NetNews Usenet Archive 1993 #3/NN_1993_3.iso/spool/comp/org/sug/621

Internet Message Format | 1993-01-21 | 4.8 KB

Xref: sparky comp.org.sug:621 comp.org.usenix:1351 comp.unix.wizards:5585 comp.unix.bsd:11647
Newsgroups: comp.org.sug,comp.org.usenix,comp.unix.wizards,comp.unix.bsd,news.sys-admin
Path: sparky!uunet!cs.utexas.edu!csc.ti.com!tilde.csc.ti.com!fstop.csc.ti.com!choctaw!peterson
From: [e-mail] (Bob Peterson)
Subject: Re: IMPORTANT: POSIX threatens our use of lp/lpr and friends
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: choctaw
Organization: Texas Instruments
References: <[e-mail]>
Date: Thu, 21 Jan 1993 12:49:47 GMT
Lines: 89

------
  Note: Posted for the person named in the signature.
------

I have been active in the standards community, both domestically and
internationally for about 15 years and have served on the IEEE
Standards board and chaired the IEEE REVCOM committee for the last
three years.  REVCOM, for those of you who don't know, is the committe
which reviews all of the IEEE standards following the balloting
process, determines whether or not the document represents the
consensus of industry and recommends to the IEEE Standards Board
whether or not a particular document should be accepted by the Board
as a standard or returned to the sponsor for further work due to
procedural (and sometimes technical) reasons.

REVCOM *STRONGLY* frowns on ballot stuffing or other attempts to
       ==========
subvert the consensus process and those doing so may find themselves
excluded from participating in future ballots at the Sponsor's
discretion or have their votes discarded or lumped together into a
single vote.

The recent message from Ran Atkinson <[e-mail]>
entitled "IMPORTANT: POSIX threatens our use of lp/lpr and friends "
has several errors which I would like to clear up.

|>   Namely get on the balloting group for the POSIX.7 standard.  ANYONE
|> who breathes can get on the balloting group.  If you are an IEEE
|> member then give your membership number.  If not, call up the IEEE and
|> say that you are "a party of interest for the POSIX standards and have
|> experience with UNIX".

While it is true that *anyone* may request to be on an IEEE Sponsor
ballot, *only* those who are IEEE or Computer Society members will
receive an actual yes/no ballot.  Others will receive an information
copy of the document and a form for responding with comments.
Although all comments (from anyone) are required to be considered and
responded to, by IEEE balloting rules, a document is approvable when 75%
of the yes/no votes are affirmative.

Obviously the higher the percentage, the better, and Sponsors are
encouraged to resolve favorable as many comments as possible to achive
.gt. 75%.  However 75% is acceptable provided everyone has had a
chance to see all of the unresolved negative comments and the
committee's reasons why they cannot be resolved favorably.

|>   Once the ballot goes out, folks in the balloting group can object to
|> basing the standard on Palladium and state that you "want the standard
|> to be based on existing practice as demonstrated by UNIX System V and
|> BSD UNIX rather than Palladium which is not existing practice and is
|> known not to work well."
|>
|>   Please recirculate this note to folks you think might be of interest
|> and circulate it at USENIX and other UNIX User's gatherings.
|>

This is the kind of thing that can be used as evidence of block
voting and which is detremental to the consensus process.

Whatever happens, you can be relatively assured that REVCOM will not
approve anything which does not represent a consensus of the industry.
But recognise that IEEE and ANSI definition of consensus is "More than
a majority, but not unananimity."  It is entirely possible for 25% of
the population to be opposed and still have a standard approved.

None of the above is intended to discourage anyone from voicing their
opinion or becoming actively involved in the committees which develop
standards.  In fact, I encourage the latter so that more of industry,
especially the military, is represented during development.  But the
most effective way to do this is to become involved in the committee
work early on as opposed to complaining when others have done all of
the work in a way you may not happen to like.

If anyone wishes to discuss this in further detail, I can be reached
below.

Regards,

Clyde


=======================================================================
                                 |
Clyde R. Camp, P.E.              |   " There is nothing so susceptible
Texas Instruments, Incorporated  |     to the secret manipulations of
[phone]                   |     evil men as a democracy. "
[phone] (Fax)             |             Plutarch, circa 120ad
[e-mail]                  |
=======================================================================