[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Updating the IEEE Scheme



The IEEE requires that their standards be updated no later than five years
from the date of publication.  The options are reaffirmation, revision, or
withdrawal.  We have been served notice that for the Scheme standard the
deadline for standards board action is December 1995.  

If we do nothing, the standards board will administratively withdraw the
standard in 12/95.  Otherwise we must revive the Scheme working group.  (Of
course the membership and chairmanship need not be the same as before.)  As
a working group our options are to recommend to our sponsor, the IEEE MSC
(Microprocessor Standards Committee), that the standard be extended,
reaffirmed, or withdrawn, or submit a PAR (Project Authorization Request)
for revision of the standard.  In that case, with no unexpected delays, to
meet the deadline we have only until early January to submit a new draft to
the MSC for ballotting, unless we obtain an extension.  (The first two
folks at IEEE headquarters I asked indicated that 5 months, a reasonable
sounding figure, should be sufficient time from draft submission to the
December standards board meeting.  But I have just received a detailed time
table from Kathy Doty, really seems to know what's going on, and she
concludes it will take 11 months!  Hence the early January date.  The
deadline for submission of a PAR is August 12th.)  

An extension requires an official ballot, with at least 75 replies, of
which at least 75% must be affirmative.  It is good for two years, and is
intended, it seems, for used to buy time for a revision.  Withdrawal and
reaffirmation do not require a ballot, just a working group recommendation.

The IEEE is looking to the Scheme working group for action.  The membership
of the previous working group was never formalized.  It might narrowly be
interpreted as those who attended one or more of the face-to-face meetings,
but at least for the present purpose I prefer the broader interpretation of
those who read the revised report or scheme standard newsgroups.  Hence
the addressing of this message.  The IEEE does not want to know who the
members of the working group are, just the chair.  The chair has to be an
IEEE member, but working group members do not.  The balloting group is
something else: they must be IEEE members (except by appeal) and they may
or may not be involved in development of the draft standard (the working
group's business).

A proposal for how to proceed follows.  Of course suggestions are welcome
from all.

For anything but an administrative withdrawal to happen, a working group
with an elected chair is required.  So we need volunteers to serve as
chair, followed by an election.  (Absent objections, we might proceed
informally, without the formality of nominations and seconds, or even
discussion of who the voting body is.)  The first business of the working
group would be to vote for withdrawal, reaffirmation, extension, revision,
or extension and revision of the Scheme standard.  A vote for revision is
really just a vote to submit a PAR.  If things don't work out the working
group could always vote for one of the other options in the future.  A vote
for extension is really a vote that the working group recommend balloting
for an extension, and indicates an intention to revise the standard within
two years.  If you favor revision, and think we should get on with
it, but do not think we can meet the January deadline, then you should vote
for both extension and revision.

  I suggest a two week period to wait for volunteers (as this is the
  summer, it may take a while for news to reach concerned parties),
  followed by a two week voting period.  The vote for withdrawal,
  reaffirmation, extension, revision, or extension and revision might be
  combined with the vote for the chair.

If we vote for revision, the next step is to agree on the wording in the
PAR.  We could virtually copy the wording of the original PAR, but I
suggest one improvement.  Last time we got only a verbal statement from an
IEEE official that the IEEE standard copyright would not infringe on the
revised reports.  It would be good if we got this in writing this time, and
the only way I expect this will happen is if we write something to the
effect into the PAR.

A new draft will require one or more editors.  To avoid the embarrassment
of a stalled working group or a compromised standard due to a rush to meet
the deadline, it may be wise to ensure we have editors that enjoy the
confidence of the group before submitting a PAR.  Therefore,

  I suggest we also solicit volunteers for editing the standard in the
  next couple of weeks and elect the editors along with the chair.

A possible problem with this suggestion is that until we have agreement on
the scope of changes to be made in a revision it will be hard to judge how
much editing will be involved, which may effect who volunteers.  If the
revision simply brings the standard more into conformance with R4, which
might be accomplished by the January deadline, much less work will be
required than if we attempt to incorporate some of the features that were
hoped would be included in R5, which will almost certainly require an
extension.  (Though many would love to see standardization of a more
complete language, including new features in a standard is dangerous and
must be approached with great caution.  If the standards process simply
inspires work on R5 to continue, that may be its larger purpose.)  Our
previous experience with multiple editors was a very good one, and I expect
that at least if the revision is ambitious a similar arrangement is highly
desirable.  It may, however, be necessary for the editors to have multiple
face-to-face meetings.

I am willing, but not eager, to serve again as chair.  I favor revision if
suitable editors volunteer, but I am unsure what extent of revision is best.