said, I've never actually tried this, but it seems plausible.  Any
thoughts?

*****************************************************************
* Chip Bennett, GDE Systems Inc | BennettC@j64.stratcom.af.mil  *
* USSTRATCOM/J64213             | Voice (402)294-7360           *
* 901 SAC Blvd, Suite 2B24      | FAX   (402)294-7912           *
* Offutt AFB, NE 68113-6600     | Proud member of Team Ada      *
* Opinions expressed here are my own _so_,  TTFWTW              *
*****************************************************************

Re: Why don't large companies use Ada?

[Bennett, Chip (KTR) ~U]

"Ole-Hjalmar Kristensen TF.DT/DELAB" <ohk@HAL.NTA.NO> wrote:

> Well, I think this illustrates some differences between C++ and Ada :-)
> (anyone knows who is the author of this amusing piece?)
>
> Shooting Yourself in the Foot
     .
     .
     .
     (funny stuff deleted)


This was hilarious.  I don't know about everybody else, but I would love to
see the rest of this article (on the other languages).

*****************************************************************
* Chip Bennett, GDE Systems Inc | BennettC@j64.stratcom.af.mil  *
* USSTRATCOM/J64213             | Voice (402)294-7360           *
* 901 SAC Blvd, Suite 2B24      | FAX   (402)294-7912           *
* Offutt AFB, NE 68113-6600     | Proud member of Team Ada      *
* Opinions expressed here are my own _so_,  TTFWTW              *
*****************************************************************

Re: Why don't large companies use Ada?

[Robert Dewar]

People are furiously cross-posting this thread, and many of the messages
are quite inappropriate outside comp.lang.ada. Please be careful,
inappropriate cross-posting quite understandably antagonizes people.

Re: Why don't large companies use Ada?

[Paige Vinall]

In INFO-ADA Digest - 3 Dec 1994 - Special issue (item 14)No.  Keith Thompson <ks
t@ALSYS.COM> writes:
>If you're writing new PL/I or Ada code to interface to existing Fortran
>libraries, all you should need to do is swap the indices; there should
>be no need to physically transpose the arrays themselves.  This should

Or you can use ISUB defining in PL/I.  It is a very cool feature which
will transpose the matrix for you.  It allows one copy of the
matrix and then you can define various views of the matrix.  For
example-

   dcl a(0:5, 0:9) float bin;
   dcl A_transpose(0:9, 0:5)   float bin def a(2sub, 1sub);

Matrix A is the real data and A_transpose is a view of the data
transposed.  You can now use A with your PL/I program but pass
A_transpose to the Fortran routine.  However, a real transposed
version of the matrix will be created at each call statement so
the performance would be bad if there were frequent calls to
Fortran.

Paige Vinall - PL/I Development Manager, IBM
Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Re: Why don't large companies use Ada?

[Paige Vinall]

In INFO-ADA Digest - 5 Dec 1994 (item 11)No.  Michael Feldman <mfeldman@SEAS.GWU
.EDU> writes:
>Yes.
>
>But unless I mis-read the post from the IBM chap the other day,
>ISUB does more than logically swap the indices, it actually
>transposes the array. I don't think I kept that post, but I do
>think that's what he said.
>

If an ISUB dfined array is passed to a procedure, PL/I
will compute an image of it.

>Has PL/1 supported ISUB all along? I do not remember it from my
>PL/1 days 20 or so years ago.

I think so.

Paige Vinall - PL/I Development Manager, IBM
Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Re: Why don't large companies use Ada?

[Norman H. Cohen]

In article <9412061309.AA02026@ajpo.sei.cmu.edu>, Paige Vinall
<vinall@VNET.IBM.COM> writes: 

|> In INFO-ADA Digest - 5 Dec 1994 (item 11)No.  Michael Feldman <mfeldman@SEAS.GWU
|> .EDU> writes: 
|> >Yes.
|> >
|> >But unless I mis-read the post from the IBM chap the other day,
|> >ISUB does more than logically swap the indices, it actually
|> >transposes the array. I don't think I kept that post, but I do
|> >think that's what he said.
|> >
|>
|> If an ISUB dfined array is passed to a procedure, PL/I
|> will compute an image of it.

The way to use ISUB when passing PL/I arrays to Fortran (or, as it was
called back then, FORTRAN) is to declare an ordinary array that is filled
in column-major (i.e., as Fortran expects it).  If, inisde the PL/I
program, you want to view the data as arranged in row-major order, then
declare an ISUB that creates a virtual transposed copy of the array.
It is the ordinary array that is passed to Fortran, so it is unnecessary
to create a transposed copy of the array.  (If the array is to be passed
from one PL/I procedure to another, the ordinary array can be passed and
the ISUB declaration repeated in the called procedure.)

--
Norman H. Cohen    ncohen@watson.ibm.com

Re: Why don't large companies use Ada?

[Michael Feldman]

In article <9412061309.AA02026@ajpo.sei.cmu.edu>,
Paige Vinall  <vinall@VNET.IBM.COM> wrote:

>If an ISUB dfined array is passed to a procedure, PL/I
>will compute an image of it.

That's what I thought.

>>Has PL/1 supported ISUB all along? I do not remember it from my
>>PL/1 days 20 or so years ago.

>I think so.

OK, fine. This is a very different animal from representing the array
in Fortran-friendly terms in the first place. Since Ada allows the
implementer to choose the representation, she can choose column-major.

Mike Feldman

Re: Why don't large companies use Ada?

[Bob Wells #402]

Michael Feldman <mfeldman@SEAS.GWU.EDU> writes..

> Don't take this personally, but the "all you should..." attitude is, in
> my experience, quite typical of computer folks who don't understand, or
> don't want to understand, what it takes to get a customer to buy.

G'day Michael,
An extension/corollary/variation to this is what I call "just jobs." You
know! When you have some, er, incredibly "competent" person who comes along
and says "Oh that's simple. You just have to ....." Then wanders off with a
look that they have solved the world.

My favourite one of these was when HP-UX was found to be too slow to run
what we wanted to do. His still echoing words were "That's OK, you just
have to rewrite the operating system!"

Yeesh!

@                      -----------------
@          //// ----- ( Flammkuchen?!?! )
@         (o o)        -----------------
@ ----oOO--(_)--OOo--------------------------------------------------------
  Bob Wells    "Hey!  Who took the cork out of my lunch??!" -- W. C. Fields
@ INTERNET: wel@eurocontrol.de                 CompuServe:      100272,3004
@ The Ada WWW Server is http://lglwww.epfl.ch/Ada/                 Team Ada
@ For exciting Ada info enter 'finger wel@s4ecawel.eurocontrol.de'

Re: Why don't large companies use Ada?

[Bob Wells #402]

G'day,

Here's a totally non-Ada question.

Way back in the dim dark past while studying at (yes, and graduating from (-: )
Uni. of NSW in Sydney I had to do some programming in PL/1. I'm sure that this
was pronounced as "pee-ell-one." I see that Robert Dewar (among others) writes
it as PL/1, yet I see that Dave Emery (along with others) writes it as PL/I.
Most of the Germans here at Eurocontrol swear "black-and-blue" that it is called
"pee-ell-eye." Yet I remember that the "eye" was supposed to be read as a Roman
numeral.

Is my memory playing tricks? What is the correct version?

BTW Sorry to take up bandwidth from that **incredibly important** "XYZ's Horribl
e
Posts" thread. Want to know about a two phase random bullshit generator? Send me
email for details! PAT. APP. FOR 12345674.

'Avagoodone!
Bob W. (-:

P.S. The PL/(I1) question is a serious one, not important one, but not a joke
question either! (-:

Re: Why don't large companies use Ada?

[Paige Vinall]

The official name is PL/I (as specified by the ANSI standards).  It
stands for "Programming Language One".  The "I" is a roman numeral.
Using an arabic one is incorrect.  Before PL/I, it was briefly called
NPL for "New Programming Language" but the acronym conflicted with Net
Propulsion Laboratory.  "PL/1" is incorrect but used so heavily that I
don't bother correcting people.

Paige Vinall - PL/I Development Manager, IBM
Internet: vinall@vnet.ibm.com,  Phone: (408) 463-2106

Re: Why don't large companies use Ada?

[Kenneth G. Hamilton]

If a compiler-maker really wanted to do it (yeah, right, with a gun to
their heads!) there's no reason why a compile-time switch couldn't
switch the order of subscripts in a Fortran array.  Of course, this
would break code that EQUIVALENCEd 2D arrays to 1D ones, but that's
a pretty  horrible thing to do, anyway.  It could even be regarded
as a different KIND by an F90 compiler.
-- 
- Kenneth G. Hamilton

Re: Why don't large companies use Ada?

[Robert Dewar]

Not just EQUIVALENCE, but much more importantly in Fortran you can pass
a two dimensional array to a routine that treats it as a one dimensional
array, and also you can pass a single column (but not a single row) of
a two dimensional array as a one dimensional array. These features are
often used.

Since we have pragma convention in Ada 9X, I don't see much need for
Fortran to muck around.

Interesting note: if you are trying to convert FOrtran code to Ada, and
it does use the column slicing trick, then you are in trouble, or more
accurately, you will have to play low level games with unchecked
conversion of access types, because Ada does not allow slicing of 
multi-dimensional arrays.

Note incidentally that the ability in Fortran to pass columns is exactly
what defines the requirement for array mapping in Fortran, whereas in
Ada there are no such semantics, so the array ordering is not specified
by the language (although it is slightly suggested by the form of array
aggregates for multi-dimensional arrays).