Ada vendors abandon Ada in November Emb.Sys.Prog.

Ada vendors abandon Ada in November Emb.Sys.Prog.

[Gregory Aharonian]

     For those of you who can afford to go to Tri-Ada next week, while you are
there and see a room full of Ada vendors saying all sorts of nice things about
Ada, ask yourself (and them): "How come these people never show up anywhere
else as enthusiastically as at Tri-Ada and STSC?".

     Case in point.  The November 1994 issue of Embedded Systems Programming
has two very well written articles about Ada.  The first is titled "Ada for
Space Applications" written by Richard Riehle, and the second is titled
"Cruising for Ada" written by Do-While Jones.  The articles clearly illustrate
some of the benefits of Ada, especially Jones' article which has some nice
source examples and cost/benefit analysis comments.

     These articles are almost good enough that someone reading them might say
"Gee - maybe I should get some info on Ada compilers and try it out."  It is
this type of comment that all Ada Dual-Use activities should be directed
towards - getting people to want to try out Ada.

     It is then once again very disappointing to see the complete apathetic
attitude of the Ada vendors to take advantage of these special Ada issues of 
magazines - NONE OF WHOM ADVERTISED IN THIS ADA ISSUE OF EMBEDDED SYSTEMS
PROGRAMMING.  This is intolerable and just further illustrates the hypocrisy
of some Ada supporters.  This issue of Embedded Systems Programming would 
have been ideal to place many Ada ads to capture the interest generated by
these articles of Riehle and Jones, and certainly as you walk the floor of
Tri-Ada you will see plenty of Ada vendors.  This issue would have been
perfect to drum up new leads and business with an ad.

     And Ada vendors can't claim ignorance of this magazine, after all, every
month Intermetrics manages to find the inside front cover of the magazine to
place a full page ad for its 68XXX real time C programming environment.  Why
not an Ada ad, especially given Intermetric's participation in Ada9X efforts?
Why not announce a seminar series by Tucker Taft about the benefits of Ada9X,
if Intermetrics doesn't want to offer Ada tools?  Why didn't any of the STARS
contractors with the great breakthrough tools advertise in this issue?  Why
didn't any of the Ada9X contractors advertise something about Ada9X?

     What's more embarassing and undermining for Ada is the Tri-Ada ad that
appears in this issue in the middle of Jones' article.  At one point the ad
talks about "a catalog of industry resources" implying vendors somewhere, but
it's hard not to conclude that the Ada industry isn't anything more than a
marginal niche, based on the nearly complete absence of Ada advertising and
promotions outside the Mandated world.

     Besides, this Tri-Ada is still poorly prepared and does nothing to
promote Ada.  For example, "That's why Ada is modular".  "Ada uses standard
interfaces between modules".   BIG FRIGGIN DEAL.  Many languages whose vendors
actually advertise claim that benefit to the extent that no one believes it
anymore.  C++, Smalltalk, Basic, and Eiffel ads and articles all claim this,
so it is marginal for it to appear in this Tri-Ada ad.  Just hope no one
reading these comments read the Magnavox consultant's trashing-Ada article.

     Or we have "Just as important is Ada's easy readability.  Programmers
can understand code written by others, no matter where or when they wrote it."
This is a very transparent lie, because most programmers know that you can
really beautiful code in any language, and you can write really ugly code in
any language.  This is an idiotic claim to make. LANGUAGES DON'T KILL READABLE
CODE, PROGRAMMERS KILL READABLE CODE.

     "And when you consider that through the life cycle of your average large
system, changes amount to eight time the original cost, you quickly see the
financial significance of Ada's software design".  Well by reading just the
text of this ad, I don't see the financial significance of Ada.  There is no
data to make such a conclusion, so claiming it not only is wrong, but plays
on the vaporware fears of many programming consumers.  Further no one is going
to believe any such claims made by something coming out of the defense world.
And finally, "life cycle" is a phrase with little meaning in the commercial
software world, unfortunately.  So selling to that need will fall on deaf
ears, especially when you mention nothing about the size and speed of 
executables, which do concern commercial programmers.  You might as well claim
that Ada cures AIDS, a claim that follows as logically from the ad as the
claim of Ada's financial benefits.

     This ad, like previous ads, is poorly prepared for the audience it is
reaching out to, and the ad agency that prepared them should be sued, as well
as the Ada vendors who refuse to butress such ads and well placed Ada articles
with their own ads.  I would die if I saw an ad for an Ada compiler from
Alsys and Rational which actually mentioned the price.  Vendors of every other
language offer me CDROMs with compiler, linkers, debuggers, browsers and other
tools for $99 to $499 or whatever, but most magazine readers wonder if Ada
compilers are sold anywhere in any form.

     So as you go through the Tri-Ada trade show, ask all of these vendors,
who will be saying all sorts of nice things about Ada to their DoD sources,
especially those receiving the bucks from DoD Ada contracts like Ada9X and
STARS and ICASE, ask them why they don't say the same things about Ada with
the same fervor, with their own dollars, outside the Mandated world. They
fool no one.  And ask yourself why the DoD continues to tolerate such
nonsense.

Greg Aharonian

Re: Ada vendors abandon Ada in November Emb.Sys.Prog.

[Kent Mitchell]

Gregory Aharonian (srctran@world.std.com) wrote:
[ snipped out a lot of stuff which is the standard stuff about Ada vendor
not spending THEIR OWN MONEY they way Greg think they should (i.e. on
advertising ]

:      "And when you consider that through the life cycle of your average large
: system, changes amount to eight time the original cost, you quickly see the
: financial significance of Ada's software design".  Well by reading just the
: text of this ad, I don't see the financial significance of Ada.  There is no
: data to make such a conclusion, so claiming it not only is wrong, but plays
: on the vaporware fears of many programming consumers.  Further no one is going
: to believe any such claims made by something coming out of the defense world.
: And finally, "life cycle" is a phrase with little meaning in the commercial
: software world, unfortunately.  So selling to that need will fall on deaf
: ears, especially when you mention nothing about the size and speed of 
: executables, which do concern commercial programmers.  You might as well claim
: that Ada cures AIDS, a claim that follows as logically from the ad as the
: claim of Ada's financial benefits.

I agree with Greg that the ad did very little to "prove" it's conclusion.
However, the statement that the term "life cycle" is not used in the
commercial world is complete bunk.  This is exactly the term used in all
cases I've been involved with in the commercial world (remember with
Greg's assertion that Rational is abandoning Ada all I ever seem to do is
visit commercial sites these days ;-) ).  To say that life cycle cost mean
nothing to the commercial world when compared to "the size and speed of
the executable" show how little Greg really knows about commercial software
development.  Commercial software development is *more* concerned with life
cycle costs as they are (to use a well known refrain) SPENDING THEIR OWN
MONEY.

--
Kent Mitchell                   | One possible reason that things aren't
Technical Consultant            | going according to plan is .....
Rational Software Corporation   | that there never *was* a plan!

Re: Ada vendors abandon Ada in November Emb.Sys.Prog.

[whiting_ms@corning.com (Matt Whiting)]

In article <39dpnv$9bt@rational.rational.com>, kdm@puppy.rational.com (Kent Mitchell) writes:
> Gregory Aharonian (srctran@world.std.com) wrote:
> [ snipped out a lot of stuff which is the standard stuff about Ada vendor
> not spending THEIR OWN MONEY they way Greg think they should (i.e. on
> advertising ]
> 
> :      "And when you consider that through the life cycle of your average large
> : system, changes amount to eight time the original cost, you quickly see the
> : financial significance of Ada's software design".  Well by reading just the
> : text of this ad, I don't see the financial significance of Ada.  There is no
> : data to make such a conclusion, so claiming it not only is wrong, but plays
> : on the vaporware fears of many programming consumers.  Further no one is going
> : to believe any such claims made by something coming out of the defense world.
> : And finally, "life cycle" is a phrase with little meaning in the commercial
> : software world, unfortunately.  So selling to that need will fall on deaf
> : ears, especially when you mention nothing about the size and speed of 
> : executables, which do concern commercial programmers.  You might as well claim
> : that Ada cures AIDS, a claim that follows as logically from the ad as the
> : claim of Ada's financial benefits.
> 
> I agree with Greg that the ad did very little to "prove" it's conclusion.
> However, the statement that the term "life cycle" is not used in the
> commercial world is complete bunk.  This is exactly the term used in all
> cases I've been involved with in the commercial world (remember with
> Greg's assertion that Rational is abandoning Ada all I ever seem to do is
> visit commercial sites these days ;-) ).  To say that life cycle cost mean
> nothing to the commercial world when compared to "the size and speed of
> the executable" show how little Greg really knows about commercial software
> development.  Commercial software development is *more* concerned with life
> cycle costs as they are (to use a well known refrain) SPENDING THEIR OWN
> MONEY.
> 
-- 

Well, as one who writes software in the "commercial world", and increasingly
buys COTS software for "commercial" use, I beg to differ.  Life cycle costs are
given much lip service in the commercial world, but I rarely see decisions made
that truly reflect consideration of much beyond the initial costs.  Most
managers in "civilian" companies know that they will be long gone on to their
next job before the maintenance costs of their decisions become visible. 
Managers are rewarded (as are engineers, BTW) most heavily for what they've
done in the past YEAR; not what they did 5, 10, or more years previous.  Many
companies don't even keep performance records on their employees for more than
5 years!  What sticks in the collective corporate memory are the events such
as on time, or under budget project completions, or introduction of some snazzy
new technology ... not the fact that a system has very low life cycle costs. 
I'll bet that most companies don't even TRACK software maintenance costs.  I've
run across only one company that does and I've talked with lots of folks from
lots of companies over the past 12 years.

It is my opinion though that the increased use of COTS software may finally
change this.  I know that my company is starting to pay attention to software
maintenance costs now that it gets yearly bills from vendors for annual
maintenance fees!  Many used to think that our internal software developers and
maintainers were inefficient and costly, but nobody ever tracked the costs so
it was largely urban (corporate?) legend.  Now that the costs are painfully
visible and ramping up fast as we buy more and more COTS products, people are
starting to take notice.  I've even heard some managers begin to talk about the
costs of COTS actually being higher on a lifecycle basis...

--
Matthew S. Whiting, P.E.   | PP-ASEL-IA | All opinions expressed herein are
Corning Incorporated       |            | strictly personal.
whiting_ms@corning.com     |            |