Data shows Top 50 Software Vendors not using Ada

Data shows Top 50 Software Vendors not using Ada

[Gregory Aharonian]

     The Mandated world has for some time ignored all data about the
dismal state of Ada use outside the Mandated world, which is a shame,
considering the large amount of data available, data useful for redirecting
DoD policies to make Ada successful.

     Case in point.  The May 31, 1993 issue of Information Week has an annual
review article on the Top 50 Independent Software Vendors.  Their collective
revenues are over $16 billion a year, they employ about 100,000 people, and
they dominate are setting standards for almost every field of computing for
the future: operating systems, applications, CAD/CAM, CASE, MIS, EIS, 
networking, communications, database management, graphics, publishing, etc.
AND NONE OF THEM USE ADA AND NONE OF THEM ARE INVOLVED WITH DOD SOFTWARE
POLICY ACTIVITIES.  Once again, when it comes to spending their OWN money,
no one is spending it on Ada.

     For how long can the DoD isolate itself from the mainstream of American
commercial software development?  Is this anyway to guarantee national
security, by ignoring the very people and industries you are defending?
I think not.  Drop the Ada Mandate and let the defense community more
cost-effectively fulfill its requirements by being able to tap into this
large, profitable, American-as-apple-pie, commercial software industry,
an industry that is doing without tax dollars what the DoD Ada contractors
are barely managing to do with tax dollars.

    Here's a list of the companies in the Top 50:

	Microsoft, Computer Associates, Oracle, Novell, Lotus, WordPerfect,
Dun & Bradstreet, SAP AG, Software AG, Borland, Ask Group, Legent, Cadence,
SAS Institute, Autodesk, Mentor Graphics, American Management Systems,
Informix, Adobe, Sybase, BMC Software, System Software, Information Builders,
Compuware, Symantec, Candle, Aldus, Santa Cruz Operation, J.D.Edwards,
Structural Dynamics, Cincom Systems, Software Publishing, Systems Center
Claris, Knowledgeware, Attachmate, Micro Focus, Interleaf, Software Toolworks,
Banyan Systems, Comshare, Cognos, Boole and Babbage, Softlab AG, American
Software, Parametric Technology, Cerner, Agency Management, Intuit and
Medical Information Technology.

No Ada, mostly C/C++, some Cobol and Assembler and Lisp.  But no Ada.
Until the DoD replaces wholescale its current managers and contractors with
people who understand and subscribe to capitalism, Ada will remain as
dead as Latin.


Greg Aharonian
Source Translation & Optimization
-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178

Re: Data shows Top 50 Software Vendors not using Ada

[David Tannen x8273]

Greg,

This time I don't think this list of companies means that much to what the DoD 
is
required to do.  Do any of these companies build systems that need to be as fau
lt
tolerant as DoD systems?  I doubt it.  If there software fails, just reboot and
go on.

On one system I worked on, if it failed - bye bye WORLD.  A serious possibility
 in 
the DoD world.  It was critical that the system worked right every time.

I use Borland products at home because they are superior for DOS/MS-Windows com
pared
to anything available in Ada for the same platform.  But I do not think I want 
my
life to be decided by a C++ program and a pointer pointing to something in memo
ry
it was not suppose to be pointing to.

David Tannen

Re: Data shows Top 50 Software Vendors not using Ada

[dog.ee.lbl.gov!overload.lbl.gov!agate!howland.reston.ans.net!europa.eng.g]

In article <C81yLF.B9p@world.std.com> srctran@world.std.com (Gregory Aharonian)
 writes:
>No Ada, mostly C/C++, some Cobol and Assembler and Lisp.  But no Ada.
>Until the DoD replaces wholescale its current managers and contractors with
>people who understand and subscribe to capitalism, Ada will remain as
>dead as Latin.
>        ^^^^^^
*Whew!*  For a minute there, I thought he was going to say "Esperanto"!
:-)

>Greg Aharonian
>Source Translation & Optimization

Yeah, we could tell right away. <big grin!>

-- 
-Comments above aren't neceessarily the opinion of the SEI, AJPO, or CAE-Link-
David Weller  |  Have you hugged your DRAGOON lately?
----I'm the Ultimate International Masochist: I speak Ada AND Esperanto!-----

Re: Data shows Top 50 Software Vendors not using Ada

[Rod Cheshire]

srctran@world.std.com (Gregory Aharonian) writes:

>     The Mandated world has for some time ignored all data about the
>dismal state of Ada use outside the Mandated world, which is a shame,
>considering the large amount of data available, data useful for redirecting
>DoD policies to make Ada successful.

>     Case in point.  The May 31, 1993 issue of Information Week has an annual
>review article on the Top 50 Independent Software Vendors.  Their collective
>revenues are over $16 billion a year, they employ about 100,000 people, and
>they dominate are setting standards for almost every field of computing for
>the future: operating systems, applications, CAD/CAM, CASE, MIS, EIS, 
>networking, communications, database management, graphics, publishing, etc.
>AND NONE OF THEM USE ADA AND NONE OF THEM ARE INVOLVED WITH DOD SOFTWARE
>POLICY ACTIVITIES.  Once again, when it comes to spending their OWN money,
>no one is spending it on Ada.

Why would a commercially oriented company spend a lot of money "tooling up"
on Ada when Languages like C and Pascal are taught in Universities?

>     For how long can the DoD isolate itself from the mainstream of American
>commercial software development?  Is this anyway to guarantee national
>security, by ignoring the very people and industries you are defending?
>I think not.  Drop the Ada Mandate and let the defense community more
>cost-effectively fulfill its requirements by being able to tap into this
>large, profitable, American-as-apple-pie, commercial software industry,
>an industry that is doing without tax dollars what the DoD Ada contractors
>are barely managing to do with tax dollars.

I agree. In the days when Software Engineering techniques, languages and
Quality Assurance
were a liitle more than a black art, the DoD and NASA desperately needed some
way to increase reliabilty, standardise and make the generation of software
more efficient. THINGS HAVE CHANGED or havent they noticed? Software
Engineering is actualy taught in universities and colleges, QA is a lot
better than it was then (thank industry leaders, sei etc) and modern 
languages/environments now alleviate the problems we had when Ada was
invented.

>    Here's a list of the companies in the Top 50:

>	Microsoft, Computer Associates, Oracle, Novell, Lotus, WordPerfect,
>Dun & Bradstreet, SAP AG, Software AG, Borland, Ask Group, Legent, Cadence,
>SAS Institute, Autodesk, Mentor Graphics, American Management Systems,
>Informix, Adobe, Sybase, BMC Software, System Software, Information Builders,
>Compuware, Symantec, Candle, Aldus, Santa Cruz Operation, J.D.Edwards,
>Structural Dynamics, Cincom Systems, Software Publishing, Systems Center
>Claris, Knowledgeware, Attachmate, Micro Focus, Interleaf, Software Toolworks,
>Banyan Systems, Comshare, Cognos, Boole and Babbage, Softlab AG, American
>Software, Parametric Technology, Cerner, Agency Management, Intuit and
>Medical Information Technology.


I reckon the Ada complier manufacturers realised that Ada, being a defence
related language, would be a gold mine (the $1000 coffee machine syndrome).
Surely by now, the cost of making Ada support environments has fallen?

>No Ada, mostly C/C++, some Cobol and Assembler and Lisp.  But no Ada.
>Until the DoD replaces wholescale its current managers and contractors with
>people who understand and subscribe to capitalism, Ada will remain as
>dead as Latin.

I personally would like to see Ada more widely used. However as an 
engineer I would like to have the flexability to use the best tool for the
job (providing I can justify using it).

Rod Cheshire
rodc@pf.adied.oz.au

Re: Data shows Top 50 Software Vendors not using Ada

[Gregory Aharonian]

>This time I don't think this list of companies means that much to what the
>DoD is required to do.  Do any of these companies build systems that need
>to be as fault tolerant as DoD systems?  I doubt it. If there software fails,
>just reboot and go on.

   You know, one of the most obnoxious conceits of the Defense world is that
it has the worst real time, fault tolerant, large scale computing problems.
There are plenty of industries using the tools and products of the 50 companies
I listed whose problems are just as bad as the DoD's.
   Consider Wall Street.  Each day these guys trade hundreds of billions of
dollars in financial products, futures, stocks, etc.  Miss one tick or
second of market movements, and you could lose millions of your OWN money.
Their systems can't break down during the trading day, their algorithms have
to handle all market exceptions, etc.  Because at stake is billions of
dollars.  Yet they seem to survive without high priced Ada systems.
   How about airlines.  A few months ago, Business Week had a glowing article
about Smalltalk,in which a large Smalltalk system developed by American
Airlines handles their daily scheduling activities.  Screw up one connection,
and the propagation throughout the system will cost the airline millions.
Over 1200 flights are handled daily. Certainly American Airlines logistics
here are as complicated as most similar DoD operations, and they do this
everyday.
   The phone systems - real time, high throughput, even help fight a few
wars with the DoD - all done in C.  Remember the invasion of Grenada?
Some guy had to use his credit card and make a phone call back to the US
over phone lines programmed with C because the DoD's Ada communications
systems couldn't talk to each other.  "Yea operator, I'm on a beach trying
to redirect the shelling - can you connect me to the military".

  The DoD has to understand that its problems are no longer unique, that
many others having just as pressing problems, and that these people are
addressing these problems with their own money without Ada, and without
spending a lot of tax dollars on over priced software tools.  And what
does the leading Ada contractor, IBM, say to its customers when they
ask how to solve such problems - "Use C++ or Smalltalk - we do".

Greg
-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178

Re: Data shows Top 50 Software Vendors not using Ada

[howland.reston.ans.net!agate!linus!linus.mitre.org!mwunix.mitre.org!m2358]

I feel that the data presented does not support the conclusion that
Ada is not being used, while it may support the conclusion that Ada
is not being used in the MAJORITY of the work.  I've worked inside of
a few companies and the government in the past, and in my experience,
such organizations seldom speak with a single voice in most matters.
There are often trends, but there are also often pockets which are doing
something different from the trend.

I feel that companies usually experience the same problems as the DoD,
but on a smaller scale.  Delivering quality software on-time, within
budget, meeting requirements, and being marketable in nature are common
objectives (even the DoD has to market within itself).  So they seek
solutions within their constraints, and their constraints often include
the time to explore alternate solutions and the budget to buy things like
Ada compilers.

This is where the universities can play a significant role.  I feel that
the AJPO is doing a really good thing by fostering the use of Ada in
universities.  Over 200 students have gone through my SE, OOD, and OOP
classes so far, and these students go out into industry with an exposure
to Ada included in their list of assets.  I also expose them to C++ in
the OOD and OOP courses, and we go through a series of exercises comparing
and contrasting the two languages.  Sometimes the students jump on the
Ada bandwagon ... sometimes they do not ... it seems to vary with their
individual experiences and needs.

In many cases, I hear back from students who jumped on the Ada bandwagon
years later.  The reports are often of the form that they explained their
perception of Ada to their bosses and won over the idea of trying Ada on
some project.  Sometimes I hear that the project was a win, sometimes not.
But at least management opened its eyes to a "new" idea (Ada) that it may
not have even considered before that student came along.

As strongly as I have come out for Ada on the net before, I'd like to note
that I frequently use Ada, C, and C++ for my personal work (at home).  Lately i
t
has been coming out about 70% Ada, 5% C, and 25% C++, but that changes with
the projects.  At MITRE, my work is about 85% Ada and 15% C (also depending on
the projects), at Monmouth College (where I currently teach) my work is
about 70% Ada and 30% C++, and in the archives my work is 100% C (there are
no Ada compilers on the DEC-20 or the DEC ALPHA, altho the ALPHA is supposed
to get one, I hear).  In all of my work, except MITRE, of course, I have a
free choice in the language.  How I choose a language for a project is another
story (to be told later, if there is interest).

Rick Conn

Re: Data shows Top 50 Software Vendors not using Ada

[David Tannen x8273]

Greg,

You missed my point entirely.  Most of your examples had to do with the loss of
money - a small loss compared to human lives or a whole planet.

DoD systems must be even more fault tolerant than most of the systems built by
the rest of the world.  Greg, please name one system you know about that if it
fails could cause the loss of life (on a large scale) or even the meltdown of
the planet.

As for your comment on DoD comms systems and Grenada.  I know people who were
there and folks who built these systems.  The reason they did not work has more
to do with how equipment is procurred by the DoD than Ada.  Each service (Army,
Navy and Air Force) is always off working in isolation or in competition for 
limited $s.  They rarely talk to each other about what they are doing (might
loss their edge w/Congress or the President for those ever shrinking $s).  So
the services often end up with systems that don't work together.

Gee, isn't that why Ada was developed in the first place, to eliminate all the
different languages being used by the DoD.

David Tannen
tannen@tigger.geg.mot.com

Re: Data shows Top 50 Software Vendors not using Ada

[agate!howland.reston.ans.net!noc.near.net!nic.umass.edu!ymir.cs.umass.edu]

In article <SRCTRAN.93Jun3232913@world.std.com> srctran@world.std.com (Gregory 
Aharonian) writes:
 > You know, one of the most obnoxious conceits of the Defense
 > world is that it has the worst real time, fault tolerant, large
 > scale computing problems.  There are plenty of industries using the
 > tools and products of the 50 companies I listed whose problems are
 > just as bad as the DoD's.

There are organizations outside the Pentagon with mission-critical
software systems out there, but the Pentagon's systems are
life-critical.  Failing to make a trade or double-booking a seat on an
airplane just isn't in the same class... To compare apples to apples
you need to talk about patient monitoring, railroad signalling, or
something...

 >     Consider Wall Street. Their systems can't break down during the
 > trading day, their algorithms have to handle all market exceptions,
 > etc.

Which of course explains why the SEC shuts down programmed trading
whenever the market acts up.  And of course there was a recent
incedent where the market management software failed and trading had
to be done with paper records.  Might not have happened if...

 > How about airlines.  A few months ago, Business Week had a glowing
 > article about Smalltalk,in which a large Smalltalk system developed
 > by American Airlines handles their daily scheduling activities.
 > Screw up one connection,

How about airlines.  The new ATC system is being written in...
Boeing is writing the flight software for its new passenger aircraft
in...

      The phone systems - real time, high throughput, even help fight a few
   wars with the DoD - all done in C.  Remember the invasion of Grenada?

Remember the Northeast long-distance blackout? It was directly
attributed to the semantics of 'break'.
        /s
--
Alexander Erskine Wise /\/\/\/\/\/\/\/\/\/\/\/\ Software Development Laboratory
/\/\/\/\/\/\/\/\/\/\/\/\/\/\ WISE@CS.UMASS.EDU /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\ This situation calls for large amounts of unadulterated CHOCOLATE! /\/\/\

Re: Data shows Top 50 Software Vendors not using Ada

[deccrl!news.crl.dec.com!dbased.nuo.dec.com!digits.enet.dec.com!brett]

In article <1993Jun4.130219.27569@linus.mitre.org>, m23588@mwunix.mitre.org (Ri
chard Conn) writes...

>about 70% Ada and 30% C++, and in the archives my work is 100% C (there are
>no Ada compilers on the DEC-20 or the DEC ALPHA, altho the ALPHA is supposed
>to get one, I hear).

>Rick Conn


The DEC Ada V3.0 for OpenVMS AXP compiler  [Hey, don't blame me for the
					    Trademark laws or product names :-)
]
is shipping today, has been for months.  Unlike the DEC Ada V3.0 for OpenVMS VA
X
compiler, it does not have the Professional Development Option available yet,
but we expect that to be in V3.1 (that is the Smart Recompilation stuff etc.)


Like all the DEC Ada compilers, it is available via our campus-wide licensing
program for media-cost [ie: free] to qualifying academic institutions.  Our
Mips Ultrix product does very well in this space.


The DEC Ada V3.0 for OSF/1 AXP compiler is currently being field tested.


/Bevin
DEC Ada team

Re: Data shows Top 50 Software Vendors not using Ada

[cis.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!agate!linus]

In article <1993Jun4.202343.26416@dbased.nuo.dec.com>, brett@digits.enet.dec.co
m (Bevin R. Brett) writes:
|> 
|> In article <1993Jun4.130219.27569@linus.mitre.org>, m23588@mwunix.mitre.org 
(Richard Conn) writes...
|> 
|> >about 70% Ada and 30% C++, and in the archives my work is 100% C (there are
|> >no Ada compilers on the DEC-20 or the DEC ALPHA, altho the ALPHA is suppose
d
|> >to get one, I hear).
|> 
|> >Rick Conn
|> 
|> 
|> The DEC Ada V3.0 for OpenVMS AXP compiler  [Hey, don't blame me for the
|> 					    Trademark laws or product names :-)
]
|> is shipping today, has been for months.  Unlike the DEC Ada V3.0 for OpenVMS
 VAX
|> compiler, it does not have the Professional Development Option available yet
,
|> but we expect that to be in V3.1 (that is the Smart Recompilation stuff etc.
)
|> 
|> 
|> Like all the DEC Ada compilers, it is available via our campus-wide licensin
g
|> program for media-cost [ie: free] to qualifying academic institutions.  Our
|> Mips Ultrix product does very well in this space.
|> 
|> 
|> The DEC Ada V3.0 for OSF/1 AXP compiler is currently being field tested.
|> 
|> 
|> /Bevin
|> DEC Ada team
Great news, Bevin.  Thank you very much ... I'll look forward to it with
relish.  WUARCHIVE is the base I'm working from for this, and it is running
OSF/1.  I think WUSTL has the appropriate agreement in place.

Rick

Re: Data shows Top 50 Software Vendors not using Ada

[John Bollenbacher]

Would we possibly save large amounts of bandwidth if we all agreed with
Greg?  Greg, we all agree that if the mandate we lifted and the DoD were
allowed to choose the cheapest language with which to create a system one
of 2 things would happen:
  1) Ada would not successfully compete in the open market and would
  disappear, or
  2) Ada would successfully compete and its prices would drop

It is also quite obvious that in either case, the DoD would be in the
position it was in when it decided it needed a mandate.  Now this is the
part that needs to be read slowly by some:

The DoD is responsible for maintaining huge amounts of embedded software
for very long periods of time.  If each procurer is free to choose the
implementation language that make his program's creation cheapest, there
will be an overwhelming number of language/dialects to be maintained by
this department (read tax-dollars).  The point of the mandate is the ease 
this management problem.

IT IS NOT THE GOVERNMENTS JOB TO MAKE ADA SUCCESSFUL IN THE OPEN MARKET.

Sorry, for the caps (and the general, frustrated, tone of this posting).

--
-----------------------------------------------------------------------------
- John Bollenbacher                                        jhb@dale.cts.com -
- Titan Linkabit Corp.                                       (619) 552-9963 -
- 3033 Science Park Rd.                                                     -
- San Diego, Ca. 92121                                                      -
-----------------------------------------------------------------------------

Re: Data shows Top 50 Software Vendors not using Ada

[Gregory Aharonian]

>You missed my point entirely.  Most of your examples had to do with the loss
>of money - a small loss compared to human lives or a whole planet.
>DoD systems must be even more fault tolerant than most of the systems built by
>the rest of the world.  Greg, please name one system you know about that if it
>fails could cause the loss of life (on a large scale) or even the meltdown of
>the planet.

   Air traffic control systems route some of their real time information
over commercial phone lines (if you remember some of the stories from ATT's
Manhattan problem a while ago).  Conceivably we could lose a few planeloads
of people if enough information was lost.   If someone really studied this
problem, I bet you would find out that the number of civilian-potential
deaths from informaiton systems accidents surpasses the number of military
potential deaths in peacetime, and probably equivalent to potential deaths
in today's sanitary wartimes.
   Thus, I still argue that the defense world does not have sole claim to
the need to prevent death.  Also, I would argue that the death rate from
unemployment psychology due to loss of job due to some information
miscalculation is also relevant.
     Soldier's lives should be protected at all costs, except for bankrupting
the country they are defending, because civilian livelihoods and lives (i.e.
the people paying the taxes to protect the soldiers) are killed.

Greg

-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178

Re: Data shows Top 50 Software Vendors not using Ada

[Michael Feldman]

In article <C8626p.HGB@dale.cts.com> jhb@dale.cts.com (John Bollenbacher) write
s:
[good stuff deleted]
>
>IT IS NOT THE GOVERNMENTS JOB TO MAKE ADA SUCCESSFUL IN THE OPEN MARKET.
>
>Sorry, for the caps (and the general, frustrated, tone of this posting).
>
Amen to both lines. Let's get back to the technical stuff, unless someone
has really good new information to continue this thread with.

Mike

Re: Data shows Top 50 Software Vendors not using Ada

[Gregory Aharonian]

John,

>The DoD is responsible for maintaining huge amounts of embedded software
>for very long periods of time.  If each procurer is free to choose the
>implementation language that make his program's creation cheapest, there
>will be an overwhelming number of language/dialects to be maintained by
>this department (read tax-dollars).  The point of the mandate is the ease 
>this management problem.

Like it or not, the DoD already has a multi-language problem.  A growing
number of those inside the DoD are ignoring the Mandate and using what's
available.  DARPA was, is and will always be funding stuff that will never
use Ada.  To get jobs when they are booted out of the DoD, soldier-types
are using C/C++.  All I am asking for is that the DoD confess to the
hypocrisy that the DoD is a one language institution.  It isn't because
based on their behavior, and not their words, the DoD is at best apathetic
towards Ada.  The Ada Mandate has done absolutely nothing to making the
DoD a one language organization.  Too many of its people act otherwise.
And the DoD doesn't have the decency to fund a honest study of what is
actually going on inside the Mandated world.

>IT IS NOT THE GOVERNMENTS JOB TO MAKE ADA SUCCESSFUL IN THE OPEN MARKET.

Read the Ada Mandate.  Congress, at someone's behest, but the phrase
"cost effective" into the law.  If for the DoD to make the Ada Mandate work
requires that Ada be successful in the open market, then the DoD has
to take steps to make it successful.  Let DARPA orchestrate it, since
they know very well how to influence markets and get contractors to do
what they want.

Besides, the DoD for the last three years has suppressed an Air Force
report stating that the country does not have the infrastructure to
support the Mandate.

So as long as this hypocrisy goes on at taxpayer expense, there is not
enough bandwidth available for such postings.


-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178

Re: Data shows Top 50 Software Vendors not using Ada

[Robert I. Eachus]

In article <SRCTRAN.93Jun5164025@world.std.com> srctran@world.std.com (Gregory 
Aharonian) writes:

  >    Air traffic control systems route some of their real time information
  > over commercial phone lines (if you remember some of the stories from ATT's
  > Manhattan problem a while ago).

  Greg you apparently have no concept of what a fail-safe system is.
The result of the ATT failure in Manhattan--which was not really a
software failure--was to leave many planes sitting on the ground for
hours BECAUSE there was no way to confirm that other centers could
handle the load.  Flights already in the air were not affected,
because they already had confirmed routings.

  >     If someone really studied this problem, I bet you would find
  > out that the number of civilian-potential deaths from informaiton
  > systems accidents surpasses the number of military potential
  > deaths in peacetime, and probably equivalent to potential deaths
  > in today's sanitary wartimes.

     Again a fundamental misundertanding.  Of course the number of
civilian-potential deaths is greater, in peace or wartime, because it
is usually civilians who suffer the consequences.  For peacetime
examples, there may have been a few military personnel on KAL 007 or
on the Iranian airliner shot down in the Persian Gulf, but most of the
casualties were civilians.

     In the Gulf War, many Iraqi soldiers were intentionally killed by
allied airstrikes.  Most of the people killed by accident were
civilians.  Keeping today's sanitary wartimes sanitary means keeping
all weapons under control and on target to the greatest extent
possible.

  >	Soldier's lives should be protected at all costs, except for
  > bankrupting the country they are defending, because civilian
  > livelihoods and lives (i.e.  the people paying the taxes to
  > protect the soldiers) are killed.

     Again a fundamental misunderstanding.  Soldier's lives are NOT
protected at all costs, what comes first is completion of the mission.
The primary mission of the Department of Defense is to keep the peace,
because that is the easiest way to protect the American people.

     Let me try to explain that differently.  Letting the other guy
shoot first is not good tactics.  But the Cold War never got hot
because many soldiers on both sides took all possible reasonable--and
many unreasonable--risks to avoid starting a shooting war.  Shooting
wars are hard to stop and tend to spread.

--

					Robert I. Eachus

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

Re: Data shows Top 50 Software Vendors not using Ada

[cis.ohio-state.edu!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.]

In <SANDY.93Jun4101033@beeker.cs.umass.edu> sandy@beeker.cs.umass.edu (Sandy Wi
se) writes:

>Remember the Northeast long-distance blackout? It was directly
>attributed to the semantics of 'break'.

Nonsense!  It was directly attributable to someone not knowing what
they were doing.  If you think using Ada is going to automatically
make programs error free, even when munged on by someone who doesn't
know the language, I don't think I want *my* life depending on what
you might produce!

Face it.  If people doing software know what they're doing, things
like that don't happen.  If you let someone who *doesn't* know what
they're doing mess with it, quite likely they will break it.  And the
language you use is *not* going to save you from that.

-- 
"Insisting on perfect safety is for people who don't have the balls to live
 in the real world."   -- Mary Shafer, NASA Ames Dryden
------------------------------------------------------------------------------
Fred.McCall@dseg.ti.com - I don't speak for others and they don't speak for me.

Re: Data shows Top 50 Software Vendors not using Ada

[news]

In article <C8626p.HGB@dale.cts.com>, jhb@dale.cts.com (John Bollenbacher) writ
es:
*Would we possibly save large amounts of bandwidth if we all agreed with
*Greg?  Greg, we all agree that if the mandate we lifted and the DoD were
*allowed to choose the cheapest language with which to create a system one
*of 2 things would happen:
*  1) Ada would not successfully compete in the open market and would
*  disappear, or
*  2) Ada would successfully compete and its prices would drop
*
*It is also quite obvious that in either case, the DoD would be in the
*position it was in when it decided it needed a mandate.  Now this is the
*part that needs to be read slowly by some:
*
*The DoD is responsible for maintaining huge amounts of embedded software
*for very long periods of time.  If each procurer is free to choose the
*implementation language that make his program's creation cheapest, there
*will be an overwhelming number of language/dialects to be maintained by
*this department (read tax-dollars).


No there wouldn't.  You really need to wake up and take a look around.  There
would be (for 90+ percent of all work) just two languages, one a variant of the
other:  C and C++, the languages which all of North America, left to their own
devices and the free market, have standardized on.  Both are well standardized
at this point, and these are standards which work in the real world.  Programs
written in generic C port, from just about anything to just about anything else
.In contrast, I could tell you horror stories all day and all night about 
trying to port applications in "standard" COBOL and FORTRAN, and the official 
Ada-woe BBS contained several really eye-opening items on portability with
Ada (the basic gist of the articles being "Forget it!").

When all of North America other than for a few poor schmucks with Ada hung
around their necks and a few Forth and Pascal die-hards standardizes on
one language, you can believe that there are a number of damned good
reasons.


-- 
Ted Holden
HTE

Re: Data shows Top 50 Software Vendors not using Ada

[David Emery]

The difference between the TedHolden program and the GregAharonian
improved program is that the Greg program knows a heluva lot more
about computing than the Ted program.  The Greg program knows that the
majority of software by volume is still COBOL, and that 4GL's are much
more common replacements for COBOL than C...  The Greg program
responds on the semantic level, while the Ted program still comes back
with the "Ada is different, therefore wrong" syntactic response.

One could only hope that the TedHolden program will eventually exhibit
the failing of most C programs, seg fault and dump core...
				dave

Re: Data shows Top 50 Software Vendors not using Ada

[Gregory Aharonian]

From: emery@dr_no.mitre.org (David Emery)
>
>The difference between the TedHolden program and the GregAharonian
>improved program is that the Greg program knows a heluva lot more
>about computing than the Ted program.  The Greg program knows that the
>majority of software by volume is still COBOL, and that 4GL's are much
>more common replacements for COBOL than C...  The Greg program
>responds on the semantic level, while the Ted program still comes back
>with the "Ada is different, therefore wrong" syntactic response.
>
>One could only hope that the TedHolden program will eventually exhibit
>the failing of most C programs, seg fault and dump core...

Dave,
        As a federal employee (MITRE is a FFRDC), it is illegal for you
to insult and criticize any member of the public. Check the federal regs.
I assume you are posting these insults under orders from someone at MITRE,
knowing what little initiative I see nowadays - I want names and an
investigation into why the Air Force is funding MITRE to character
assassinate civilians.  Would that you turn your invective on people
spending taxdollars, maybe the TedHolden and GregAharonian programs could
be archived.

	However, I still think you can be a useful member of society, and
in that view, please read the following Biblical passages, substituting
"federally funded employee" wherever you see the word "woman", "comp.lang.ada"
wherever you see "the churches", and "taxpayers" wherever you see "husband"
or "man".  May these inspirational thoughts help you mature in life.

			Proverbs 21:9
			Proverbs 21:19
			Proverbs 27:15-16
			1 Corinthians 11:8
			1 Corinthians 14:34-35
			Collossians 3:18-19
			1 Timothy 2:11-12

-- 
**************************************************************************
 Greg Aharonian
 Source Translation & Optimization
 P.O. Box 404, Belmont, MA 02178

Re: Data shows Top 50 Software Vendors not using Ada

[agate!howland.reston.ans.net!darwin.sura.net!uvaarpa!murdoch!holmes.acc.V]

In article <SRCTRAN.93Jun14140053@world.std.com>, srctran@world.std.com (Gregor
y Aharonian) writes:
|> 
|>    What has killed any chances of Ada surviving outside the Mandated world
|> is the rampant hypocrisy inside the Mandated world with regards to Ada,
|> hypocrisy that festers because of the silent of professinaly reputable
|> public servants.
|> 
|>  Greg Aharonian

  Having read your many posts to this newsgroup, Mr. Aharonian--and not to
start any more flame wars, which lend little to the technical discussion of
this group--I suspect you have ignored the wide usage of Ada in the non-U.S.
computing community.  You may be in the majority in the U.S. as a proponent
of C/C++, but in Europe, for instance, Ada is far more widely used than C,
and they have developed what are probably the most sophisticated software
validation and safety procedures in the world.  Ada does include many extra
features that encourage modern software engineering principles, which is not
to say that they cannot be implemented in C/C++, just that the ANSI/ISO C and
C++ standards contain nothing of the sort, leaving such additional libraries
or preprocessors to the individual compiler vendors and thus impeding
portability.  I realize you have lambasted Ada's supposed portability as well,
and I will not dispute with you the problems inherent in porting any software
code between machines.  I believe some of the edge the U.S. has lost in the
software quality race has been caused by our refusal to keep up with standards,
whether Ada or C++.  Europe uses Ada, and Japan uses Prolog (in some part),
but the U.S. uses ???.  Granted, C/C++ has obtained the largest share of the
market, simply because a C compiler is included with every Unix machine, and
Unix, while certainly not the best operating system in existence, is used
because it is familiar, because it's been around for so many years.  However,
the following languages also have large shares of the U.S. market: Ada, COBOL,
Lisp, FORTRAN, and assembly in all its manifestations (used even where a
higher-level language would have made the project easier and more
maintainable).

  The Ada mandate and documentation standards were intended to make code
for DoD projects relatively uniform, even though several exceptions
might be granted for the use of other languages.  This way, another
contracter can pick up the project where the first one left off, for updates
or maintenance.  Though certainly Ada is NOT the only language used today
in the DoD, it HAS achieved a degree of homogeneity impossible without a
mandate, and the so-called polylingual problem the DoD faces has been lessened,
not increased.  Ada may not be the best language for EVERY project, but
neither is C/C++ or Lisp or FORTRAN, but Ada was designed to support most of
the programming applicable to the DoD, and as such, it is better to side with
homogeneity and the resulting increased maintainability by writing all
general-purpose applications that do NOT demand exceptions in Ada.  That they
did not side with a C++ mandate is quite understandable, especially given
that all Ada proposals were derivative of Pascal and NOT C.  Language-
enforced readability is a lot more effective than programmer-specific
readability.

-- 
Doug lamb@Virginia.EDU             U.Va. Dept. of Electrical Engineering
All uncited opinions herein, express or implied, reflect the position of
          no other organization or individual than myself.

Re: Data shows Top 50 Software Vendors not using Ada

[James Crigler]

Doug Lamb wrote in response to Greg Aharonian:
:   Having read your many posts to this newsgroup, Mr. Aharonian--and not to
: start any more flame wars, which lend little to the technical discussion of
: this group--I suspect you have ignored the wide usage of Ada in the non-U.S.
: computing community.  You may be in the majority in the U.S. as a proponent
: of C/C++ [...]

Doug, if you think Greg is a C/C++ proponent
  1.  You haven't been reading what he writes
  2.  So is Prof. Mike (Feldman).
Greg has complained a lot about what the DOD has/hasn't done to help Ada,
and to a large extent he's been right.

:   The Ada mandate and documentation standards were intended to make code
: for DoD projects relatively uniform
[...]
:  Language-
: enforced readability is a lot more effective than programmer-specific
: readability.

Ada can't enforce readability.  Only people can.  Some of the hardest to
read code in the world was written with the intention of following some
(usually reasonable) set of coding standards (a subject recently discussed
in this group).  Control of the language namespace is the key, not the
choice of language per se.

Jim Crigler
-----------------------------------------------------------
Wow!
    --  Drew Kaplan

Re: Data shows Top 50 Software Vendors not using Ada

[Ian Wild]

dtl8v@holmes.acc.Virginia.EDU (Douglass T. Lamb) says:
>... but in Europe, for instance, Ada is far more widely used than C, ...

As it stands, this statement is not true of any Europe I've ever seen.  Have
I missed some qualifying context somewhere?  A brief scan of the job ads
here shows C/C++ still hugely more in demand than Ada.  (Of course, both
are way behind COBOL.)

ian

Re: Data shows Top 50 Software Vendors not using Ada

[cis.ohio-state.edu!math.ohio-state.edu!uwm.edu!caen!uvaarpa!murdoch!holme]

In article <1993Jun16.095953.18705@cfmu.eurocontrol.be>, ian@cfmu.eurocontrol.b
e (Ian Wild) writes:
|> dtl8v@holmes.acc.Virginia.EDU (Douglass T. Lamb) says:
|> >... but in Europe, for instance, Ada is far more widely used than C, ...
|> 
|> As it stands, this statement is not true of any Europe I've ever seen.  Have
|> I missed some qualifying context somewhere?  A brief scan of the job ads
|> here shows C/C++ still hugely more in demand than Ada.  (Of course, both
|> are way behind COBOL.)
|> 
|> ian

  Forgive me if I erred in my statement.  It seems to me the volume of data
and of publications in and on Ada are vastly higher in Europe than in the U.S.
European contracters have used Ada in building aerospace and train systems
(for instance, the European Airbus and the [now-defunct?] European Fighter
Aircraft), research systems, nuclear systems, you name it.  It appears from
the literature that Ada is far more widely taught, as are software engineering
and software validation principles, but how a typical European school's
curriculum compares to a typical American school's, I couldn't say.  Several
authors have been led to make the following statement: "In the European
computer engineering community, Ada has become the de facto standard for
_____ (you fill in the blank) systems," where various answers are safety-
critical, real-time, reliable, or distributed/multiprocessing.
-- 
Doug lamb@Virginia.EDU             U.Va. Dept. of Electrical Engineering
All uncited opinions herein, express or implied, reflect the position of
          no other organization or individual than myself.

Re: Data shows Top 50 Software Vendors not using Ada

[Jo Uhde, aka DrJo]

In article <SRCTRAN.93Jun12101402@world.std.com>, srctran@world.std.com
(Gregory Aharonian) wrote:
> 
> 
> From: emery@dr_no.mitre.org (David Emery)

[Humourous (? ;-) things written by David Emery about Greg Aharonian
deleted]
 
> Dave,
>         As a federal employee (MITRE is a FFRDC), it is illegal for you
> to insult and criticize any member of the public. Check the federal regs.
> I assume you are posting these insults under orders from someone at MITRE,
> knowing what little initiative I see nowadays - I want names and an
> investigation into why the Air Force is funding MITRE to character
> assassinate civilians.  Would that you turn your invective on people
> spending taxdollars, maybe the TedHolden and GregAharonian programs could
> be archived.

[Greg's biblical references deleted]

Huh?  Since when are employees of The MITRE Corporation considered to be
federal employees (aka civil servants (aka "silly" servants))?  Not all
MITRE sites are FFRDC's (such as those in Europe!), and there is no
indication in the domain name of Mr. Emery's e-mail account to indicate
which site it originated from.  My husband works at MITRE/Houston where's
he's just another one of those money grubbing contractor scum 8-D. 
However, he does get to call me "M'am", and I get to be on top whenever I
want ;-)

DrJo

________________________________________________________________
Jo Uhde-Lacovara, PhD   	       	NASA/JSC
with standard_disclaimer;   use standard_disclaimer; (Yes, this is stolen)
   _          |  
 _| ~-.       | *** Yankee by birth.  Texan by choice. ***
 \,  _*       | ***      Californian by dreamin'       ***   	
   \(         |  

Re: Data shows Top 50 Software Vendors not using Ada

[Phil Thornley , BAe]

In article <C8puz5.E9L@murdoch.acc.Virginia.EDU>, dtl8v@holmes.acc.Virginia.EDU
 (Douglass T. Lamb) writes:

> (for instance, ..... [now-defunct?] European Fighter Aircraft), 
                        ^^^^^^^^^^^^
I hope not! - It's still there in a hangar (but not yet in the air) 
and still being worked on.
(Although it has had some interesting times recently :-})

-- 
Phil
---------------------------------------------------------------------
 -- phil_thornley@eurokom.ie
 -- tel: +44 772 855725, fax: +44 772 855216
Usual disclaimers apply