Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!samsung!uakari.primate.wisc.edu!dogie.macc.wisc.edu!uwvax!umn-d-ub!umn-cs!meccts!viper!dave
From: dave@viper.Lynx.MN.Org (David Messer)
Newsgroups: news.groups
Subject: Re: Well, use it... (Actually net topology)
Keywords: get the facts, jack
Message-ID: <2980@viper.Lynx.MN.Org>
Date: 6 Dec 89 15:27:31 GMT
References: <7551@fear+loathing.UUCP> <128533@sun.Eng.Sun.COM> <17386@umn-cs.CS.UMN.EDU> <128633@sun.Eng.Sun.COM>
Reply-To: dave@viper.Lynx.MN.Org (David Messer)
Organization: Lynx Data Systems, Eagan, MN
Lines: 23

In article <128633@sun.Eng.Sun.COM> briang@sun.UUCP (Brian Gordon) writes:
 >Usenet, as I understand it, can be viewed as a ring (ignoring the leaf nodes
 >and the relatively uncommon site with multiple feeds).  Messages posted from a
 >site are sent in both directions around the ring (again ignoring leaf nodes)
 >and the message-id is used to keep it from "passing itself" in either
 >direction.

Usenet, in fact, is a directed graph.  Near the leaf nodes it
tends to resemble a tree, but near the 'backbone' it is much
more fully connected.  Minnesota, for instance, has two (at
least) full feeds coming into the area, which feed a top
level of sites, each of which feed each other.  (I hear we
are more organized than most local areas however.)

In any event, it would take at least two key sites to stop
a newsgroup from reaching a significant fraction of the local
sites.  Refusing to carry a group such as sci.aquaria is more
symbolic than practical.  Besides, why should it be any other
way?  Just because I don't want to carry sci.aquaria it
doesn't make it right to force my opinion on others.
-- 
Remember Tiananmen Square.           | David Messer       dave@Lynx.MN.Org -or-
                                     | Lynx Data Systems  ...!bungia!viper!dave