Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!cbosgd!mark
From: mark@cbosgd.UUCP (Mark Horton)
Newsgroups: net.news.adm
Subject: Re: child care
Message-ID: <2070@cbosgd.UUCP>
Date: Thu, 1-May-86 14:53:13 EDT
Article-I.D.: cbosgd.2070
Posted: Thu May  1 14:53:13 1986
Date-Received: Sat, 3-May-86 19:06:49 EDT
References: <905@decuac.UUCP> <150@fai.UUCP>
Organization: AT&T Bell Laboratories, Columbus, Oh
Lines: 26

Well, actually, you see so many wrong postings from AT&T in NJ
for two reasons I can think of:

(1) The level of support of Usenet by AT&T in NJ is lower than in
other areas.  There are only two backbone sites (clyde and ulysses)
in all of AT&T in NJ, and the distribution arrangements for the rest
of the machines tend to be pretty ad-hoc.  The typical system administrator
is pretty busy with other things, and doesn't really have time to
participate in the electronic community or Usenet.  (Many of them won't
even answer mail.)  As a result, the users have no formal instruction
from the SA, no written documentation, just word of mouth.  The typical
posting comes from someone who doesn't read news, has something for sale,
and heard through the grapevine that they should try to post it to Usenet.

(2) AT&T accounts for about 1/3 of all of Usenet, and over half of that
is in NJ.  So of the erroneous postings you see, a large chunk are from
there because a large chunk of the net is there.

Of course, any administrative or software fix to the problem won't work,
because the SAs of the offending machines won't install the new software.
In this case, the only possible fix is to start a new network with a human
filter in the loop (in appropriate newsgroups) to prevent such mistakes
from making it to the general readership.  This is what we're hoping to
do with Stargate.

	Mark Horton