On Sat, 23 Aug 2008 02:31:50 -0700, Van Snyder <[log in to unmask]> wrote:
> The view of the US delegation to the ISO
>working group is that parallelism will become so important so soon that
>CoArrays should not be delayed, and should not be optional (optional
>stuff has never been successful in Fortran).
The two optional parts of the current standard are unsuccessful not because
they are optional, but because there is no demand for them.
Part 2 (varying length character strings) was created principally to help
buy votes for Fortran 90. It also served as a feasibility proof for
modules. Serious usage was not the prime concern.
Proposals for conditional compilation arose partly from an informal request
from HPFF in the early 1990s and partly as it was realised that cpp and fpp
were being widely used but were not considered sufficiently Fortran-like.
Conditional compilation was rejected as a requirement for inclusion in the
base language but was accepted in 1995, with less than wholehearted
enthusiasm, to be an optional part 3. The standard was published in 1999,
by which time anyone who wanted this facility was almost certainly using
other tools.
The previous option in Fortran was subset Fortran 77 which never really took
off, probably because it was overtaken by increasing machine size and
improved compiling techniques.
Optional parts are however fully part of the standard and are there to be
used by those vendors and users who want them. There is no logic in saying
that because coarrays are important, they must be in the base language.
Important for whom? If there is demand they will be used. There are a
number of reasons why they should be an option. A primary one is that, when
it is realised that the coarray model is not the best one to pursue, the
Fortran base language will not be lumbered with lots of redundant syntax and
complexity.
David Muxworthy
|