Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!bellcore!decvax!decwrl!glacier!kestrel!king
From: king@kestrel.UUCP
Newsgroups: net.news.group,net.abortion,net.misc,net.motss,net.philosophy,net.politics
Subject: reorganization of soapbox newsgroups
Message-ID: <8433@kestrel.ARPA>
Date: Sat, 24-May-86 19:36:46 EDT
Article-I.D.: kestrel.8433
Posted: Sat May 24 19:36:46 1986
Date-Received: Mon, 26-May-86 01:38:29 EDT
Organization: Kestrel Institute, Palo Alto, CA
Lines: 17
Xref: watmath net.news.group:5712 net.abortion:3197 net.misc:9703 net.motss:2942 net.philosophy:5491 net.politics:16384


Has anybody determined how much volume would be reduced if there were a
protocol that would allow messages to incorporate pieces of other
messages by references?  Something like 
<here goes lines 123-178 of <2139@cbosgd.UUCP>, reindented by ">> ">
(placed in an outgoing message which would have a header line
Incorporations: <2139@cbosgd.UUCP>
which would be inserted by the operation that includes the text of the
original.)  My feeling is that the reduction would be considerable in
exactly those groups that have a problem.  After all, the people who
send these messages are a small group with finite typing speed.

Such inclusions could be found fairly quickly by hashing each line of the
outgoing message and of potential inclusions on the last 40 characters of
the line.  A simple scheme to find these things could be used, since
there is no motivation to dodge the mechanism.

comments?