I wrote:
> > One such standard is the Posix one. But just because something is
> > standardized doesn't necessarily make it particularly portable.
José Rui Faustino de Sousa wrote:
> AFAIK all the main platforms have at least some degree of POSIX support.
Of course, having some degree of support doesn't mean that a Fortran
programmer can actually write code and expect it to work (particularly
as on some platforms, the support may be provided by add-ons that
aren't installed on most machines).
> IMHO it was somewhat of a lost opportunity for X3J3 (now J3 right?)
> committee to drop the effort to standardize a Fortran 90 binding to
> POSIX (1003.19) that would have replaced the original POSIX.9 bindings
> to Fortran 77.
Well, it is right that X3J3 is now J3, but it isn't right that
POSIX 1003.19 was an X3J3 effort. There was liason between the
POSIX committee working on that and X3J3, but X3J3 was definitely
not the primary development body.
Indeed, (speaking based on my personal observations rather than as any
kind of official statement), the inputs from X3J3 weren't given very
much weight in the matter. I recall that either X3J3 (or several of
its members - I don't recall whether or not it was an official
committee position) had a lot to say about 1003.9 (the f77 binding),
and that that input was almost entirely ignored. One part of that
input was that the binding should have been to f90 in the first place.
There were lots of other things - many of them relating to specific
"issues" caused by f77's lack of features needed to support a good
job. Even in the context of f77, I think that many on the committee
thought some of the choices made were questionable at best.
If you want details on my opinion, read the rationale annex of
1003.9. Then invert almost everything in it. I don't think there
is a single thing in the rationale that I agree with...so it isn't
too surprising that I don't like the result. Indeed, the rationale
annex appears to have been the main response to public comments;
instead of incorporating them, a rationale annex was added "explaining"
why they weren't done.
In my personal opinion (which I know that some people on the list
don't share), 1003.9 was a fiasco and its poor reception was a major
contributor to the lack of adequate support to keep 1003.19 alive.
Yes, I think that 1003.19 could have been valuable...particularly
if it had been done in the first place instead of what became 1003.9.
--
Richard Maine | Good judgment comes from experience;
[log in to unmask] | experience comes from bad judgment.
| -- Mark Twain
|