Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ukma!tut.cis.ohio-state.edu!triceratops.cis.ohio-state.edu!karl From: karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) Newsgroups: news.admin Subject: Re: The Dynamics of Debate on USENET Message-ID:Date: 28 Sep 89 14:47:21 GMT References: <35033@apple.Apple.COM> <46115@bbn.COM> <35037@apple.Apple.COM> <147@isgtec.UUCP> Sender: news@tut.cis.ohio-state.edu Organization: OSU Lines: 24 In-reply-to: bmw@isgtec.UUCP's message of 27 Sep 89 12:15:39 GMT bmw@isgtec.uucp writes: >o If it's already been said, don't say it again. 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. Now, this is by no means true for everyone. UUCP sites which are fed by every-other-hour (or whatever time frame) batching still have the delays. But it seems to me that delays are less of a concern now than ever. --Karl