Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!brutus.cs.uiuc.edu!coolidge
From: coolidge@brutus.cs.uiuc.edu (John Coolidge)
Newsgroups: news.admin
Subject: Re: The Dynamics of Debate on USENET
Message-ID: <1989Sep28.173157.22915@brutus.cs.uiuc.edu>
Date: 28 Sep 89 17:31:57 GMT
References: <35033@apple.Apple.COM> <46115@bbn.COM> <35037@apple.Apple.COM> 
Sender: news@brutus.cs.uiuc.edu
Reply-To: coolidge@cs.uiuc.edu
Organization: U of Illinois, CS Dept., Systems Research Group
Lines: 39

karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) writes:
>bmw@isgtec.uucp writes:
>   Interestingly, USENET debates have a property that doesn't exist with
>   face-to-face debates: multiple participants separated by huge
>   temporal delays.

>I tend not to give much credence to this argument these days, because
>of the low propagation times of the NNTP massfeed hubs, one of which
>is my site.  Since at least last winter, there's been a pronounced
>tendency to run nntpxmit _often_.  Erik Fair/Apple&UCB has nntpxmit
>going off every single minute.  I'm doing it every other.  Brian
>Kantor/UCSD has a modified nntpsend script which sends _continuously_
>until it runs out of things to send (which, I suspect, is almost
>never).  It is not at all uncommon for me to see an article arrive
>here which has already come through 6 or 7 hops with a Date: that's
>less than 20 minutes in the past.

Count me in with the continuous people (running an experimental sender
that is nowhere near ready for release, alas). I see the same sort of
thing (6-7 hops or more in 20 minutes). I expect this will actually get
faster with new news software (TMNN or C) that puts less burden on the
news machines and makes people more willing to run fast transmissions.

On the other hand, Usenet is a heterogeneous network and lots of readers
are on sites away from the NNTP 'backbone'. This leads to an even more
interesting phenonminon: those away from the backbone can often be
presented with an entire discussion, four or five messages-worth, that
came over in one big lump. This will probably affect responses some way
or another: it might inhibit a good response, because the recipient thinks
they'd be butting in on an ongoing conversation, or it might provoke a
better response because the responder knows what's already been covered.

--John

--------------------------------------------------------------------------
John L. Coolidge     Internet:coolidge@cs.uiuc.edu   UUCP:uiucdcs!coolidge
Of course I don't speak for the U of I (or anyone else except myself)
Copyright 1989 John L. Coolidge. Copying allowed if (and only if) attributed.
You may redistribute this article if and only if your recipients may as well.