Re: Current status of Ada?

[Gary Scott <garylscott@sbcglobal.net>]

adaworks@sbcglobal.net wrote:
> "Harald Korneliussen" <vintermann@gmail.com> wrote in message 
> news:1187850312.375316.57440@z24g2000prh.googlegroups.com...
> 
>>On Aug 22, 2:53 am, "Jeffrey R. Carter"
>><spam.jrcarter....@acm.nospam.org> wrote:
>>
>>
>>>And some projects using Ada commercially consider it a competitive
>>>advantage and keep it secret.
>>
>>Are you sure? It seems to me it's in the companies' best interests to
>>say whether they are using Ada, since Ada developers are tricky to get
>>hold of. Keeping it secret seems to me both difficult (what do you
>>write in the job ads?) and counterproductive.
>>
> 
> I once had a commercial client that required a non-disclosure agreement
> about their use of Ada because of competitive reasons.   In their view,
> their competitors would use this fact against them as a sales gimmick. The
> fear was that the competitors would ridicule them for "using a language
> that was not part of the mainstream and had been rejected by the Department
> of Defense."
> 
> Note that, when Emmett Paige wrote his famous memo abrogating the
> Ada mandate, the memo was widely interpreted as the equivalent of the
> DoD admitting that Ada was a mistake, and direct abandonment of its
> use for future DoD projects.   Although that was not the intent of the
> memo, that interpretation is now widespread both within and outside
> the DoD.
> 
> It is unfortunate that the memo was written in a way that left it open to
> Ada's enemies to misinterpret.   The damage done is widespread.  The
> educational institution where I teach once required Ada of its students.
> Now the language is almost non-existent except in a two-week portion
> of an eleven week class that I teach.  No one else in our computer science
> department gives it any credibility at all.
> 
> The real-time software projects are now being written in Java.   The funding
> for research will not support anything with the Ada language involved.   The
> newly-hired faculty members regard Ada as a quaint era of the past, not
> something to be taken seriously.
> 
> I have been an Ada advocate for about twenty years, but it is becoming clear
> that, without some miracle or absent someone in the DoD coming to their
> senses, the use of the language will continue to decline both in the commercial
> world and in the DoD.   When I was still consulting and teaching Ada, one of
> my major clients, a DoD contractor building one of our major weapons systems,
> switched from Ada to C++.   It was a massively stupid decision.   But the man
> who was previously in charge, who understood the value of Ada, retired.  His
> successor knew little about Ada and was a strong advocate of C++.   Without
> the mandate in place, he could blithely ignore the wisdom of using Ada and
> demand that everything be written in C++.
> 
> I asked the question, at the time, "What makes you think you can use a language
> such as C++ that is inherently error-prone, and expect a result that is 
> error-free."
> My credibility suffered from my resistance to C++.    The more I saw of, and
> continue to see of, C++, the more I realize how dangerous the language is and
> how wrong-headed it is to use C++ for military software systems, but
> my opinion carries no weight.   At the same time, in an effort to offset the
> known dangers of C++, many DoD organizations and their contractors have
> chosen Java.   This is also a dumb decision, but the new real-time features
> of Java make it more difficult to clarify the points that make Ada a better
> choice.
> 
> There is no single strong advocate for Ada at present.   There is no powerful
> corporate sponsor as there is for Java.  There is no major Ada project that
> is visible to the larger community of software developers.   The language is
> seen as "old-fashioned" and out-of-date by those who have graduated within
> that past ten to fifteen years.    It is an oddity.
> 
> The damage to Ada was the result of many factors.   The AJPO never quite got
> it right.   The DoD certainly never got it right.   The infighting between Ada 
> vendors
> never helped.   The fact that Ada compiler vendors charged outragesous prices 
> for
> their compilers helped to discourage commercial organizations from using Ada:
> COBOL, C, C, Pascal, were more affordable.   Most PC versions of Ada had
> less capability for building PC applications directly than BASIC.  With 
> exception
> of the Meridian Compiler, there were no good libraries for creating MS-DOS
> applications.   Even Meridian got it wrong by defining the data type for system
> address incorrectly.
> 
> With Ada 95, the designers and contributors to the design of the language did 
> get
> a lot of things right.  Ada finally became a language for the ordinary 
> programmer.
> The time was also right.   A lot of people renewed their interest in the 
> language.
> Then, grabbing defeat from the jaws of potential victory, the letter from Mr. 
> Paige
> muddled the entire decision-making process.    A delay of two or three years
> before writing that kind of letter might have made a difference.   Instead, the
> developer community ran as fast as it could to find other options.
> 
> JSF is being developed in C++.  A truly dumb decision.    Missile Defense Agency
> has completely abandoned Ada.
> 
> As noted in an earlier post, I made an inquiry some time ago about the current 
> state
> of Ada usage.   I am constrained from publishing the names of projects that are 
> using
> Ada, but I was suprised to find that there are still quite a few. 
> Unfortunately, such
> constraints do not help to promote the awareness that Ada is real and continues 
> to
> be a valuable tool for building software systems.  I promote it whenever I can 
> for
> my own students and have had thesis students do their M.S. thesis using Ada.  I
> make it clear in all of my software engineering classes that Ada continues to be 
> the
> most effective language when one needs to take an engineering view of the 
> software
> process.
> 
> But individual professors of computer science are of little importance in the 
> effort to
> improve the state of Ada utilization and awareness.  We need some kind of larger
> effort.  The Ada Resource Association (or whatever it is currently called) has 
> proven
> ineffectual.   The AdaIC web site, while in capable hands, has no pro-active 
> role.
> And the Ada compiler publishers seem to be ashamed to admit, broadcast, or
> let anyone know that they have Ada products.   When is the last time that 
> Rational
> had any information about its Ada compiler at a conference or trade-show?  When
> is the last time that any Ada compiler publisher had a booth at a trade-show? 
> When
> have we last seen any publicity about the value of Ada for some major project? 
> Where
> has anyone seen an Ada textbook for sale in a bookstore?  Even the 
> computer-centric
> bookstores have no books on Ada -- none.
> 
> As long as Ada remains invisible the number of projects will decline.   As long 
> as officials
> in the DoD believe that Ada is not supposed to be used for military projects 
> anymore
> (many believe just that), Ada will be in decline.
> 
> This is truly unfortunate.  Ada continues to be the best hope as a language for 
> software
> engineering.   In my view, it is still the best language for use in 
> safety-critical, mission-critical,
> and military software systems.   It offers a lot to commercial software 
> developers, as well.
> How we get that message out, now that there is no powerful sponsor and no 
> effective
> Ada consortium, I don't know.   At one time, I used to write a lot of articles 
> about the
> value of Ada for software magazines such as JOOP, HP Professional, Embedded 
> Systems
> Programming, and others.  That seemed to help a little.   I have yet to see 
> anyone publish
> an article about the Ada 2005 standard -- even in DoD publications.   It is as 
> if it never
> happened.
> 
> I no longer have the time to devote to Ada since my role has changed.    I am no 
> longer
> directly involved in Ada, though I continue to promote it whenever I can.   I 
> can still
> teach it in some of my classes, but I get the question from my colleagues, "Why 
> are
> we bothering with that old language?"     At present, I am the last hold-out for 
> keeping
> Ada in some small part of our curriculum.  When I am gone, Ada will also be 
> gone. Or
> as newer faculty members take over my courses, Ada will vanish entirely.
> 
> I wish I could outline an action plan instead of posting a tale of lament. 
> Perhaps someone
> from this forum can come up with a solution for improving the situation.   I 
> wonder if
> someone might write and publish some articles about the new standard and the 
> continuing
> viability of the language?    Maybe we can get someone in the DoD, someone with 
> a brain
> in their head who understands software, to reinvigorate and reinstate the 
> interest and
> committment to Ada.   I would hope so, but it is a faint hope at this point.
> 
> Richard Riehle
> 
> 
Correction, JSF does use a mixture of Ada and C++.

-- 

Gary Scott
mailto:garylscott@sbcglobal dot net

Fortran Library:  http://www.fortranlib.com

Support the Original G95 Project:  http://www.g95.org
-OR-
Support the GNU GFortran Project:  http://gcc.gnu.org/fortran/index.html

If you want to do the impossible, don't hire an expert because he knows 
it can't be done.

-- Henry Ford