GNAT vs other compilers

GNAT vs other compilers

[Patrice Saintonge]

Hello !

I'm currently trying to convince my boss that GNU tools are worth being 
used, compared to other often expensive proprietary tools.

This is why i'm looking for comparative evaluations of the GNAT 
compiler vs the other compilers in the hood.

Many thanks in advance.

Patrice Saintonge.
cs-defense-ceo@calva.net

Re: GNAT vs other compilers

[Dave Wood]

Patrice Saintonge wrote:
> 
> Hello !
> 
> I'm currently trying to convince my boss that GNU tools are worth being
> used, compared to other often expensive proprietary tools.

I'm all in favor of GNAT - in spreading Ada 95, it provides a
good service for the community.  But, I wouldn't be too hasty
to assume that all the Ada products from the commercial vendors
are expensive.  In fact, you might find them quite cost 
effective, particularly when issues like environment, speed,
support, and services are factored into the equation.

It is my opinion that strong, competitve commercial vendors
and usable freeware are both crucial ingredients to a vibrant
Ada community.  They serve different but complementary 
purposes.

> This is why i'm looking for comparative evaluations of the GNAT
> compiler vs the other compilers in the hood.

We're pretty comfortable with this, but you might find
substantive comparative evals hard to come by for at least
several more months.  What you're more likely to get now
are anecdotal accounts.

-- Dave Wood
-- Product Manager, ObjectAda for Windows
-- http://www.thomsoft.com

Re: GNAT vs other compilers

[Robert Dewar]

Dave Wood says

"I'm all in favor of GNAT - in spreading Ada 95, it provides a
good service for the community.  But, I wouldn't be too hasty
to assume that all the Ada products from the commercial vendors
are expensive.  In fact, you might find them quite cost
effective, particularly when issues like environment, speed,
support, and services are factored into the equation."

First of all, Dave does not mean commercial here, he means proprietary.
GNAT is definitely a commercial product, and ACT is a commercial vendor,
it is just that GNAT is not a proprietary product!

I agree with Dave that you should factor in all these issues into the
equation, particularly support, which can be critical when you are
using new technologies. We are very happy to compete in all these
areas, and we (and many of our customers) see GNAT as much more than
just of value for spreading Ada 95. For example, GNAT is still the
only high quality Ada technology in sight on many machines, and is
still the only full language technology (all other Ada 95 compilers
so far do not address the full set of special needs annexes).

As always, choosing a software technology requires work on the part of
users, and it is rare that you can say absolutely "technology X" is
superior to "technology y", for any x,y. Rather you end up deciding
that in *this* particular situation one or the other makes better
sense.

Incidentally, Dave is also quite right that you should not assume that
all the Ada products from the proprietary vendors are expensive, or that
GNAT is necessarily inexpensive. Sometimes a GNAT-based solution with
appropriate levels of support may be more expensive than some proprietary
solution. Whether it is worth paying this higher amount depends as always
on whether the added capability is sufficiently valuable.

Robert Dewar
President
Ada Core Technologies

P.S. This is pushing the limits of what is appropriate to this newsgroup,
but, I think, no more than Dave is pushing them :-)

Re: GNAT vs other compilers

[Robert A Duff]

In article <dewar.839546529@schonberg>, Robert Dewar <dewar@cs.nyu.edu> wrote:
>P.S. This is pushing the limits of what is appropriate to this newsgroup,
>but, I think, no more than Dave is pushing them :-)

I can't speak for everyone, obviously, but *I* don't see anything wrong
with this discussion.  Lots of people on this group want to know stuff
about various vendor's Ada products, so why can't the vendors tell them?
It's not like you guys are hawking memory chips or something -- now
*that* would be inappropriate.  (Or, say, get-rich-quick pyramid
schemes!)

- Bob

Re: GNAT vs other compilers

[Dave Wood]

Robert Dewar wrote:
> [snip]

> 
> P.S. This is pushing the limits of what is appropriate to this newsgroup,
> but, I think, no more than Dave is pushing them :-)

Yeah, but you know, sometimes I like to shake things
up on comp.lang.gnat...

:)

-- Dave Wood
-- Product Manager, ObjectAda for Windows
-- http://www.thomsoft.com