Re: Ichbiah's letter to Anderson: Here it is

Re: Ichbiah's letter to Anderson: Here it is

[news]

I love analogies...  several in this group were less than thrilled with my
last one (the Viennese ball...);  try this one:

Picture the biggest lemon of all times in the automotive industry,  No, I
don't mean the Edsel, which was a lemon from a marketing and timing standpoint,
but nevertheless ran and did what it was advertised to do more or less.  The
British are really better in this department than we are;  somewhere between
the Humber Super-Snipe, the Rover 3500-S, and the Triumph Stag ("Swagger up
to a Stag, and stagger away!"), is probably the all time prize.

Picture the designer and chief engineer of the Humber Super-Snipe coming out
and saying:

   "You know, I'm really a pretty bright fellow, and I really do know a lot
   about building cars, you see, and I don't really want to saya whole lot 
   about the Humber Super-Snipe which is out there now...  sort of embarassing
   if you know what I mean, not one of my better efforts and I've learned
   a lot since then, but what I really want to tell you people is that the
   group of clowns who are working on NEXT YEAR's model Humber Super-Snipe
   have really messed up BAD!!!  I mean, there's just no way you want to
   buy one of those rascals!!  In fact, I don't even want to be associated 
   with Humber any more!!!  In fact, I hereby wash my hands of this whole
   sordid affair, just like Pontius Pilate, here, gimme that bucket of
   water and that towel...


There are better roads to travel out there.  One of which is the programming
language which comes with Borland's new MS-Windows version of Paradox.  This
is called ObjectPAL, and is a really beautiful implementation of a point-
and-shoot, object oriented language which is oriented towards database
applications.  

Two things I'm noticing about Ada applications:  One is that the really
big database applications appear to be the ones which get messed up the
worst, and two is that, when all is said and done, the really biggest
problem with Ada is simply the likelihood of procuring 30 or 50 competent
Ada programmers out in the middle of Podong Indiana or Oklahoma somewhere
on an army base.  C++ is obviously not a solution to that problem either.
ObjectPal just might be.  Borland has a $140 special on PdoxWin now and,
in my estimation, it would be hard to find a better way to spend $140.


-- 
Ted Holden
HTE

Re: Ichbiah's letter to Anderson: Here it is

[Tucker Taft]

In article <SRCTRAN.93Apr6103639@world.std.com> srctran@world.std.com (Gregory 
Aharonian) writes:

>    By contacting Dr. Ichbiah, I have received permission to post his letter
>to Christine Anderson to comp.lang.ada.  Here it is.  While I can't comment
>on the technical aspects of his letter, from a marketing point of view, Ada9X
>sounds like a disaster.  
 ^^^^^^

I encourage you to study the Ada 9X documents and the Ada 9X
"marketing" literature (both of which are freely available from
the Ada 9X project office, and most of which are posted on
ajpo.sei.cmu.edu as well as the Ada 9X bulletin board), 
and to go to one of our many public presentations, and draw your own 
conclusions.

Obviously Jean Ichbiah is entitled to his own view of the 9X process,
the 9X product, and the related marketing efforts, but there are many other 
more impartial, informed observers whose views are worth considering as well.  

(I certainly don't consider myself impartial, so I will spare you my views.)

S. Tucker Taft   stt@inmet.com
Ada 9X Mapping/Revision Team
Intermetrics, Inc.
733 Concord Avenue
Cambridge, MA  02138

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

>>    By contacting Dr. Ichbiah, I have received permission to post his letter
>>to Christine Anderson to comp.lang.ada.  Here it is.  While I can't comment
>>on the technical aspects of his letter, from a marketing point of view, Ada9X
>>sounds like a disaster.  
 ^^^^^^

>I encourage you to study the Ada 9X documents and the Ada 9X
>"marketing" literature (both of which are freely available from
>the Ada 9X project office, and most of which are posted on
>ajpo.sei.cmu.edu as well as the Ada 9X bulletin board), 
>and to go to one of our many public presentations, and draw your own 
>conclusions.

Tucker,
	You have been on the dole to long.  I as the potential "customer"
should not have to go out and find out about your "product", Ada9X.  You
as the vendor must seek me out by advertising and direct mailing to me
these literature on file at AJPO.  Your public presenations are all
inside the Mandated world, which as a potential customer of Ada9X does me
no good.
	It continually amazes me on the lack of interest by the Mandated
world to spend any of its own money promoting Ada outside the Mandated
world.  The assumption, I guess, is that everyone outside the Mandated
world is going to beat the door down to get to Ada9X.
	If the rest of Intermetrics were run according to this marketing
style, the results would be disasterous for Intermetrics and you would
be fired.  In the Mandated world, this marketing style gets you more
contracts.  Amazing.

	Whoever pushed for and pushes for the Mandate must accept the
responsibility of insuring its success outside the Mandated world.
Otherwise Ada will be more expensive to use under any conditions than
other languages.  Unfortunately, this responsibility has been shirked.

Greg



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

Re: Ichbiah's letter to Anderson: Here it is

[Tucker Taft]

In article <SRCTRAN.93Apr12145214@world.std.com> srctran@world.std.com (Gregory
 Aharonian) writes:
>>>    By contacting Dr. Ichbiah, I have received permission to post his letter
>>>to Christine Anderson to comp.lang.ada.  Here it is.  While I can't comment
>>>on the technical aspects of his letter, from a marketing point of view, Ada9
X
>>>sounds like a disaster.  
> ^^^^^^
>
>>I encourage you to study the Ada 9X documents and the Ada 9X
>>"marketing" literature (both of which are freely available from
>>the Ada 9X project office, and most of which are posted on
>>ajpo.sei.cmu.edu as well as the Ada 9X bulletin board), 
>>and to go to one of our many public presentations, and draw your own 
>>conclusions.
>
>. . . I as the potential "customer"
>should not have to go out and find out about your "product", Ada9X.  You
>as the vendor must seek me out by advertising and direct mailing to me
>these literature on file at AJPO.  

My point was simply that if you go out of your way 
to seek out and publicize Jean Ichbiah's letter, then
you personally might want to take the time 
to read and study the 9X proposals and literature 
and presentations, before speculating on what 
9X marketing "sounds" like.  I certainly agree that we can't
expect the general public to make such an effort, but you seem
to have almost endless energy in poking fun at the Ada community, 
and you might want to devote some of that energy to learning what 
is really out there on Ada 9X, if only to help inform the "masses"
rather than to publicize only half of the story.

> . . . Your public presenations are all
>inside the Mandated world, which as a potential customer of Ada9X does me
>no good.

Ada 9X has been presented to several organizations
that are non-DoD, non-government, and non-mandated.  Once there
are implementations, the focus will shift heavily to promoting
Ada to new markets.

>	It continually amazes me on the lack of interest by the Mandated
>world to spend any of its own money promoting Ada outside the Mandated
>world.  The assumption, I guess, is that everyone outside the Mandated
>world is going to beat the door down to get to Ada9X.

You will find that as Ada 9X transitions from vaporware to software, 
the level of marketing by the Ada 9X project to the 
non-mandated, non-government world will increase
dramatically.  Most people out there shopping at Object World,
OOPSLA, etc., are looking for a solution that they can test-drive
now, and implement in the next year or so.  Until we have some
reasonably robust implementations of Ada 9X that potential
users can get quickly, we are just blowing smoke and potentially 
wasting our marketing dollars.

Currently we are hitting the various conferences with technical
presentations.  Once we have "product" (probably late 1993 for the
GNU Ada9X "GNAT") you can be sure that we will be hitting the 
non-mandated world with more of a "marketing" orientation.

>Greg Aharonian
>Source Translation & Optimiztion
>P.O. Box 404, Belmont, MA 02178

S. Tucker Taft  stt@inmet.com
Intermetrics, Inc.
Cambridge, MA  02138

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

>	It continually amazes me on the lack of interest by the Mandated
>>world to spend any of its own money promoting Ada outside the Mandated
>>world.  The assumption, I guess, is that everyone outside the Mandated
>>world is going to beat the door down to get to Ada9X.
>

>You will find that as Ada 9X transitions from vaporware to software, 
>the level of marketing by the Ada 9X project to the 
>non-mandated, non-government world will increase
>dramatically.  Most people out there shopping at Object World,
>OOPSLA, etc., are looking for a solution that they can test-drive
>now, and implement in the next year or so.  Until we have some
>reasonably robust implementations of Ada 9X that potential
>users can get quickly, we are just blowing smoke and potentially 
>wasting our marketing dollars.

This is still an unacceptable response from people troughing tax dollars.
As I pointed out in an earlier posting, in four years of the existence
of the commercial magazine "Embedded Systems Programming" there was not
one ad from Intermetrics concerning Ada, while every month there was one
full page ad (sometimes two) publicizing your C products.  Thus, Tucker,
I do not take seriously anything you say about marketing Ada9X.  You (as
part of Intermetrics) had your chance with Ada83, a good language for
embedded systems programming, and did absolutely nothing.  Its a decent
language (with or without extensions) for object oriented programming
(according to a large amount of public posturing over the years), and you
and most others refused to push Ada in your commercial advertising.

     Waiting for Ada9X is too late, the windows of opportunity will be
all shut, as more articles about problems with Ada projects (STANFINS,
FAA, RCAS [probably]) hit the press, while glowing stories about Smalltalk
and C/C++ are there for contrast.  Given that Ada83 is a pretty good
language is, there is no excuse for all those in the Mandated world not to
be pushing it as much as possible, if nothing else, for the practice of
fostering Ada.

     But I see nothing where it involves spending your own money.  Please
explain why not once in four years, Intermetrics did not place a single
ad mentioning its Ada capabilities in Embedded Systems Programming.
Was it embarassment for being associated with Ada?  Was it the realization
that Ada can't be profitably marketed outside the Mandated world, which
does not have captive markets that can be charged exorbitant compiler
prices?  Was it that Intermetrics management figured that commercial Ada
only has a negative return on investment?  

    Not that Intermetrics is alone in this hypocrisy.  All I know is that
the company with the best record for marketing Ada outside the commercial
world, Alsys, through its leader, Jean Ichbiah has expressed the most
reservations about the marketability of Ada9X.  I question the agenda of
any representatives of Intermetrics on the Ada9X effort - the company's
prior history makes it suspect in the future success of Ada.

    In fact, given the past dismal performance of the Mandated world in
pushing Ada commercializing, one could cynically propose that some on the
Ada9X effort are developing an Ada9X that will keep its installed, captive
market effort with improvements that address the limited needs of the
Mandated world but provide little in the way of marketing advantages in
the non-Mandated world.  This way the captive market is kept, allowing the
continual charging of compiler prices out of sync with the free markets.
"We tried to make a more marketable Ada but people just didn't give us
a chance" - I can hear the future moaning and wailing.
Fortunately, I am not a cynic, so I won't propose such a viewpoint.

     Much like IBM has continually refused to include Ada in its commercial
product line advertising and marketing, Intermetrics refusal to advertise
its Ada product and services says much more about their true beliefs about
Ada than the endless public posturing.  Stop taking tax dollars to profess
your love for Ada and do whatever you want.

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

Re: Ichbiah's letter to Anderson: Here it is

[Charles H. Sampson]

In article <SRCTRAN.93Apr26204948@world.std.com> srctran@world.std.com (Gregory
 Aharonian) writes:
>
[Words about the lack of Ada marketing by Ada vendors, including a
 response by Tucker Taft that Ada 9X will be marketed.]

>This is still an unacceptable response from people troughing tax dollars.
> ...
>     Much like IBM has continually refused to include Ada in its commercial
>product line advertising and marketing, Intermetrics refusal to advertise
>its Ada product and services says much more about their true beliefs about
>Ada than the endless public posturing.  Stop taking tax dollars to profess
>your love for Ada and do whatever you want.

     Greg makes enough valid points that I continue to read his posts,
repetitious though they may be, but this is too much.

     Companies that are "troughing tax dollars" do so on contract.  In
almost all cases these contracts have been won through a competitive bid-
ding process.  (That the process is inherently wasteful and often unfair is
an interesting story, but not particularly pertinent to Ada.)  Those con-
tracts spell out the contractors' obligation.  I doubt that any Ada-related
DoD contract has ever required the contractor to shill for the Ada lan-
guage.  Certainly the ones I've been involved with do not.

     I'm willing to listen to the argument that failure to promote Ada is
short-sighted on the part of Ada vendors, although I don't consider that a
settled issue.  However, since the government itself has no right to demand
that they spend any of their profits on promoting Ada, I have no idea why
Greg has appointed himself to this role.

                                   Charlie

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

>     I'm willing to listen to the argument that failure to promote Ada is
>short-sighted on the part of Ada vendors, although I don't consider that a
>settled issue.  However, since the government itself has no right to demand
>that they spend any of their profits on promoting Ada, I have no idea why
>Greg has appointed himself to this role.

    The government has the right to demand contractors do whatever is in
the best interests of the country.  If the contractors doesn't want to
comply, they can get business elsewhere.  Look at the current administration,
with its "let's make the government a venture capital investor".  It's
telling companies how to do technology transfer.  The government is always
intervening in the marketplace - just look at how Japan's governmment
turned the Nikkei index around.

    You can't be that naive to not realize that if someone from the DoD
called Intermetrics management and "suggested" that they place some Ada
ads, that Intermetrics would place these ads.

    But with the Ada Mandate as a national law, the DoD (which I assume
supports the Mandate and pushed for it), has to make sure that the
language is thriving enough to insure a sufficient supply of programmers,
tools, and advanced software technology compatibility.  Outside the
Mandated world, Ada is practically non-existent to the extent that
TRW's Mosemann study prediction that Ada will eventually cost more to use
than C++ is coming true quicker and quicker.

    Given the Mandate, and the reality of Ada, when all of its contractors
and compiler vendors drop the ball promoting Ada, its up to the DoD to
assume the responsibility, especially to demand more from anyone taking
Ada money.

   Right now, ignoring the Cobol and Fortran communities, Ada makes up
less than four percent of all non-Mandated software development activities,
jobs, programmers, tools, products and companies, after ten years of
billion dollar budgets going into programs using Ada.  The language is
dead, and Ada9X with its upward incompatibility and wait-till-its-ready
marketing strategy won't make a bit of difference, GIVEN THE DOD'S USE
OF ITS CURRENT ADA SOFTWARE ENGINEERING CONTRACTORS, and given success
stories like JINTACCS, internal DoD efforts cost-effectively and
successfully using C/C++.

     And getting back to Tucker, I would much rather see the Ada9X efforts
in the hands of Ichbiah and Alsys, who spend their money promoting Ada,
as opposed to Taft and Intermetrics, who spend their money promoting C.


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

Re: Ichbiah's letter to Anderson: Here it is

[cis.ohio-state.edu!news.sei.cmu.edu!ajpo.sei.cmu.edu!falis]

Just as a point of fact, Greg's statement about Alsys' not being committed 
to aggressive marketing of Ada9x, is inaccurate.  Jean Ichbiah resigned
from a regular role at Alsys in 1991.  His opinions do not reflect
the Alsys corporate stance on Ada9x.  - Ed Falis, Alsys

Re: Ichbiah's letter to Anderson: Here it is

[Charles H. Sampson]

In article <SRCTRAN.93Apr27141025@world.std.com> srctran@world.std.com (Gregory
 Aharonian) writes:

   [Previously, I wrote:]

>>     I'm willing to listen to the argument that failure to promote Ada is
>>short-sighted on the part of Ada vendors, although I don't consider that a
>>settled issue.  However, since the government itself has no right to demand
>>that they spend any of their profits on promoting Ada, I have no idea why
>>Greg has appointed himself to this role.
>
>    The government has the right to demand contractors do whatever is in
>the best interests of the country.  If the contractors doesn't want to
>comply, they can get business elsewhere.  ...

     Wrong.  The government only has the _right_ to demand that contractors
fulfill their contracts.  I certainly hope that the government considers
the best interests of the country, in spite of occasional bits of evidence
to the contrary.  If they decide that promoting Ada in the non-Mandated
world is worthwhile, they can write that into their RFPs (Request for Pro-
posal) and let the contractors factor in the cost when they bid.

>    You can't be that naive to not realize that if someone from the DoD
>called Intermetrics management and "suggested" that they place some Ada
>ads, that Intermetrics would place these ads.

     I'm not so naive and I've seen similar acts by the government.  I
consider them totally unconscionable, not too terribly far removed from a
kickback.  Are you claiming that it's something the government should do as
considered policy?

>    But with the Ada Mandate as a national law, the DoD (which I assume
>supports the Mandate and pushed for it), has to make sure that the
>language is thriving enough to insure a sufficient supply of programmers,
>tools, and advanced software technology compatibility. ...

     For decades the DoD has trained programmers in its arcane languages. 
When the Ada effort began it was a purely DoD project aimed a producing a
purely DoD language, with the implicit understanding that the DoD would
continue to train programmers to use its single new language.  Just because
Ada turned out to be an excellent general purpose software engineering
language doesn't mean that the DoD has suddenly been saddled with a respon-
sibility to promote it on the outside.  A cost/benefit analysis might prove
that for the DoD it is better to ignore the outside world.  (To my knowl-
edge such an analysis has not been done.)  This is not to deny that the
computing community at large might be better served by a wider acceptance
of Ada.  I believe that to be the case.  What's not clear to me is why some
people consider it obvious that the DoD should spend part of its budget
promoting this wider acceptance.

>    Given the Mandate, and the reality of Ada, when all of its contractors
>and compiler vendors drop the ball promoting Ada, its up to the DoD to
>assume the responsibility, especially to demand more from anyone taking
>Ada money.

     Again, who is "taking Ada money"?  The DoD awards contracts and the
contractors try to fulfill them.  If the DoD wants the contractors to do
something, put it into the contract.

                                   Charlie

Re: Ichbiah's letter to Anderson: Here it is

[Robert Kitzberger]

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


>    The first responsibility of the DoD is to develop systems as cheaply
>as possible (without comprimising safety, quality, etc). 

Surely you don't mean this.  The first responsibility is to defend our
country.  Developing systems hopefully helps support that goal.
Within systems development, the responsibility is to develop
effective, safe systems, as inexpensively as possible.  Perhaps you've
spent so much time arguing about Ada (usually effectively) that you've
suffered from Priority Inversion? ;-)

	.Bob.
--
------------------------------------------------------------------------
Bob Kitzberger                          Internet:   rlk@rational.com
Rational, Grass Valley, CA              CompuServe: 70743,1550
type Opinion is private; 

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

>    But with the Ada Mandate as a national law, the DoD (which I assume
>>supports the Mandate and pushed for it), has to make sure that the
>>language is thriving enough to insure a sufficient supply of programmers,
>>tools, and advanced software technology compatibility. ...
>
>     For decades the DoD has trained programmers in its arcane languages. 
>When the Ada effort began it was a purely DoD project aimed a producing a
>purely DoD language, with the implicit understanding that the DoD would
>continue to train programmers to use its single new language.  Just because
>Ada turned out to be an excellent general purpose software engineering
>language doesn't mean that the DoD has suddenly been saddled with a respon-
>sibility to promote it on the outside.  A cost/benefit analysis might prove
>that for the DoD it is better to ignore the outside world.  (To my knowl-
>edge such an analysis has not been done.)  This is not to deny that the
>computing community at large might be better served by a wider acceptance
>of Ada.  I believe that to be the case.  What's not clear to me is why some
>people consider it obvious that the DoD should spend part of its budget
>promoting this wider acceptance.

    The first responsibility of the DoD is to develop systems as cheaply
as possible (without comprimising safety, quality, etc).  If it is indeed
true that over the life-cycle it is cheaper to develop systems in C/C++
than in Ada (and there is enough circumstantial evidence like JINTACCS
to study the problem), than I feel as a taxpayer that the DoD has two choices.

   Choice one is to insure that Ada is as cost-effective as C/C++, which
requires many work outside of the Mandated world in terms of greater Ada
acceptance in programmer supply, tools, products and companies.  Either
on its own, or by making its contractors do so, it has to make Ada competitive.

   Choice two is to allow C/C++ to be used in defense projects, to altar[sic]
the Mandate to allow C/C++ to be used without waivers (if this is not
already true , as in the JINTACCS and who knows what else).

   I have been gathering many statistics, and outside the Mandated world,
C/C++ activities in a macroeconomic sense are at least ten to twenty times
larger than for Ada.  With all of this market attention on C/C++, over
time C/C++ programmers, tools and libraries are going to get better and
better, making their use in development very cost effective.  At the same
time, for Ada things will stagnate and it will drop farther and farther
behind C/C++ to the point the differential will be so significant that
those currently battle against Ada inside the DoD will have the upper hand.

    The whole point of all of my messages to date is that the Ada Mandate
is a gross distortion of the marketplace with very significant economic
implications that have never been fully examined by the DoD.  The one
study to touch upon the subject, the Mosemann studies, when the substudies
weren't contradicting each other, were highly deficient on the demographic
statistics of the economics of C/C++ and Ada as to render their conclusions
irrelevant.

    Given the related business mistakes such as the software reuse efforts
and the STARS welfare project, it is obvious that there is a need for serious
economic analysis of the microeconomics of defense software development.
I have much raw data that does not paint a pretty picture.  At best, given
what I have seen, over the life-cycle, using C/C++ is slightly more cost
effective than Ada.  At worst, it's much better.  To stick with the Ada
Mandate and one language is then a betrayal of the American taxpayer.

>>    Given the Mandate, and the reality of Ada, when all of its contractors
>>and compiler vendors drop the ball promoting Ada, its up to the DoD to
>>assume the responsibility, especially to demand more from anyone taking
>>Ada money.
>
>     Again, who is "taking Ada money"?  The DoD awards contracts and the
>contractors try to fulfill them.  If the DoD wants the contractors to do
>something, put it into the contract.

   Exactly.  The DoD is not a business.  We expect them to be a bunch of
socialist killers, that is a top-down, centralized entity of soldiers with
weapons.  I have no problem with that.  What I do have a problem with is
when these same people try to act like capitalist business men.  They
aren't and they can do it very well.  The Ada Mandate forces the DoD to
become businessmen - to have to deal with marketing, competition, free
market supply and demand, cost/benefit analysis, investment, etc.  The
DoD has never done this very well, and the one agency that does this
some of the time, DARPA, has to violate the Mandate to fulfill its
mission, since none of the advanced software they fund (AI, parallel,
nat lang, distributed) is being seriously done in Ada anywhere in the
country.

    The Mandate is not supposed to be a tool of control for the DoD over
its contractors, nor should it be an artificial prop for a questionable
economic proposition.

           IF ADA IS AS COST-EFFECTIVE AS ITS PROPONENTS PROCLAIM,
           THE MANDATE WOULD BE IRRELEVANT, SINCE FOR EVERY BIDDED
           PROJECT, THE ADA BIDS WOULD ALWAYS BE LOWER.

It isn't, and the Mandate is a disservice to the American public.

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

Re: Ichbiah's letter to Anderson: Here it is

[Charles H. Sampson]

In article <SRCTRAN.93Apr29102415@world.std.com> srctran@world.std.com (Gregory
 Aharonian) writes:

>    The whole point of all of my messages to date is that the Ada Mandate
>is a gross distortion of the marketplace with very significant economic
>implications that have never been fully examined by the DoD. ...

     I don't have to stretch too far to accept the claim that the Ada man-
date is a gross distortion of the marketplace.  However, I would rather
characterize it as simply ignoring the marketplace.  The mandate reaffirms
the basis of the entire effort that led to Ada: the DoD will save substan-
tially by focussing on a single language for its tactical systems.  Fur-
ther, it chooses Ada as that single language.  Usually it's not a good idea
for a legislature to decide technical issues, but in my prejudiced opinion
they lucked into the right decision this time.

>                                                       ...  At best, given
>what I have seen, over the life-cycle, using C/C++ is slightly more cost
>effective than Ada.  At worst, it's much better.

     [referenced later]

>                                 ...  The Ada Mandate forces the DoD to
>become businessmen - to have to deal with marketing, competition, free
>market supply and demand, cost/benefit analysis, investment, etc.

     To the contrary, the mandate doesn't require the DoD to be businessmen
at all, or at least not much.  There's generally no decision to be made;
use Ada.  They are only required to be businessmen to the degree that they
can evaluate a cost/benefit analysis in support of a waiver.

>           IF ADA IS AS COST-EFFECTIVE AS ITS PROPONENTS PROCLAIM,
>           THE MANDATE WOULD BE IRRELEVANT, SINCE FOR EVERY BIDDED
>           PROJECT, THE ADA BIDS WOULD ALWAYS BE LOWER.

     That's not the way military contracts are bid.  Usually implementa-
tion, IV&V, and a succession of maintenance contracts are bid independent-
ly.  It's well-known that implementation costs are a small part of the
life-cycle costs of military software.  If contractors were allowed to
choose the implementation languages, they might well say, "We can do this a
lot cheaper in B- than in Ada," provided they didn't have to consider the
costs of the follow-on maintenance contracts.  As a matter of fact, during
the process that led to Ada it was accepted that the costs of implementing
in this new language might be higher than in some other languages.  It was
decided that this was acceptable if total life-cycle costs were lowered.

     Finally, the Ada mandate has not precluded other languages.  The only
requirement is that use of another language on a particular project must be
shown to be cost-effective over the entire life of the software.  Based on
your earlier statements, this shouldn't be too hard to do.  I wonder why it
hasn't happened often.

                                   Charlie

Re: Ichbiah's letter to Anderson: Here it is

[Mike Ryer]

Greg complains about Intermetrics not advertising in embedded computer
magazines.  There is one overwhelming reason for that: we don't have
any Ada compilers for commercial embedded chips, just military ones.
Perhaps we should, but we invested our money elsewhere for Ada '83.

Even if we had Ada compilers for the Intel, Motorola, and similar chips,
we wouldn't necessarily advertise in magazines.  We run ads for our C
compilers to a) inform people that they exist, and b) transmit the message
that our products are reputable and popular.  We don't expect anyone to
choose C over Fortran, Smalltalk, or Ada because they saw a C compiler
ad.  We don't expect to convince anyone that our product is the best via
an ad either.  Ads yield little more than "awareness".

For Ada, people find out about compilers by checking the validated compiler
list.  We don't need the "we exist" sort of advertising.  The presence of
our C compiler ads gives us the "we're reputable" image which carries over
to Ada and our other business areas.

Have you noticed that the major Ada vendors are reporting greatly increased
sales this year?  The use of Ada is at an all-time high.  Reports of her
death are greatly exaggerated.

Greg, I agree that more marketing of Ada outside the mandated world is needed.
I just don't favor magazine ads for compilers.  The videotapes that we'll 
produce, and the other marketing activities already in progress, plus the
availability of GNAT, plus the entry of Ada into the OOP fray, will make a
difference.  Be patient a little longer, we're going as fast as we can on
all of these activities.

Mike Ryer
Deputy Manager
Development Systems Department
Intermetrics, Inc.

Re: Ichbiah's letter to Anderson: Here it is

[Tucker Taft]

In article <SRCTRAN.93Apr26204948@world.std.com> 
  srctran@world.std.com (Gregory Aharonian) writes:

> . . .
>     But I see nothing where it involves spending your own money.  Please
>explain why not once in four years, Intermetrics did not place a single
>ad mentioning its Ada capabilities in Embedded Systems Programming.

As Mike Ryer pointed out, Intermetrics was never in the off-the-shelf,
Ada 83 cross-compiler market.  Our biggest Ada 83 market has
been IBM/370 MVS.  Rest assured that we spent plenty of our
own money productizing, maintaining, and marketing our MVS Ada compiler.
(We developed the basic Ada compiler technology under a competitive
contract with the Air Force.  That covered significantly less than
half of the ultimate cost to bring it to market.)
But as you might guess, we didn't feel that Embedded Systems Programming
was quite the perfect fit for our expected MVS Ada customer.

Every company makes its own decisions about what products
to develop and how to sell them.  You are certainly mistaken
if you think the Ada compiler vendors can just sit around on their
duffs and pull in the money from "captive" compiler sales.  First
of all, looking strictly at the "captive" DoD market, there is not
a single significant host/target combination that does not have
multiple vendors competing.  Secondly, even on the "cushy" government
contracts, it is easy to lose money because of the long drawn-out
competion phase that precedes contract award, followed by the
frequently shifting requirements thereafter.

I totally agree with you that Ada 83 as a language was not
marketed well to the "world."  However, unlike you, I don't
lay the blaim on individual stupidity or laziness, but rather on 
a more community-wide "group think" in the early Ada days that we
had an obviously "better mousetrap" and the world would beat a path
to our doorsteps.  By the time some Ada companies began to break
through to a more realistic assessment of Ada and the required
marketing, they had dug themselves so deep into debt that they couldn't
afford to start over with a more market-oriented strategy.
(It is interesting to look at 10-year old copies of CACM.  SofTech
used to advertise about Ada every month.  It didn't get them very far...
So much for advertising without a market-ready product to back it up.)

Ultimately, many Ada companies just hunkered down with their current
market niche and hoped for the best.  It seems that now, partly
as a result of consolidation in the industry, some Ada companies
are getting themselves onto a sound enough financial footing to
think about (re)launching a more broad-based marketing strategy,
perhaps using Ada 9X as the excuse to get potential customers
to take a "new" look at Ada.  But this takes time and money,
both of which are still in short supply at most Ada companies, partly
due to past misreading of the market.  

At this point, rather than harping on the many mistakes of the
Ada community in the past, we should begin focusing on how
to improve things in the future.  I believe that Chris Anderson
of the 9X project office is doing just that.  AdaNet and the AdaIC
also both seem to be taking a more proactive role in trying to
spread the word.  Other government-sponsored Ada-related organizations,
such as ASSET, seem still to be in a "hunkered-down" or "come-find-us"
mode, which is admittedly a shame.  Of course if they don't read
comp.lang.ada, perhaps we need to find a more constructive way
to wake them up than via periodic flames.  

>Greg Aharonian

S. Tucker Taft  stt@inmet.com
Intermetrics, Inc.
Cambridge, MA  02138

Re: Ichbiah's letter to Anderson: Here it is

[Mike Ryer]

In article <SRCTRAN.93Apr29102415@world.std.com> srctran@world.std.com (Gregory
 Aharonian) writes:

>
>           IF ADA IS AS COST-EFFECTIVE AS ITS PROPONENTS PROCLAIM,
>           THE MANDATE WOULD BE IRRELEVANT, SINCE FOR EVERY BIDDED
>           PROJECT, THE ADA BIDS WOULD ALWAYS BE LOWER.


NOT!

Languages are chosen at the beginning of the life cycle, based on RFPs for 
Development and/or Production.  These RFPs never include life-cycle 
maintenance, enhancements, and support of the software.  That is a separate
RFP.  Both DOD and its contractors have incentives to make decisions based
on *development* costs rather than total costs.  As long as these incentives
(profits and promotions) are not changed, some other mechanism is needed to
ensure that the ultimate maintenance and supportability of the software gets
considered in the language decision.  That is the purpose of the mandate.
Have you got a better approach?

Particularly given the DOD's tendency to have software maintenance done
by "seven dollar engineers" under labor-ordering contracts, the real cost
of using an undisciplined language (or set of languages) in development
is immense.

Mike Ryer

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

Mike Ryer from Intermetrics writes:

>Even if we had Ada compilers for the Intel, Motorola, and similar chips,
>we wouldn't necessarily advertise in magazines.  We run ads for our C
>compilers to a) inform people that they exist, and b) transmit the message
>that our products are reputable and popular.  We don't expect anyone to
>choose C over Fortran, Smalltalk, or Ada because they saw a C compiler
>ad.  We don't expect to convince anyone that our product is the best via
>an ad either.  Ads yield little more than "awareness".
>
>For Ada, people find out about compilers by checking the validated compiler
>list.  We don't need the "we exist" sort of advertising.  The presence of
>our C compiler ads gives us the "we're reputable" image which carries over
>to Ada and our other business areas.

    If this was any other language, and you any other company, I would
not give a damn what Intermetrics does.  By Ada is a federally mandated
language on which national security is somewhat dependent, and you are
acompany acepting tax dollars in support of this policy.  While your
advertising strategy might be  in the best interests of  Intermetrics,
it is not in the best interests of the country, for the following 
reason.  Many in the non-Mandated world tend to be interested in 
buying a technology based on the exposure they see for it in ads and
articles, independent of any company's particular products.
    People are not seeing Ada anywhere in the non-Mandated press, while
they do see other languages like C++ and Smalltalk.  Thus your self
interested advertising strategy, which I would normally support, is
unacceptable as long as you remain an Ada contractor.  The language,
despite vendor claims, is ten to thirty times less thriving than C/C++
in the non-Mandated world.

    So I stil disagree.  The Ada vendors and contractors have to make an
extra effort to evangelize Ada in the non-Mandated world, as long as you
accept Ada tax dollars dollars and support a questionable national policy
(i.e. the Mandate).  Get off the Ada dole, and I don't care what you do.
But as long as you are part of the effort, especially something as
marketing-ly important as the Ada9X redesign, you have an ethical 
obligation to push Ada whereever possible.

     And your hopes for Ada93 in the OOP fray make little business sense.
You will be offering too little too late to a community making its
decisions now.


>Have you noticed that the major Ada vendors are reporting greatly increased
>sales this year?  The use of Ada is at an all-time high.  Reports of her
>death are greatly exaggerated.

    Until I see the breakdown of your statistics, I don't believe this.
Come to Wadas, where I will hopefully be presenting endless statistics
showing that by any measure of thriving in the non-Mandated world, that
Ada is ten to thirty time less thriving than C/C++ (in terms of
products offered,  corporate use, new job offers, advertising lineage,
trade show presence).

    Because of the Mandate, and DoD tolerance of contractor and vendor
behavior, Ada is and will be a niche languge like Forth and Prolog in
the non-Mandated world.   As an independent observer noted in his
Business Week article, "Ada is an obscure DoD language".

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

Re: Ichbiah's letter to Anderson: Here it is

[Gregory Aharonian]

Tuck writes:

>At this point, rather than harping on the many mistakes of the
>Ada community in the past, we should begin focusing on how
>to improve things in the future.  I believe that Chris Anderson
>of the 9X project office is doing just that.  AdaNet and the AdaIC
>also both seem to be taking a more proactive role in trying to
>spread the word.  Other government-sponsored Ada-related organizations,
>such as ASSET, seem still to be in a "hunkered-down" or "come-find-us"
>mode, which is admittedly a shame.  Of course if they don't read
>comp.lang.ada, perhaps we need to find a more constructive way
>to wake them up than via periodic flames.  
   ( If they don't read comp.lang.ada, they should be fired.  These
efforts are not on the job training programs on how to exchange
information in the electronic era.  Let more qualified people run
these efforts. )


   Since current policy is being made by many of those who made the
mistakes in the past, it is very important to harp on these issues.
I am Bayesian, and by arguing that the past is a good predictor of the
near future, I would say Ada is in deep trouble.
   The Ada9X office has no idea of the true views of Ada in the non-Mandated
world, either the raw demographics of acceptance and usage, or in terms
of public opinion.  I will bet you an all-chocolate ice cream sundae at
the Friendly's on Concord ave that the Ada9X office does not have and
has not funded any sort of marketing/demographic survey of Ada usage and
recognition outside the Mandate world.  The office is developing
policy in the blind.  Chris Anderson's comments in Utah were way off
the mark when it comes to non-Mandated OO interest in Ada.

   AdaNet, ASSET, VCOE and the rest of them are all bureaucratic
enterprises gathering little real-world business experience to help
future Ada efforts.   If you want to see this first hand, ask ASSET
for their components schema and look at it from a free market point
of view - their schema would make Karl Marx happy, not Adam Smith.
It's a shame no one is overseeing what is going on at ASSET.  To
rest any of Ada's future on these efforts is to admit defeat now.

    There is a much simpler way to resolve all of these issues and
eliminate my endless pestering of everyone on c.l.a:  first get the
DoD to factor life-cycle mainentance and support costs properly into
contracts, and then get rid of the Ada Mandate.  Let market forces
decide who is right and ensure that the DoD gets the most cost 
effective support to meets its development needs.
    The defenders of free markets and capitalism should not rely on
such a socialist piece of legislation as the Mandate.  Isn't that why
we spent this country into debt fighting the Cold war?

Greg Aharonian

P.S. I don't do lunch, so I will want to collect my sundae after work.

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

Re: Ichbiah's letter to Anderson: Here it is

[news]

In article <rlk.736095268@pelton>, rlk@pelton.Rational.COM (Robert Kitzberger) 
writes:
*srctran@world.std.com (Gregory Aharonian) writes:
*
*
*>    The first responsibility of the DoD is to develop systems as cheaply
*>as possible (without comprimising safety, quality, etc). 
*
*Surely you don't mean this.  The first responsibility is to defend our
*country. 

Having to spend ten times the going rate in both money and time for everything
you ever do will prevent you from doing that.  That's a hell of a good reason
for deep-sixing Ada.



-- 
Ted Holden
HTE