Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!bfmny0!tneff From: tneff@bfmny0.UU.NET (Tom Neff) Newsgroups: news.config Subject: Re: Unregistered-but-referenced sites (was Re: deleted UUCP sites) Message-ID: <15066@bfmny0.UU.NET> Date: 4 Jan 90 15:34:13 GMT References: <4195@convex.UUCP> <30273@mcdchg.chg.mcd.mot.com> <43496@improper.coherent.com> Reply-To: tneff@bfmny0.UU.NET (Tom Neff) Distribution: news Lines: 34 I understand Mel's idea of the conditionally registered site being blown away when some site decides to grab a real #N registration of the name. But what I don't get is this: Suppose TWO or more sites conditionally register the SAME name, by being mentioned in two distinct map entries for sites who've never heard of each other. Doesn't pathalias get terminally confused? In fact, won't mail to either site get randomly mixed depending on how the sending site costs-out the path to one map reference versus another? Say, for instance, I feed my brother-in-law's site 'spungg' in NYC, and just mention it on my maps: spungg(DEMAND). Meanwhile someone in Oregon feeds his proctologist's machine which is similarly named 'spungg' and HE says: spungg(DIRECT). Neither machine has a #N entry. Now to pathalias, these are the same machine, right? If someone explicitly routes through mysite!spungg or hissite!spungg and is lucky enough to avoid rabid rerouters, it might work, but otherwise my brother-in-law will get some of the proctologist's mail and vice versa, depending on which path looks cheaper to the sender. That's why I don't think 'conditional registration' ought to be allowed. If pathalias sees a node name mentioned in a neighbor-list, but has no #N entry for it, it ought to report it out separately as an error, methinks. I haven't quite figured out whether smart-host affects this. -- "We must never forget that if the war in Vietnam \ $ Tom Neff is lost... the right of free speech will be X tneff@bfmny0.UU.NET extinguished throughout the world." -- RN 10/27/65 $ \ uunet!bfmny0!tneff