NetNews Usenet Archive 1992 #30/NN_1992_30.iso/spool/comp/lang/ada/3665

Text File | 1992-12-14 | 5.4 KB | 103 lines

Newsgroups: comp.lang.ada
Path: sparky!uunet!walter!att-out!pacbell.com!sgiblab!zaphod.mps.ohio-state.edu!cs.utexas.edu!csc.ti.com!tilde.csc.ti.com!mksol!mccall
From: [e-mail] (fred j mccall 575-3539)
Subject: Re: Open Systems closed to Ada?
Message-ID: <[e-mail]>
Organization: Texas Instruments Inc
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Date: Mon, 14 Dec 1992 17:28:41 GMT
Lines: 92

In <[e-mail]> [e-mail] (Michael Feldman) writes:

>In article <[e-mail]> [e-mail] (fred j mccall 575-3539) writes:
>>
>>>IMHO, DoD is doing the right thing by opting for a strong and enforceable
>>>standard. Shooting at a moving target is no fun. I don't often defend
>>>Defense, but dammit, I think they are right on target here. Contractors
>>>who want to experiment with a moving state of the art with MY tax money
>>>are just outta luck.
>>
>>Can we actually document any savings, or are we still working on
>>guesswork and theory in this area?  Anyway, wouldn't it make more
>>sense to freeze on a per-contract basis rather than to Mandate frozen
>>technology for a dozen years at a time?
>>
>My experience is that the projects are sufficiently complex that the numbers
>can be cooked to show savings or lack of savings according to the motive
>of the writer. Lots of people are trying to show that Ada saves money,
>and because they want these numbers to come out right, they do. Others
>want just as desperately to show no savings, and I'm sure the numbers
>would come out their way too. Maybe I'm too cynical, but I don't put
>all that much faith in these arguments, one way or the other.

Probably quite true.  I just think that appeals to all the savings
that Ada provides when we don't seem to have a lot of proof over the
long term that it's providing such savings are just slightly less than
convincing.

>The Mandate give a starting date (June 1, 1991), but no end date. What
>makes you think that DoD will NECESSARILY "freeze technology" for the
>next eleven years? If the technology changes such as to make it
>manifestly obvious that the Mandate should go away, I'm sure that
>it can be made to do so. Lots of clever people at DoD and on the Hill.

Because I expect that once it is in effect, it will remain in effect.
Bureaucratic inertia, if nothing else, which means the technology can
only change when the language definition is changed; about every 10-12
years.

>In the meantime, in effect the decision IS being made project by project.
>How else? Each project is contracted for individually. If the manager
>can make the case that Ada is manifestly not cost-effective, (s)he
>can get that waiver or exception. The practical effect of the mandate
>is to make the nay-sayers defend their case, instead of making the yea-sayers
>defend _their_ case. I see nothing wrong with that.

I think it makes more sense to simply do the thing in the best
language for the job, and make EVERYBODY defend their case.  The idea
that there's a 'default' language that you have to make a case AGAINST
strikes me as a problem.

>The _psychological_ effect of the 1990 Mandate is why I think it was a dumb
>idea. The rest of the world says "how good could Ada be if DoD has to
>force it on its own contractors?" and shuts off discussion, figuring
>it's not worth investigating further.

>I happen to believe that the jury is still out on cost-effectiveness _on
>a single project_ - that one can cook the numbers any way one likes,
>because the differences are not well-understood and probably at the margins
>of the project, and also because the language/OS field is very fluid
>these days and compiler version K running on system version P could
>give VERY different numbers from K+1 and P+1. It comes down to a
>question of will. People who wish to operate in good faith and give
>Ada their best shot without endlessly nay-saying will, in my idealistic
>view of the world, manage to come up with a good and cost-effective
>project. Those who wish to keep sabotaging an effort will, I'm sure,
>find endlessly creative ways to do so.

Quite true.  However, the fact that something can be made into a "good
and cost-effective project" using Ada still doesn't say that Ada was
the best choice.  It's almost a "well, you may as well lay back and
enjoy it" attitude.  You can't do anything about the weather, so you
may as well just deal with it.  The same kind of thinking gets applied
to the Mandate, I suspect.

>The _global_ cost-effectiveness of doing a large number of
>projects with a small number of languages seems obvious to me.
>Frankly, I really wish people would settle down, accept the Mandate
>as given for the time being, shut up and get the work done.
>That is, IMHO, the best way to spend my tax dollars.

I think we'd all be better served by OPEN discussion, rather than a
Mandate.  The "soldier, shut up and soldier" argument is one with
which I'm quite familiar; I'm just not convinced that it's the best
approach to producing the best products for the tax money we spend
doing 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
------------------------------------------------------------------------------
[e-mail] - I don't speak for others and they don't speak for me.