James Giles wrote:
>
> "Van Snyder" <[log in to unmask]> wrote:
> ...
> > Another problem is that, out of the million or so Fortran users world wide,
> > 14 are working on the next standard.
> ...
> > If you want J3 to do something different, sign up to do some of the work.
> >
> > I had a dream that at least some of the work could be done by e-mail,
> > without actual attendance at meetings.  That hasn't worked out -- and I
> > don't know why not.
>
> There are those of us who wouldn't mind working on the standard
> but simply can't afford to travel (as stay at expensive hotels) for four
> weeks out of the year (or even the two or three time a year required
> to mantain full member status).  Other than looking for typos, what
> can anyone do?
>

Look for logical problems and incompleteness or inconsistencies.

Try to think of something you'd like to do, for example call several
of the MPI or WINDOWS GUI routines, or do I/O on a complicated
mutually recursive data type (like a family where the head of the house,
the wife, and the kids are all types that mutually point to each other,
with the kids being a linked list).  Try some "real world" problem
and see if it works reasonably well.

It's a fine line between "incomplete" and "complete I guess, but it
doesn't do as much as I like."  But people can take a stab at it.
Same thing for "reasonably well".  The things to look for would be
obvious holes.  It's a judgment call as to what can be fixed at
this stage.  But most J3 members would be receptive to comments
along the line of "I agree with what you are trying to do here,
but it doesn't quite solve a large class of problems because of a
small thing, blah, blah..."

That, and as someone else said, look for things that don't make
logical sense between chapter 5 and 12, etc.

Dick Hendrickson
> --
> J. Giles