Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!uunet!munnari.oz.au!cs.mu.oz.au!ok
From: ok@cs.mu.oz.au (Richard O'Keefe)
Newsgroups: comp.lang.prolog
Subject: Re: Prolog standardization
Keywords: Paranoia
Message-ID: <2673@munnari.oz.au>
Date: 10 Nov 89 11:03:57 GMT
References: <2609@munnari.oz.au> <696@sce.carleton.ca> <2643@munnari.oz.au> <1281@quintus.UUCP>
Sender: news@cs.mu.oz.au
Lines: 69

In article <1281@quintus.UUCP>, dave@quintus.UUCP (David Bowen) writes:
> Richard O'Keefe wrote:
> >Let's be honest about this:  democracy is rule BY the same people
> >who are affected.  Some of us (like me) can't vote for a representative
> >even in principle.  Those of us who belong to countries which are
> >"represented" did not get any opportunity to pick their "representatives".
> 
> This last statement is not true in general.  In the U.S., at least, the
> future delegates to WG17 will be chosen by the ANSI working group on Prolog
> standardization.  Membership of this working group is open to anyone who
> lives in the U.S., or works for a U.S.-based organization, and is able to
> attend the meetings.

In fairness to Roger Scowen and the others, I should make it clear that I
did have the opportunity to be on the BSI committee:  starting shortly
after I left the UK.  Now that I've left the USA, there's an ANSI committee.
Perhaps I should leave some more countries...

Dave Bowen's last sentence appears to say that USA citizens & workers can
have some say in who represents them on WG17 _if_ they are rich enough.
"Put your money where your mouth is" is a reasonable criterion, but it is
not what we call "democracy".  I do not claim that WG17 *ought* to be a
democracy; that criticism was a response to another posting.

Let's compare this with the C and Fortran 8X standards.  The C standard was
developed as an ANSI standard.  You could become an "observing member" by
paying a fairly modest sum (I think it was about US$100) and this meant that
you got copies of the documents (that was a really worthwhile investment,
because buying a copy of the draft from Global Engineering Documents cost
about US$70 in California, and there were several drafts).  However, that
didn't get you representation.  To have a voice on the committee, you had
to pay rather more money, and you had to go to the meetings.  The result is
that a lot of small companies like Quintus just had to trust X3J11 to do
the right thing, yet C is _vital_ to Quintus' operations.  Fortunately,
X3J11 took it as a firm rule not to break working code if they could
possibly help it, and when I left Quintus at the end of last year it was
clear that ANSI C was not going to hurt Quintus one little bit.  The X3J3
team working on Fortran 8X were similarly determined not to break any
aspect of Fortran 77 -- that meant some rather contorted, not to say
unpleasant, syntax in Fortran 8X.  There is a lot of fear and fury about
Fortran 8X, but the worst anyone has been able to claim with respect to
old code is that new compilers are likely to be slow and buggy and to
generate slow code, which one would expect no matter what 8X looked like.
One of the most serious charges against X3J3 has been that copies of the
draft standard were expensive.

If the standard were being developed entirely by an ANSI committee, or if
WG17 delegated the task to an ANSI commitee as was done with X3J3, I for
one would have no apprehension about the result.  ANSI committees in the
past have done a pretty good job of heeding letters from the public.  But
that is not the way it is being done.

One thing which could be a big help in making the development of the
Prolog standard accessible to Prolog users generally would be setting
up a computer in the USA (and ideally one in Europe and one in Australia
and one in the Near North (ok, the Far East for most of you)) to act as
a distribution point for current drafts of the standard.  This is not a
heck of a lot of work:  the mail server "netlib" software already exists
and can be picked up free.  The main work required is the work of putting
the documents in machine-readable form, which is a good idea anyway as it
would make them easier to proof-read, index, and otherwise work on.  Then
publish the E-mail addresses of the server sites in SIGPLAN Notices and
the Logic Programming Newsletter and comp.lang.{prolog,misc}.  (I did
suggest this in 1984, having been able to check Ada questions on
UCSC-ECLB.  Great idea.)  I'm sure that sites could be found which would
volunteer to do this if the drafts and comments were made available to
them in machine-readable form.  It has to be made easy for people to find
out what the state of the standard is and cheap for WG17 to tell them, and
that seems like a good way.