Re: DARPA admits Ada gets in the way

Re: DARPA admits Ada gets in the way

[Michael Feldman]

In article <25222@oasys.dt.navy.mil> tdsmith@oasys.dt.navy.mil (Timothy Smith) 
writes:
>
>    DARPA (Defense Advanced Research Project Agency) directs basic RESEARCH
>projects. In general, DARPA funds high risk/high payoff projects.  THey
>do not DEVELOP hardware.  Once the concept is proven DARPA transitions
>it to some other part of the DOD for development into a military system.
>    There is no contradiction betwee DARPA wanting to keep it's options
>open and the rest of DOD wanting to have systems written in Ada.

In a democracy I see no problem with this. It's not clear to me why DARPA
should make a pronouncement on the mandate. Washington is like that: a
guy (who won't speak for attribution) expresses his own opinion, claiming
he speaks for his sub-agency. (If he were really speaking for DARPA, why
won't he let his name be used?) Then the rest of us think that the sub-agency
(DARPA, in this case), speaks for the whole agency (DoD, in this case).
It's all part of the Washington power game. Greg's assertions notwithstanding,
little of this has much to do with Ada. 

>>Like Mike Feldman noticed quite recently, and as I have posted based on
>>analyzing all of the entries in my government software database, Ada is
>>not being used, called for, and hired for.
>
[stuff deleted]

Oh, all right. Since you paraphrased me, I'll chime in again. Ada fan
though I am, I am no fan of the mandate. It's diversionary and
ultimately foolish. DoD can make Ada stick to the projects for
which it needs Ada, without Congress getting its fingers in the pie.

I have said since the mandate was cranked into the law, in the middle
of the night in 1990 just before the last election, that the mandate
would allow Ada vendors to be lazy - because their market was protected -
and cause many people to spend their days and nights to keep the mandate
in the law, instead of working on substantive matters. It's clear to me
now that my understanding of the Washington game was reasonably good,
even if I am a pointy-headed academic. Watch them scurry to keep Congress
from deleting the mandate. Whatever happens will happen in the middle
of the night when nobody is looking. That's life in the fall of even-
numbered years.

Crazy idealistic eggheaded professor that I am, I actually believe that
Ada can stand on its own, without a legal mandate. Closed-minded people
will resist it, as such people always resist change. Open-minded people
will see that Ada offers them something. 

The mandate is a bit like the auto industry in microcosm. Lee Iacocca
demands protection from Toyota even as he imports millions of Mitsubishis.
The Ada vendors want the mandate to protect them from the yellow peril
of C++, even as they diversify into C++ because they can't figure out
how to make a big profit from Ada.

If the vendors of compilers and tools will get to work building quality 
products, and sell them at a fair price (not the order-of-magnitude price 
difference found, for example, in Rich Pattis' list of RISC-6000 compiler 
prices), the non-DoD part of the world will treat Ada with respect. If 
the attitude remains "soak the DoD contractors and ignore the rest of 
the world," no DoD mandate will save Ada from failure in the larger world.
This has NOTHING to do with technical issues. It has to do with business
decisions. Technically, Ada can compete, with or without type extension.
Will the Ada business ever give Ada the chance it deserves?

Mike Feldman

PS - someone characterized my last few postings as my being in a new
"attack mode." New? I sorta thought I always _was_ in attack mode...

Re: DARPA admits Ada gets in the way

[Randy Jordan]

In article <1992Sep18.004238.2689@mathsoft.com>, aha@mathsoft.com (Greg Aharoni
an) writes:
|>Path: pnl-oracle!ogicse!usenet.coe.montana.edu!rpi!think.com!mathsoft!aha
|>From: aha@mathsoft.com (Greg Aharonian)
|>Newsgroups: comp.lang.ada
|>Subject: DARPA admits Ada gets in the way
|>Message-ID: <1992Sep18.004238.2689@mathsoft.com>
|>Date: Thu, 17 Sep 92 17:42:38 GMT+8:00
|>Article-I.D.: mathsoft.1992Sep18.004238.2689
|>Sender: usenet@mathsoft.com (Usenet News Administration)
|>Organization: MathSoft, Inc.
|>Lines: 82
|>Nntp-Posting-Host: fresnel
|>
|>
|>
|>That it has taken this long for people in DARPA finally to say something
|>tells alot about the supression of dissent inside the DoD.  After all,
|>when Ada first was really mandated, DARPA knew (or should have known)
|>what languages its contractors were using and how likely any of these
|>people would be to switch to Ada (trying converting some of these large
|>Lisp systems to Ada).  Given existing language transitions in general

The LAW does NOT require conversion of existing programs written in ANY
language .... only of NEW developement to be done in Ada (Thank GOD).

|>(which are low, and mostly in technical community Fortran --> C/C++)

NO one in their right mind converts FORTRAN to C anything. I delivered
a FORTRAN (2167) program to the Army several years ago (Before Ada) even THEN
C language was UNacceptable.

|>and with a mandate from the DoD, it should been obvious that not many of
|>DARPA's contractors would switch to Ada.  Obviously in the years since

Almost ALL DoD contractors use Ada now.

|>they haven't, and DARPA's current policy downgrading Ada reflects this.
|>
|>    And if these people are stating these things publicly, imagine the rest
|>of the dissent that is still being suppressed within the DoD with regards
|>to Ada.  If you read the program to the next Tri-Ada Conference, you will
|>not see anything that reflects this dissension at all.
|>
|>    What is so pathetic about the defense community botching of Ada is that
|>with less money and attention, it's VHDL language is catching on like crazy
|>in the circuit design world.  VHDL is being accepted in that community like
|>Ada advocates can only dream about.  What's funny is that the the languages
|>(Ada, VHDL) are similar enough syntactically that many of the defenses of
|>Ada's non-acceptance, in light of VHDL, are false.
|>

VHDL... Never heard of it.

|>
|>		WHEN IT COMES TO SPENDING THEIR OWN MONEY, PEOPLE
|>		DON'T SPEND IT ON ADA OR REUSABLE SOFTWARE.
|>

No, they spend it on Debugging and fixing C bugs, tring to read and maintain C
source and retrain jr programmers to updata unstructured C systems.

|>Greg Aharonian
|>Source Translation & Optimization
|>
--

RJ

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
	To Thyne Own Self be true...
	                             myne opinions are.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Re: DARPA admits Ada gets in the way

[Carl Kaun]

I am a contractor who has worked on a number of DARPA projects.  My
comments reflect this experience, but otherwise are my own opinions,
(and especially not those of my employer).

aha@mathsoft.com (Greg Aharonian) writes:

  ... <deleted material>

>>  That it has taken this long for people in DARPA finally to say something
>>  tells alot about the supression of dissent inside the DoD.  After all,

Suppression of dissent?  My experience is that at DARPA it has been
basically a non-issue, because (at least in the areas of DARPA where I
have been involved) DARPA has not been forced to use Ada.  Even where
there was encouragement to do so, it was relatively easy to write
justifications avoiding its use.

>>  And if these people are stating these things publicly, imagine the rest
>>  of the dissent that is still being suppressed within the DoD with regards

Again, "these people" (referring to DARPA) discuss things, publicly or
not, when it becomes an issue to them, and when not otherwise
proscribed by security needs. In a 25 year career in DoD and working
as a DoD contractor, I have never detected (IMHO)
any significant "suppression" of discussion on non-security matters. I
object to the post-hoc arguments whose premise (unsupported by
evidence) is that if you do not (or have not) hear(d) others express
opinions matching your own, then it is because those who hold such
opinions are (or have been) suppressed.  Such intellectual flabbiness
leads shortly to unreality or dishonesty.

>>  to Ada.  If you read the program to the next Tri-Ada Conference, you will
not see anything that reflects this dissension at all.

Post-hoc again. The content probably reflects the selections of the
organizers, who may wish to promote the use of Ada, and not
suppression as the statement seems to imply.

>> What is so pathetic about the defense community botching of Ada is that
>> with less money and attention, it's VHDL language is catching on like crazy
>>   ... <material deleted>

Is it botched?  Why is it pathetic that a good language with features
similar to Ada is catching on?  ... <many other questions>.

----- 
Bottom line: I don't believe DoD is suppressing discussion, I would
hate to have others think that it is unless there is other evidence
for it.  I do not address the issue of whether DoD policy concerning
Ada use has been premature, excessive, or heavy-handed.  That is a
different issue than the "suppressing discussion" issue.

---------------------------------------------------------------------
      Carl F. Kaun       ckaun@ads.com        415/960-7420
---------------------------------------------------------------------

Re: DARPA admits Ada gets in the way

[Mike Black]

In article <1992Sep18.184122.8984@ads.com> ckaun@ADS.COM (Carl Kaun) writes:
>Bottom line: I don't believe DoD is suppressing discussion, I would
>hate to have others think that it is unless there is other evidence
>for it.  I do not address the issue of whether DoD policy concerning
>Ada use has been premature, excessive, or heavy-handed.  That is a
>different issue than the "suppressing discussion" issue.
>
>---------------------------------------------------------------------
>      Carl F. Kaun       ckaun@ads.com        415/960-7420
>---------------------------------------------------------------------

After 20 years in the USAF I can attest to the fact that discussion
at the WORKER level is not suppressed.  However, at the management
level and in the public forum it most definitely is suppressed.
Any comments an individual makes in the public forum are cleared
and (most assuredly) must be deamed politically correct before
released.  

I also understand DARPA's position.  I also worked in research and
development for the last 11 years and appreciate their thought
that they shouldn't be restricted to ANYTHING!!  One doesn't
advanance technology using yesterday's tools.  I went thru a 
period when the Ada manadate was first created when our software
shop said EVERYTHING was going to be developed in Ada with full
mil standards applied (they never heard of tailoring specifications).
They didn't seem to understand about leveraging R&D dollars by
allowing contractor-format-acceptable performance clauses.  In our
R&D work we didn't care what language/hardware/whatever the contractor
wanted to use as long they proposed a technically sound program.

P.S. DARPA is not the only DOD organization that does research.
Basic research and R&D dollars are spread throughout DOD at many
of the laboratories.


-- 
-----------------------------------------------------------------------------
 black@beno.CSS.GOV             land line: 407-676-2923  | I want a computer
 real home: Melbourne, FL       home line: 407-242-8619  | that does it all!
 CSI Inc, Computer Software Innovations, Palm Bay, FL

Re: DARPA admits Ada gets in the way

[agate!spool.mu.edu!caen!zaphod.mps.ohio-state.edu!cs.utexas.edu!csc.ti.co]

In <1992Sep18.163435.9956@oracle.pnl.gov> jordan@warped.pnl.gov (Randy Jordan) 
writes:

Well, it looks like we've found another language bigot who wants to
play 'flamewar'.

>In article <1992Sep18.004238.2689@mathsoft.com>, aha@mathsoft.com (Greg Aharon
ian) writes:
>|>Path: pnl-oracle!ogicse!usenet.coe.montana.edu!rpi!think.com!mathsoft!aha
>|>From: aha@mathsoft.com (Greg Aharonian)
>|>Newsgroups: comp.lang.ada
>|>Subject: DARPA admits Ada gets in the way
>|>Message-ID: <1992Sep18.004238.2689@mathsoft.com>
>|>Date: Thu, 17 Sep 92 17:42:38 GMT+8:00
>|>Article-I.D.: mathsoft.1992Sep18.004238.2689
>|>Sender: usenet@mathsoft.com (Usenet News Administration)
>|>Organization: MathSoft, Inc.
>|>Lines: 82
>|>Nntp-Posting-Host: fresnel
>|>
>|>
>|>
>|>That it has taken this long for people in DARPA finally to say something
>|>tells alot about the supression of dissent inside the DoD.  After all,
>|>when Ada first was really mandated, DARPA knew (or should have known)
>|>what languages its contractors were using and how likely any of these
>|>people would be to switch to Ada (trying converting some of these large
>|>Lisp systems to Ada).  Given existing language transitions in general

>The LAW does NOT require conversion of existing programs written in ANY
>language .... only of NEW developement to be done in Ada (Thank GOD).

And what does it say about little details like changes or updates to
existing code, legacy code, etc., etc.?

>|>(which are low, and mostly in technical community Fortran --> C/C++)

>NO one in their right mind converts FORTRAN to C anything. I delivered
>a FORTRAN (2167) program to the Army several years ago (Before Ada) even THEN
>C language was UNacceptable.

Must be a lot of people out there who aren't in their right minds,
then.  Personally, I think the idea of 'language conversion' is rather
silly, anyway, since code run through converters (or even if
'converted' by hand) is typically much harder to maintain.  Better to
redesign it and make use of the features of the new language.

>|>and with a mandate from the DoD, it should been obvious that not many of
>|>DARPA's contractors would switch to Ada.  Obviously in the years since

>Almost ALL DoD contractors use Ada now.

Yes, with a gun to their heads.  They also use a lot of other things. 

>|>they haven't, and DARPA's current policy downgrading Ada reflects this.
>|>
>|>    And if these people are stating these things publicly, imagine the rest
>|>of the dissent that is still being suppressed within the DoD with regards
>|>to Ada.  If you read the program to the next Tri-Ada Conference, you will
>|>not see anything that reflects this dissension at all.
>|>
>|>    What is so pathetic about the defense community botching of Ada is that
>|>with less money and attention, it's VHDL language is catching on like crazy
>|>in the circuit design world.  VHDL is being accepted in that community like
>|>Ada advocates can only dream about.  What's funny is that the the languages
>|>(Ada, VHDL) are similar enough syntactically that many of the defenses of
>|>Ada's non-acceptance, in light of VHDL, are false.
>|>

>VHDL... Never heard of it.

Don't do anything having to do with hardware, do you?

>|>
>|>		WHEN IT COMES TO SPENDING THEIR OWN MONEY, PEOPLE
>|>		DON'T SPEND IT ON ADA OR REUSABLE SOFTWARE.
>|>

>No, they spend it on Debugging and fixing C bugs, tring to read and maintain C
>source and retrain jr programmers to updata unstructured C systems.

In a (polite) word, hogwash!  Didn't they ever teach you that you
should pick the proper language and tools for each specific job.  This
is what I see as the biggest problem with the 'Ada Mandate'.  It
requires you to use a hammer, even if your problem doesn't happen to
look like a nail.  As Dennis Ritchie once said (referring to something
else entirely), "If you want PL/I, you know where to find it."

-- 
"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: DARPA admits Ada gets in the way

[Frank Manning]

In article <1992Sep21.170317.28618@mksol.dseg.ti.com> mccall@mksol.dseg.ti.com
(fred j mccall 575-3539) writes:

>Didn't they ever teach you that you should pick the proper language and
>tools for each specific job.  This is what I see as the biggest problem
>with the 'Ada Mandate'.

How come everybody uses English when they mention this in Usenet? Aren't
there other languages better suited for complaining?

-- Frank Manning                                -- Think global. Act loco.
-- College of Engineering and Mines             --
-- Civil Engineering 100, University of Arizona --                 -Zippy
-- Tucson, AZ  85721    frank@evax2.arizona.edu --

Re: DARPA admits Ada gets in the way

[dog.ee.lbl.gov!hellgate.utah.edu!cs.utexas.edu!csc.ti.com!tilde.csc.ti.co]

In <1992Sep21.161408.3733@arizona.edu> frank@evax2.engr.arizona.edu (Frank Mann
ing) writes:

>In article <1992Sep21.170317.28618@mksol.dseg.ti.com> mccall@mksol.dseg.ti.com
>(fred j mccall 575-3539) writes:

>>Didn't they ever teach you that you should pick the proper language and
>>tools for each specific job.  This is what I see as the biggest problem
>>with the 'Ada Mandate'.

>How come everybody uses English when they mention this in Usenet? Aren't
>there other languages better suited for complaining?

Yes, but not on Usenet/Internet -- which is primarily native English
speakers and for which the 'standard' language is English.

[Complaining isn't writing software.  Back to the mines until you
learn something, Frank.  :-)]

-- 
"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.