On Thu, Oct 22, 1998 at 01:08:12PM -0000, [EMAIL PROTECTED] wrote:
> > They were, however, not designed by committee, which is why much of
> > the bulk of Ada is excess, useless, cumbersome baggage, and why much
> > of the bulk of C++ and Java is actually useful.
> 
> What do you call ANSI ?  Aren't They a committee ?  The ANSI committee
> is in control of the C++ Std.  Not Biojrn Strustrup (sp).

I recognize the difference between design and standardization.  I recognize
that the original design is quite elegant and don't hold him responsible
for what ANSI did to it (good or bad).

In other words, I recognize that "design by committee" almost always
leads to disaster, while "standardization by committee" is not quite as bad.
It's not great, but it's not sure doom.

> Well, the "validation" will help ensure that "Feature A" in 2 different
> compilers will generate code that does the same thing.
> I have seen 2 different "C" compilers generate massivly different code
> (when doing assembly code debugging).

So have I -- in fact, I've seen the *same* C compiler generate massively
different code depending on what debugging/optimization options were
supplied to it.  But I don't see this as a problem needing to be solved;
I don't really care if the code is different.  What I *do* care about is
whether or not it correctly performs its function.

> > There's a reason you don't see the people who have a choice (i.e. the
> > freeware community) rushing to port Ada and write things in Ada and create
> > applications in it, while you *do* see them doing the same in C++ and
> > Java (and C and Perl and tcl/tk).
> 
> I guess that you have never heard of gnat (the GNU Ada Translator).

I was aware that the GNU folks had a couple of bits and pieces of
Ada-related code -- I didn't know the names of any of them.  That
hardly constitutes a "rush".

In other words, if you look in any of the Usenet comp.sources.* archives,
or monitor the releases of software announced in any of the net's mailing
lists or web sites, you'll find that the lanaguages I listed above
account for +95% of everything that comes out.  Ada's presence is
below insignificant, and it's pretty clear that the reason for that
is that people who have a choice *don't* choose Ada.  (Nor have they ever:
even when all the hub-bub about Ada happened in the 80's, the people
with a choice paid little attention to it.)

Given that the 'net community has a long history of choosing the best
tools they can get their hands on -- to the point of using alpha releases
if that's all that's available -- why do you think that is?

> Would you want to be riding in the space shuttle with 100 % new code that may 
> or (more realisticly) may NOT have been fully debugged ?

I wouldn't want to ride in the space shuttle with new code *in any language*.
There's nothing magic about Ada (or C++ or C or anything else) that makes
it less likely to be buggy, despite the trumpeting of its endorsers.
I've seen elegant code written in Fortran and garbled junk written in C++.

What makes code less buggy, more maintainable, and all those other
Good Things is the use of skilled programmers given adequate time
to design, code and test their work -- the language is secondary
to this effort, although obviously some are more suitable for certain
tasks than others.  (Example: natural language processing in Fortran
would make your head explode.)

---Rsk
Rich Kulawiec
[EMAIL PROTECTED]
____________________________________________________________________
--------------------------------------------------------------------
 Join The Web Consultants Association :  Register on our web site Now
Web Consultants Web Site : http://just4u.com/webconsultants
If you lose the instructions All subscription/unsubscribing can be done
directly from our website for all our lists.
---------------------------------------------------------------------

Reply via email to