Newsgroups: comp.parallel
From: eugene@pioneer.arc.nasa.gov (Eugene N. Miya)
Subject: News group maintenance was: intel iPSC newsgroup
Organization: NASA Ames Res. Ctr. Mtn Vw CA 94035
Date: 6 Apr 1995 21:45:17 GMT
Message-ID: <3m95j9$klc@usenet.srv.cis.pitt.edu>

In article <3lrkh6$amf@usenet.srv.cis.pitt.edu> abali@watson.ibm.com
(Bulent Abali) writes:
>This news group doesn't have the level of discussion it deserves.
>Many articles are about CFPs or about people asking "Is there an XYZ
>package/machine..." and threads stop there.  I would like to see more
>activity, however I think moderation is not helping with that.  We
>should perhaps discuss pros and cons of having an unmoderated
>newsgroup.
>
>[So what topic would you like to discuss?  -Mike (a moderator...)]

I'll comment on that as one of the earliest posters to this group,
one of my posts was cited in a referred journal (possibly one of the
earliest citations of net.news in the print media), and I maintain the
parallel biblio (which extends beyond this group).

0) I think moderation made this group one of the better groups on the
net.  We don't see spam postings, calls for cards to Craig Shergold,
or ads from Canter & Siegel [usually].  The past moderators had some
idea when to cut off certain discussions.

0a) It would be nice if real editing took place and not mere moderation,
but that's asking too much.

1) Ah? what level of discussion do you think it deserves?
The Usenet isn't a magazine.  People who post are busy.
People have also learned to be very sensitive about what they post in
this group.

2) Many queries posted here could be better answered with an adequate
FAQ, literature searches.  I helped a tiny bit with Steve's.  The new
moderators have said they were going to compose one.  I coined the term
FAQ in the mid 1980s.  I believe it's the responsibility of the entire
group not merely a moderator or an FAQ maintainer to compose and
maintain FAQs.  Where's your contribution?  That's rhetorical.

You can't sit back passively and expect the expertise of the world to
come to you.  The net has never worked that way.  This isn't a magazine.
Don't you wish the world were like the film Annie Hall where you would
pull experts from the side lines at no cost?

3) It would be possible to summarize some past discussion using FAQ type
files: "What is a Class 6 computer?" "What constitutes a supercomputer?"
etc. are all pretty basic posts of past circular discussions.

The net lacks any real sense of community memory.  Experts get tired of
rehashing old topics.  That's where my biblio comes in, but it's now a
ragged tag volunteer effort.

People would almost rather have you view their Web pages (almost, I can
think of quite a few professional lurkers).

4) Working toward referred papers gives people more credit.
Really in depth discussions and "discoveries" rarely appear here.
I would in fact venture, they have never appeared here.
People volunteer information here.  Some of it is wrong, some from
differing opinions.

This is not the place to merely say, "The Shuttle blew up" or "Cray
Computer declared chapter 11" a hundred times which is a problem with
unmoderated groups.

5) You get discussion by sticking your neck out and hopefully the 
participants aren't too tired of seeing said topic.

1a) You work for IBM.  This is a special note: from the earliest days
of the net from around the time of the 3090 when super and parallel computing
were the providence of a few small companies, and the VNET was not
attached to the wider internet, a perception exists that IBM people are
among the biggest trollers of information on the net.  This is a market of
some sensitivity, and that engendered some harsh feelings (some justified
and others not justified) among certain communities.  You should at
least be aware of this.  Some analysts claim IBM's entries into parallelism
"legitimized" the market, that infuriated certain companies.  Some will
welcome you with open arms, others will see you as "the competition."
The mere fact you posted using these hosts:
@usenet.srv.cis.pitt.edu @watson.ibm.com
doesn't look good to long time observers.

The sensitivity is between companies, it's between competing university,
government and industrial groups, and it's between countries.  We
can claim that we wish to work together, but I have yet to see a lot of
heavy cooperation.  It partially has to do with the merit system (we get
greater merit for sole accomplishments).

You want to consdier yourself lucky if you get email answers.

This question in fact is also an FAQ.
The issue has nothing to do with whether the group is moderated or not.


