Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!adm!cmcl2!lanl!lambda!jlg
From: jlg@lambda.UUCP (Jim Giles)
Newsgroups: comp.lang.fortran
Subject: Re: FIRST public review - response finally
Message-ID: <14177@lambda.UUCP>
Date: 15 Dec 89 23:44:20 GMT
References: <589@unmvax.unm.edu>
Lines: 27

From article <589@unmvax.unm.edu>, by brainerd@unmvax.unm.edu (Walt Brainerd):
> [...] 
> In any case, preliminary news is that there were a lot more favorable
> comments on the second review than on the first,
> enough to form a clear majority of those responding
> (also, there were fewer responding).  [...]

Most of the people I'm acquainted with who sent negative comments on the
first public review didn't bother to send comments on the second.  This
was usually explained to me as being a waste of time since the committee
appeared to ignore the bulk of the first commentary.  There is some
justification to this remark.  For example, most of the last public
review comments rejected the "specified precision" feature of the
first draft.  But, not only did the committee keep it all (though it
was moved to intrinsic procedures), but they actually made it harder to 
use by interpolating some processor dependent KIND parameters into the
feature.

If this observation is true, then the second review comments will be
fewer and more positive - not because of support of the new proposal,
but because of opposition to the committee.  It's too bad too.  I think
the present proposal (if adopted) will be the death knell for Fortran
as a standard language.  Many users will not ask for, nor will some vendors
supply, the full language.  Instead, there will be a patchwork of subsets
plus local extensions to the standard (like genuine POINTERs - as opposed to
the descriptors from hell in the proposed standard).  This is the kind of
situation which standards are supposed to avoid!