Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!decuac!avolio From: avolio@decuac.DEC.COM (Frederick M. Avolio) Newsgroups: comp.mail.uucp Subject: Re: mod.map problems and directions Message-ID: <1117@decuac.DEC.COM> Date: Wed, 3-Dec-86 09:18:46 EST Article-I.D.: decuac.1117 Posted: Wed Dec 3 09:18:46 1986 Date-Received: Wed, 3-Dec-86 20:10:34 EST References: <3129@cbosgd.ATT.COM> <6685@decwrl.DEC.COM> Organization: DEC SWS, Landover, MD Lines: 30 In article <6685@decwrl.DEC.COM>, reid@decwrl.DEC.COM (Brian Reid) writes: > ... > The thing that I am wistful about is that I have just finished writing a very > elaborate program that can make high-quality geographic maps of USENET, using > the #L and #U fields. This information is not present in the d. files, and in > the case of the #L field it is often meaningless, since domains do not have a > specific geographic location. > > The u. files, though intended for mail routing maps, serve an immensely > valuable purpose as a "white pages" directory; the d. files cannot possibly > provide this information. I agree with Brian. While the UUCP `map' is getting (or has gotten) unwieldy, it does provide useful information. The location for each site, is the obvious one. For Brian's purpose, of course, but also it is used to point people to possible UUCP links in their area. Rather than tossing out all entries for all sites in, say, AT&T, excepting cbatt (or whatever), it'd be nice to have a way to eliminate entries for many sites at the same location. This is different than saying "in the same domain." I cannot imagine that the suggestion is really "For all X in X.domain123 remove all entries except where X == domain authority for `domain123'." This would certainly produce terrible routings of mail over the phone lines. I know many folks in the UUCP Project work for AT&T but.... :-). I think clarification is in order. (I believed that I was the only one bewildered, but Brian is a pretty sharp guy, so if *he* doesn't quite get it there must be something wrong. Fred