Re: Ada saved by gnat (was Re: Where's Aetech?)

[rajat@bugs.skipstone.com (Rajat Datta)]

In article <dewar.798335596@gnat>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>"I do wonder though, at why gcc and g++ have not put a damper on the C and C++
>markets despite far greater acceptance in the commercial computing world than
>Gnat.  Commercial chip companies pay to get gcc and g++ ported to their
>architecture and don't feel like they're paying homage to the late, great USSR."
>
>I don't particularly see why you would think that gcc/g++ should put a
>damper on the market place. They are just one of many competitive products.
>...stuff deleted

Well, that was my point (no doubt badly expressed).  gcc and g++ have not
caused the kind of disruption in the C community that the original poster
was asserting GNAT is doing for Ada.  And, far from complaining about the
availability of a free compiler, commercial companies seem to find availability
of gcc to be an advantage.  Certainly the contribution list to gcc spans not
just companies like Cygnus that live off free software, but also "pure"
commercial companies like IBM, Intel, amongst others.

The argument of capitalism vs. socialism made by the original poster reminds
me of the tactics of comparing to Hitler and Nazism.  A variant of condemnation
by guilt of association.  Since your actions don't agree with my notion of
capitalism you must be--horrors--a communist.

>In the case of Ada 95, I think by 1996 you will see some pretty lively
>competition. By then GNAT will have some nice environments wrapped around
>it on the PC and Mac, and the Intermetrics educational compiler will also
>appear, presumably at a low, accessible price. Educational users will
>hopefully be able to sit back and take advantage of this competition. I
>don't know if others will enter this low price arena, but two is not
>a bad value to stimulate some technical competition!
>

But what will be the advantage of Ada that will make users sit up and take
notice?  C was not automatically embraced by all the commercial companies
when it became available, despite all the claims that it was so much better
than using assembler.  C succeeded because Unix and Borland slowly converted
a lot of programmers, one at a time, to using it.  C++ is seen as a better C
and so there's less resistance to switching to it.

How do you convince people that Ada is better than C?  Or even, that it is so
much better that it is worth the pain of converting?  Even though it is very
fashionable to decry the current state of software and how awful the bugs are,
it is a huge and profitable industry and is very successful.  It will not be
enough just to say that things stink and Ada is so much better.  Why is Ada
so much better?  I've only recently started looking at Ada (only because of
the availability of a free compiler, so that speaks well of GNAT, I think),
and have not yet seen why it's such a big difference.  That's not to say that
C isn't full of flaws (it is), or that C++ is so great (I'm not convinced of
that), or that PL/I was awful.

It is not surprising that with Java, the Sun folks are also releasing HotJava
and a host of other applications.  They know that to make Java successful they
must make it a "neat" language to switch to, and convince people that since
there are so many goodies available, it must be possible to program in it.
Otherwise, only compiler weenies/language groupies pay attention.

rajat