Xref: utzoo news.admin:8111 news.software.nntp:483 Path: utzoo!utstat!helios.physics.utoronto.ca!jarvis.csri.toronto.edu!mailrus!usenet.ins.cwru.edu!tut.cis.ohio-state.edu!ukma!david From: david@ms.uky.edu (David Herron -- a slipped disk) Newsgroups: news.admin,news.software.nntp Subject: Re: arbitron/nntp Message-ID: <13695@s.ms.uky.edu> Date: 16 Jan 90 04:46:10 GMT References: <%YJ$W$@masalla.fulcrum.bt.co.uk> Reply-To: david@ms.uky.edu (David Herron -- a slipped disk) Organization: U of Kentucky, Mathematical Sciences Lines: 28 When I wrote the scripts which became arbitron NNTP wasn't even a twinkle in anybody's eyes, at least not that I know of. Think about the information that's available to an NNTP-less system, you've got the active file and peoples' .newsrc files. In an NNTP-full system you have nearly the same thing -- the .newsrc files are elsewhere though. As Ambar suggested, if you can 'control' the reading hosts then you can somehow bring the .newsrc & active file information into one place. erm, I haven't studied the NNTP protocol very closely, despite having made heavy use of it and having been on the mailing lists involved in it's development for years, so I'm not certain how possible this is. As I recall there's a way of telling a difference between a newsreader client and a news-neighbor client, if only that a newsreader client will use a different set of commands than a news-neighbor client. If this is do-able then bugging your NNTP server to log information about what a person read could provide you arbitron-like information. In fact, it would be more accurate... (Arbitron is easily fooled by rn's "catch-up" command) Seperate protocols for news reading and news-neighbor-activities would be useful if only from an administrative point of view. -- <- David Herron; an MMDF guy <- ska: David le casse\*' {rutgers,uunet}!ukma!david, david@UKMA.BITNET <- <- New official address: attmail!sparsdev!dsh@attunix.att.com