Xref: utzoo news.software.b:4253 news.newusers.questions:1571 Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!usc!zaphod.mps.ohio-state.edu!rpi!tale From: tale@cs.rpi.edu (David C Lawrence) Newsgroups: news.software.b,news.newusers.questions Subject: Re: How to kill superseded articles Message-ID: Date: 28 Feb 90 03:13:27 GMT References: <1640@sparko.gwu.edu> Followup-To: news.software.b Organization: Rensselaer Polytechnic Institute, Troy NY Lines: 33 In article <1640@sparko.gwu.edu> se208104@seas.gwu.edu (William Lai) writes: > Is there a way to KILL the superceded article upon receipt of the > new one? I believe that one of the header lines is "supercede > such-and-such". Maybe that could be used? What news system? C News, as of the 10 Jan 90 patch, now will handle the Supercedes: header directly in relaynews. The original article is cancelled. Prior to this you had to use $NEWSBIN/expire/superkludge to get the supercedes header to work, and then it only did it in the groups named on the command line. Now superceding is done a little bit more efficiently at the cost of a slightly less efficient relaynews. As far as I know, which in this case is only memories of when we were a B News site, supercedes are not done by B 2.11 but they might be in the most recently releases. (If so, someone who hopefully point that out -- if not, then you could use a modified superkludge.) The possibility that Supercedes: is not done by B News is somewhat corroborated by the fact that the header is not mentioned in Section 2 of RFC 1036. One further note on superkludge: it only removes the file of the superceded article. References to the file in lib/history* will remain until the article expires normally. I don't know how other news systems deal with Supercedes:. Dave -- (setq mail '("tale@cs.rpi.edu" "tale@ai.mit.edu" "tale@rpitsmts.bitnet")) "If a machine can be made so that an idiot can use it, then only an idiot will use it." -- Tadao Ichikawa