correctness and comprehensiveness of compilers [was: Re: C compiletime]

[Scott Robert Ladd]

This will be my last note on the subject, but I felt some sort of 
response was required for the list. I will not reply to any further 
responses on list, although I'll be more than happy to discuss these 
issues in private e-mail.

Joe Buck wrote:
> I know that Joe Wolf and others on this thread mean well (and that Joe was
> answering claims that there was no bug report mechanism), but it's gotten
> well into out-of-bounds territory for the gcc list.  I regret that no one
> said anything earlier.

Joe Wolf provided non-partisan information for someone confused about 
reporting bugs for Intel's compilers; I'm sure GCC appreciates that some 
of us have worked in other forums to dispel myths and misinformation 
about the GNU compilers.

Since it was not an advertisement, I see Joe's message as an honest 
attempt at clarification; given that several people here also use 
Intel's compilers, I see nothing wrong with a singular, polite provision 
of non-commercial information.

I will be one of the first to scream bloody murder if a commercial 
company used a free software mailing list for selling product -- but 
that is not what happened in this case.

> Unfortunately, the fact that we archive the gcc list means that the FSF
> web site will now link to support pages for the competition; I'm afraid
> that RMS will be even less amused than Joe's management would be if an
> Intel support forum had links to AMD.

While I agree that the discussion has diverged, I also think it reflects 
poorly on GCC when we archive inaccurate statements without reasonable 
and polite rebuttal.

Buying Intel's compilers has absolutely no impact, financial or 
otherwise, on GCC; therefore, it is not a matter of "competition" so 
much as focus. This is the GCC list, for discussion of GCC-related 
issues -- and, at times, how other compilers "work" is germaine to GCC. 
Discussion of such issues, in moderation, has value.

GCC is often frustrated by false ideas about the quality of "free" 
software and the GNU compilers; I wrote my compiler comparison with the 
explicit purpose of showing the GCC is, indeed, a professional tool. If 
we expect the courtesy of fair representation for GCC, perhaps we should 
extend it to others on a reasonable scale. We're engaged in a balancing 
act, between focus and fairness. People's mileage will, I'm sure, vary.

-- 
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Professional programming for science and engineering;
Interesting and unusual bits of very free code.