From keld  Fri Jan 22 18:26:25 1999
Received: (from keld@localhost)
	by dkuug.dk (8.8.7/8.8.7) id SAA06505;
	Fri, 22 Jan 1999 18:26:25 +0100 (CET)
	(envelope-from keld)
Date: Fri, 22 Jan 1999 18:26:25 +0100 (CET)
From: Keld J|rn Simonsen <keld>
Message-Id: <199901221726.SAA06505@dkuug.dk>
To: dave@rc.gc.ca, sc22wg15@dkuug.dk
Subject: Re: (SC22WG15.1378) (SC22WG15.1377) (SC22WG15.1376) (wg15tag 2218) (SC22WG15.1374) Comments on minutes of January meeting

> > I don't think it is the proper way, as said earlier.
> > For a fasttrack, ISO is obliged to make a ballot resolution meeting,
> > and resolve the balot comments. Just saying "unfortunately, no,
> > there were enough yes-es" is not an adequate disposition of comment
> > on a fasttrack ballot. The FDIS procedure on the other hand is
> > specificaly designed to have a yes or no vote, without comments.
> > 
> > Keld
>
> I believe that Keld is correct.  Further, there would be nothing to
> prevent ISO from making changes and reballoting a revised draft that
> could be adopted as an International Standard.

Yes, the ISO ballot resolutions commitee may be beyound the control of
the common group, and those ISO bastards could invent anything ;-)

> A better approach might
> be to start on the normal five-step process now and establish a sync
> plan that would get us to an FDIS ballot on the final draft.  This way
> international comments can be fed back into the process and resolved (or
> not) well before we reach a final vote.  The fast track process carries
> too many risks beyond the reach of WG15.  Finally, WGs (and SCs) do not
> normally get involved in fast tracked documents.  They progress directly
> to JTC1.  This could have the effect of eliminating the participation of
> one of the three parties.

As some of you may have guessed, I support Dave's opinions of the
advantages of involving ISO experts and ballotting in the development
process.

> I would suggest that PASC might want to consider a similar course of
> action for similar reasons.  However, I'm not as familiar with PASC/IEEE
> procedures to know if the risk of failure is as great.  I believe that a
> PASC PAR(s) and WG15 NP(s) should be submitted for the July meetings, if
> only to allow these two groups to manage the process within their
> respective organizations.

As I said at the WG15 meeting, there is no need for new ISO NPs,
as these are already approved work items, and we are only doing
revisions. This eliminates the NP voting, and we could go 
directly to a concurrent CD registration and FCD ballot (given
that we have asked SC22 for such permission in time).
So the time to adopt the common POSIX standards in ISO would be
at best 4 months for the concurrent CD registration and FCD ballot,
and 2 months for the FDIS ballot, plus some time in between to
resolve comments and subsequent editing. Comparing to fasttrack,
which is a 6 month ballot and then a ballot resolution meeting
to resolve comments, and resulting editing, it seems to be quite
the same in time consideration.

The bit about no NPs goes as long as we do not introduce vast
new functionality - we can introduce some. Did we decide to undertake
some big chunk of networking APIs?

Keld