Re: Fortran 2008 (was Re: New style DO syntax?)


Dan Nagle wrote:

The single greatest reason why Fortran 2008 is so large is Co-arrays.
Personally, I favor co-arrays. So you see the sort of compromises
needed to get things done. Again, if you believe co-arrays is being
rushed (they've been implemented for ten years now), come vote
against them.

The problem with co-arrays is that there's only one implementation,
and that is by a vendor whose market share is minuscule. There can't
be more than a couple dozen people who have significant code
development
experience with co-arrays, and the spectrum of applications with which
they
have worked is necessarily limited. Given that co-arrays will take up
a
large fraction of valuable committee time, and their implementation in
practice
may be quite expensive, why is it wise to proceed with a
standardization effort
now?

If you ask me, I think it would be much better for purposes of F2008 to
put
some effort into a better set of Fortran bindings to MPI, say as a
technical report.
There are thousands of people who work with MPI on a regular basis, and
several orders of magnitude more MPI codes than co-array codes.
Compared to
co-arrays, the effort involved would be much less and would be a
greater service
to the community.

It may well be that co-arrays turn out to be useful - but nobody really
knows because
almost nobody has experience with them. We already have FORALL, which
adds no capability that couldn't be gotten from vectorized DO loops or
similar
techniques, and it was derived from an experimental Fortran dialect
that was already dead
by the time the F95 standard was completed.

--Eric

.

Quantcast