Re: Open Systems closed to Ada?

[mcsun!uknet!yorkohm!minster!mjl-b@uunet.uu.net]

In article <1992Dec4.163301.1791@mksol.dseg.ti.com> mccall@mksol.dseg.ti.com (f
red j mccall 575-3539) writes:
>In <1992Dec4.074834.24047@gvl.unisys.com> lonjers@prc.unisys.com (Jim Lonjers)
 writes:
>
>>It may come as a surprise to many folks that yes, even when people have a
>>choice, many choose Ada, and it is not limited to the more highly
>>publicized cases of Boeing (777) and Motorola (the Cellular Telephone
>>system).
>
>I think the question is not whether many do or not, but relatively how
>many when compared to other choices, and what are the jobs it is
>chosen for.

No, I think the quesion is _why_ did they made their choices.

My reasoning: Suppose many people may make a bad choice. Does this make the
choice any better?

>>[Just a few words of support for Dave Emery who seems to be under attack
>> for his views -- I too have observed the same pro-C militant attitudes.
>> Shall I say it?  Yes, it seems to be pretty much out of ignorance.  Most
>> of those who are militantly in favor or against any particular thing
>> are not all that well versed in the alternatives.
>
>And here we see the militant Ada attitude; if you don't agree with
>them that Ada is superior to everything, it must be because you're
>ignorant. 

This is an extreme response to Jim's statement of his experience.

>> About the only thing that C has going for it is that it has a large
>> trained base of programmers (how well trained, I do not know).  This
>> is because most of the schools now teach C as part of the curriculum,
>> or when a curriculum does not teach a particular language, C is
>> encouraged.  It is also easier to write little C programs than it is to
>> write little Ada programs.  
>
>The only thing?  Gee, that sort of begs the question of how it got
>started, then, doesn't it?  I mean, why did all those schools start
>teaching C and how did it get so popular?

It would indeed be interesting to find out why, as this sort of question is
the crux of why Ada is not more widespread. Again, you seem to be equating
"popular" with "good". My favourite refutation analogue follows:

From: rlk@VisiCom.COM
Date: 22 Apr 92 17:23:18 GMT

I've always wondered at the argument of of "programming language X must be
great because so many programmers write in it".  Hurrumph.  A similar
argument can be made for eating at McDonalds -- it must be the world's best
food, because more people eat at McDonalds than any other restaraunt.

But for mission-critical dining experiences, such as a two-year anniversary,
only a fool would go to McDonalds.  The added expense of a four-star
restaraunt is worth it, even though it may take longer.

	.Bob.

p.s. I hope McDonalds' lawyers don't read comp.lang.ada! ;-)

>> It is interesting that C++ has invented an inter-language
>> calling mechanism.  
>
>Why 'interesting'?  Any language that does name mangling (as C++ does)
>is going to need an inter-language calling mechanism to tell the
>compiler NOT to mangle certain identifiers so that the linker can find
>them, particularly if backward compatibility and the ability to link
>to modules of a language that does not do such mangling (C) is one of
>the design goals.  This was hardly 'borrowed' from Ada.

You've missed Jim's point about inter-operability of langauges. Yours is an
implementation issue and very specific to Unix-style linkers. With a simple
language like C, you only need a simple linker. But for more advanced
langauges, you're going to need a more advanced linker, or more "mangling".

>> Also that the C community is now re-inventing
>> tasking (threads), but doing it differently, but not better than,
>> Ada.  
>
>I wasn't aware that Ada 'invented' threads -- I don't believe that
>putting them into C++ was borrowed from Ada, either, but rather driven
>by the POSIX 'lightweight process' concept.

I wasn't aware that anyone was claiming that Ada invented threads.

Ada was one of the first languages to include a tasking model. All the
tasking stuff was pretty new and untried back in the early 80s. The POSIX
lightweight threads effort was driven by broadly similar goals to the Ada83
language team, and it's no surprise that they ended up with a very similar
model.

>Fred.McCall@dseg.ti.com - I don't speak for others and they don't speak for me
.

Mat

| Mathew Lodge                      | "I don't care how many times they go    |
| mjl-b@minster.york.ac.uk          |  up-tiddly-up-up. They're still gits."  |
| Langwith College, Uni of York, UK |  -- Blackadder Goes Forth               |