Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!wuarchive!texbell!splut!jay From: jay@splut.conmicro.com (Jay "you ignorant splut!" Maynard) Newsgroups: news.admin Subject: Re: Old Group Deletion Procedures Keywords: delete obsolete newsgroups Message-ID: <2920@splut.conmicro.com> Date: 5 Oct 89 19:28:40 GMT References: <3137@ur-cc.UUCP> <1989Sep26.212755.8458@elroy.jpl.nasa.gov> <4492@ncar.ucar.edu> <2908@splut.conmicro.com> <5629@tank.uchicago.edu> <2915@splut.conmicro.com> <4578@ncar.ucar.edu> Reply-To: jay@splut.conmicro.com (Jay "you ignorant splut!" Maynard) Organization: Confederate Microsystems, League City, TX Lines: 70 In article <4578@ncar.ucar.edu> woods@handies.UCAR.EDU (Greg Woods) writes: > Yes, ASSUMING that foo.bar is in the "logical" place in the namespace. >Most flame wars over group creations, with very few exceptions, have centered >around this aspect of it, rather than against the idea of the group itself. More specifically, most flame wars over group creations have centered around which high-level name should apply. Let's take the current discussion as a case in point: the discussion isn't over whether or not "aquaria" is appropriate, but whether or not "sci" is appropriate. It's a lot easier for someone to find "aquaria". The high-level name is becoming more and more a distribution classification as well as a subject classification. > What you have done here is convincingly argue that there is such a thing >as too few newsgroups. I agree. However, this does NOT prove that there >ISN'T such a thing as too MANY newsgroups. I think that there is indeed such a point. We obvoiusly differ on where that point is: it's a question of the shape of the utility vs. groups-to-articles-ratio curve I described last posting. >The "there is no foo.bar group" >could just as likely result from the fact that the group is named baz.foo.bar >instead, and the user couldn't find it. The likelihood of this happening >increases with the number of newsgroups, and in particular it will be worse >if the organization of the namespace is worse. I agree here. This problem should be amenable to software solution (or at least assistance), though: as has been suggested before, a program to do an intelligent 'grep foobar /usr/lib/news/newsgroups' would help this problem a great deal. >So, looking at BOTH Jay's and my arguments, I would conclude: >1) Groups which are really needed SHOULD be created. This means an absolute > limit on the number of newsgroups is not a good idea. Absolutely. >2) There should be SOME kind of limit on how many newsgroups we have. This > means that there should be some kind of determination that a group IS in > fact needed before it is created. It also means that it would be a good > idea to have some way of getting rid of groups that are no longer needed. Again, I agree, though I have serious doubts about the practicality of the decision process. The voting procedure we have attempts to answer this question, with varying degrees of success. I would argue that, in case of an ambivalent result, that the group should be created, since it's more likely to solve the missing group problem, but this assumes the presence of a reliable way to remove newsgroups - which we don't have. >3) We can handle a lot more newsgroups if they are placed in the namespace > properly. This means naming issues ARE important and SHOULD be considered > when creating a new group. This implies that there should be some way > of preventing the creation of misnamed groups, or forcing a name change > to something more reasonable. I would prefer a means of changing the name during the proposal process to simply preventing the creation of misnamed groups. To take my favorite case of sour grapes, I think we'd be much better off with soc.women.computing than no group at all, but that either of those are preferable in the context of this discussion to comp.society.women. It would be a lot easier to find, and would fit much better into the context of the net. -- Jay Maynard, EMT-P, K5ZC, PP-ASEL | Never ascribe to malice that which can jay@splut.conmicro.com (eieio)| adequately be explained by stupidity. {attctc,bellcore}!texbell!splut!jay +---------------------------------------- America works less when you say..."Union Yes!"