Re: Ada's (in)visibility in the engineering community

[sun-barr!cs.utexas.edu!csc.ti.com!tilde.csc.ti.com!mksol!mccall@ames.arc. nasa.gov (fred j mccall 575-3539)]

In <OBRY.92Sep14094032@cheesesteak.flash.bellcore.com> obry@flash.bellcore.com 
(Pascal Obry) writes:


>I like very much Ada. And I tried to convince in 2 differents researh center t
o
>use it, but I failled ...

>Why ? 

>because they don't want to learn a new language, 

Bad reason.  Professionals should be both prepared and able to pick up
a new language without incredible difficulty and without huge learning
curve times (although there will, of course, be some impact on
productivity for a while).

>because they don't or can't afford an Ada compiler,

Good reason.  Perhaps it isn't seen so by someone who has never worked
in a budget-constrained atmosphere, but this is perhaps one of the
BEST reasons.

>because they don't see any reason to buy an (expensive) Ada compiler whereas
>they have a free C compiler in any of their computer, here GNAT will do
>something very good.

Good reason, and a good statement with regard to GNAT.  I think that
this may be the best thing that ever happened to Ada.  Given access to
an inexpensive (as in free) compiler, perhaps there will be more
interest in 'free Ada sources', as we see in the case of C on the net. 

>because they think Ada is a too complex language and too big. Here they make
>the big mistake to mix up language and compiler. My point is that it is not
>because the compiler is hard to make that the language is difficult. And in
>this case to make a compiler is difficult because the language is *powerfull*
>not because it is complex to use ..

It is actually something of both (powerful AND complex to use).  This
is not necessarily a bad thing, but 'large' languages like Ada have
significant barriers to adoption at least in part because of their
sheer size and the increased time required to really know them well.

>because they don't like to think a lot about the conception of their projects.
>they want to make it without conception (OO or others ...) and C for this is
>very good because you can do every thing you want, there is alway (a bad) way
>to reach your goal.

Spoken like someone who has never done a major OO project.  If you
think you can just sit down and write a major application in an OO
language "without conception", you are greatly mistaken.

>because they didn't know Ada at all so they chose C++ (yes, they think it's th
e
>good old C plus something so it should be good too ... bad mistake no!)

It is if they are assuming that they can just sit down and 'hack out'
an application in C++ "without conception".  Of course, one of the
advantages of C++ is that all the people who know C can use that
subset of the language, and you can gradually increase in-house
expertise in disciplines (e.g. OO design) that are required to use it
well. 

>------------------------

>Also all these guys sometime use a kind of *meta* language to describe an 
>algorithm ... And you know what : this meta language is something very
>close from Ada ...

Well, mine tends to look more like C, but that's just a product of
past environment.

>So I tried and I failled ... But I can't even understand why !!
>They have no good reasons to don't use Ada, they have no good reason to choose
>C++ but they chose it ...

No, they had good reasons for both.  Ada was too expensive and they
couldn't do anything with it until they learned it.  C++ was a lot
cheaper, and they could at least be somewhat productive with it
immediately, if they already knew C.  

[If they didn't already know C, I'm not sure what the point would be
of choosing one over the other, other than compiler cost.]

>-------------------------

>I learn Ada by myself. I like very much to learn and compare different 
>languages, today I know C, C++, Eiffel, Ada, (plus a lot of other languages
>like Basic, Fortran, Cobol). But after all, I definitly like very much Ada ...

>Like Edmond Schonberg said in is comparison Ada 9x and C++ :

>" Although the C++ community would never state it so baldly, it appears clear
>to us that C++ is to some extent a reaction to Ada. By extending C with some
>of the best ideas of Ada, C++ did in some measure catch up to Ada "

Well, it would seem that Mr. Schonberg was incorrect, since Bjarne
Stroustrup has stated publicly numerous times that Ada was the
inspiration for some of the features that he put in C++.  I'm not sure
why this is supposed to be such a telling point.  Isn't that one of
the IDEAS behind language development; to take the best features of
other languages and try to make them work together?  Other parts of
C++ came from things like Smalltalk.  I'm not sure about this "catch
up to Ada" statement, though.  Just doesn't seem to track to me.

-- 
"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.