Path: utzoo!utgpu!utstat!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!samsung!gem.mps.ohio-state.edu!sunybcs!sbcs!stealth!brnstnd
From: brnstnd@stealth.acf.nyu.edu (Dan Bernstein)
Newsgroups: news.groups
Subject: The right way to run a trial hierarchy
Message-ID: <3923@sbcs.sunysb.edu>
Date: 13 Nov 89 03:50:36 GMT
Sender: news@sbcs.sunysb.edu
Reply-To: brnstnd@stealth.acf.nyu.edu (Dan Bernstein)
Distribution: usa
Lines: 17

Someone creates trial.foo.bar. After a month, if he thinks trial.foo.bar 
has proven useful, he asks the Namespace Committee to make it into a
real group, foo.bar. The Committee's only job is to keep the namespace
sensible, so that a user can find the right group to read or to post in;
by default all groups pass.

Notice that the trial name includes the entire real name. That lets the
readers decide the right name by deed rather than word. For example,
Richard creates trial.sci.aquaria and Bryce creates trial.rec.aquaria.
After a month, depending on how fish owners have used the groups, one or
both proposers go to the Committee; and either sci.aquaria or rec.aquaria
is created as the trial groups disappear.

The choice of people for the Committee is unimportant; probably Spaf,
Greg, and whoever thinks he's Czar at the moment would suffice.

---Dan