(j3.2006) (SC22WG5.5869) [ukfortran] N2126

[Steve Lionel]

On Mon, Jun 5, 2017 at 5:16 AM, Anton Shterenlikht <mexas at bris.ac.uk> wrote:

>
> Talking about comp.lang.fortran, which
> apart from vendor specific groups and lists
> seems to be the most active Fortran
> discussion platform, is it in WG5 interest
> to engage more directly with CLF?
>

I can think of at least five WG5 members who regularly participate in CLF,
and it is not uncommon for issues raised in CLF to be discussed by J3 and
WG5.

I think Dan's paper, pessimistic as it is, overstates the resources
available to Fortran implementors. It does little good to the language if
lots of shiny features are added in a standard, but nobody can use them
because of a lack of compiler support. I have been told by many users that
they aren't allowed to use a feature unless it is supported by at least
three if not four different compilers. What happens then is that
decision-makers frown upon continued use of Fortran since implementations
are "so far behind" the standard.

F2015 was much larger than originally envisioned, and look at the current
state of compilers - by the end of this year there will be, what, a
whopping two full F08 compilers, seven years after publication? Is this
what we want for the language?

Yeah, exceptions would be nice. Back when I worked on VMS Fortran it was
great to be able to leverage the language-independent OS support for SEH,
but that vision doesn't exist on other platforms. I can think of many other
features I'd like to see. But what I want even more is increased acceptance
of Fortran as something other than ("do people still use that?") Allowing
the compilers to catch up is a large part of that.

Steve


-- 
.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.j3-fortran.org/pipermail/j3/attachments/20170605/f473e4fc/attachment.html