drw@culdev1.UUCP (Dale Worley) (10/13/87)
kyle@xanth.UUCP (Kyle Jones) writes: > p.s. I feel the same way about the the USENET maps. Once every six > months would be often enough, since new or changed map entries > can be posted to news.config or news.newsite . Oooof! Please! The great advantage to the maps is that they're distributed in a way that allows a site to maintain the maps *automatically*. For those of us who really aren't allowed to spend the time to read and edit in zillions of small changes each month, the automatic distribution is a godsend. On the other hand, if someone wants to work out a reasonable system where the changes are applied to the site's maps automatically, I'm all for it. Dale -- Dale Worley Cullinet Software ARPA: culdev1!drw@eddie.mit.edu UUCP: ...!seismo!harvard!mit-eddie!culdev1!drw Give me money or kill me! Exercise your childishness -- remember, we are all 10 in some base.
guardian@laidbak.UUCP (Harry Skelton) (10/14/87)
In article <1645@culdev1.UUCP> drw@culdev1.UUCP (Dale Worley) writes: >kyle@xanth.UUCP (Kyle Jones) writes: >> p.s. I feel the same way about the the USENET maps. Once every six >> months would be often enough, since new or changed map entries >> can be posted to news.config or news.newsite . > >On the other hand, if someone wants to work out a reasonable system >where the changes are applied to the site's maps automatically, I'm >all for it. > To Kyle: As new sites are up and running, many are either without the maps or are new to USENET and want maps so they can find out who is where and conduct the business of e-mail and news at top speed. The biggest problem I have seen is when a site is activated, and the maps were just posted to the net, it seems that no one is willing to supply the maps to the new sites. I tell them to wait a month or two and it'll get posted soon enough. Dale: I think if someone was to write a program to scan net.newsites for information, then that would solve your problem. Otherwise I agree with you, I'd rather have the full supply of sites rather than pick names out of the stack. .---------. Harry Skelton : .-. : --- other mail drops --- guardian@laidbak.UUCP : `-'o : ihnp4!laidbak!ugh!bear ihnp4!laidbak!guardian : O : ihnp4!laidbak!laipc!bear `---------' As in the words of Socrates "I drank what*......."
merlin@hqda-ai.UUCP (David S. Hayes) (10/15/87)
In article <1193@laidbak.UUCP>, guardian@laidbak.UUCP (Harry Skelton) writes: > The biggest problem I have seen is when a site is activated, and the maps > were just posted to the net, it seems that no one is willing to supply the > maps to the new sites. I tell them to wait a month or two and it'll get > posted soon enough. New sites on the net do not need the maps. They *do* need smail, which is much smaller (only 170k). I've had no trouble installing smail (BSD system with sendmail, uucp links). The new site designates one of it's uucp neighbors as it's mail relay, giving that site's name to smail as SMARTHOST. All mail for which smail cannot find a path (that's all mail, since they don't yet have maps) is punted to the relay. Presumably, the relay has been on the net for a while, and does have the maps. I should point out that the above strategy is applicable to almost every site. Smail/pathalias were designed so that most sites would need only local routing information, plus paths to a few higher-echelon gateways. -- David S. Hayes, The Merlin of Avalon PhoneNet: (202) 694-6900 UUCP: *!uunet!cos!hqda-ai!merlin ARPA: ai01@hios-pent.arpa
allbery@ncoast.UUCP (Brandon Allbery) (10/19/87)
As quoted from <1645@culdev1.UUCP> by drw@culdev1.UUCP (Dale Worley): +--------------- | On the other hand, if someone wants to work out a reasonable system | where the changes are applied to the site's maps automatically, I'm | all for it. +--------------- Correct me if I'm wrong, but don't we already *have* this? I refer to Larry Wall's greatest gift to the net, "patch". Yes, I know context diffs can be big -- but, assuming the maps are updated automatically rather than manually, regular diffs will work just as well. -- Brandon S. Allbery, moderator of comp.sources.misc {{harvard,mit-eddie}!necntc,well!hoptoad,sun!mandrill!hal}!ncoast!allbery ARPA: necntc!ncoast!allbery@harvard.harvard.edu Fido: 157/502 MCI: BALLBERY <<ncoast Public Access UNIX: +1 216 781 6201 24hrs. 300/1200/2400 baud>> "Just one word, Data: _it_didn't_happen_!" - Tasha Yar
aeusemrs@csun.UUCP (Mike Stump) (10/20/87)
In article <4902@ncoast.UUCP> allbery@ncoast.UUCP (Brandon Allbery) writes: +------- | As quoted from <1645@culdev1.UUCP> by drw@culdev1.UUCP (Dale Worley): | +--------------- | | On the other hand, if someone wants to work out a reasonable system | | where the changes are applied to the site's maps automatically, I'm | | all for it. | +--------------- | | Correct me if I'm wrong, but don't we already *have* this? I refer to Larry | Wall's greatest gift to the net, "patch". Yes, I know context diffs can be | big -- but, assuming the maps are updated automatically rather than manually, | regular diffs will work just as well. +------- I would have to second this motion. With diffs, the powers that be could post diffs every [small unit of time; maybe day, or 3 days], and post the entire maps every [large unit of time; maybe 3-6 months]. Advantage: Maps would be more uptodate then they currently are. Some of the net bandwidth would be returned back to us. '* | patch -d /usr/lib/maps' seams to me like a nice easy command to type. With diffs one does not need to worry about truncated files so much. Disadvantage: I don't have patch (Yes, real disadvantage). I have not expolored all of the advantages, and disadvantages, but I would like to see people consider diffs. -- Mike Stump, Cal State Univ, Northridge Comp Sci Department uucp: {sdcrdcf, ihnp4, hplabs, ttidca, psivax, csustan}!csun!aeusemrs
rob@array.UUCP (Rob Marchand) (10/22/87)
In article <839@csun.UUCP> aeusemrs@csun.UUCP (Mike Stump) writes: > >I would have to second this motion. With diffs, the powers that be could post >diffs every [small unit of time; maybe day, or 3 days], and post the entire >maps every [large unit of time; maybe 3-6 months]. > I agree that this might not be a bad way to go; however, one to three days would likely be too small a granularity for posting of diffs. I know that here (and no doubt at other sites as well), article propagation comes by a variety of routes and delay times. Quite often we receive followups and replies well before the original article. I guess if they were always coming from the same newsgroup and same site, this might not be a problem. This would (could?) make patching a problem, as you would have to wait around until all intermediate patches arrived (if they arrived :-) before patching.... Something to keep in mind, anyways.... Rob Marchand rob@array.UUCP -- Rob Marchand UUCP: {mnetor,utzoo}!lsuc!array!rob Array Systems Computing ARPA: rob%array.UUCP@uunet.UU.NET 200-5000 Dufferin Street Phone : +1(416)736-0900 Fax: (416) 736-4715 Downsview, Ont CANADA M3H 5T5 Telex : 063666 (CNCP EOS TOR) .TO 21:ARY001