Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!iuvax!purdue!decwrl!vixie From: vixie@decwrl.dec.com (Paul A Vixie) Newsgroups: comp.mail.uucp Subject: Re: Pathalias and routing of mail Message-ID: Date: 13 Dec 88 11:31:34 GMT References: <1144@tank.uchicago.edu> Sender: vixie@decwrl.dec.com Organization: DEC Western Research Lab Lines: 30 In-reply-to: matt@oddjob.uchicago.edu's message of 12 Dec 88 20:03:25 GMT I understand that various people have me in their KILL files because they don't care about rerouting. May their mail be rerouted to Italy because of their intentional ignorance and lack of participation :-(. [Crawford] # I stand by my practice of jumping ahead when 'c' is fully qualified. # Since I am at an internet site, my route to 'c' ought to be almost as # up-to-date as humanly possible, Unless it's an MX that's on the other side of some slow IP or frequently congested IP links or mailbridges, in which case the UUCP path you were given could be better. # (Also, I think that jumping to a fully-qualified 'c' does More Good Than # Harm - i.e., it wins in more cases than it loses.) But when it loses, it loses BIG. And the wins? The problem peeking or any kind of RE-routing is designed to solve can be better solved at the source -- write good source routers and teach news software, news admins, and news users how to use them. This is mostly done already, and if anyone has effort for solving problems caused by replies which travel over reversed news paths, I humbly suggest that they spend that effort making the problem go away rather than making room for unpredictable lossage further downstream. And for the record, it isn't clear to me that it _is_ a win more often than not. -- Paul Vixie Work: vixie@decwrl.dec.com decwrl!vixie +1 415 853 6600 Play: paul@vixie.sf.ca.us vixie!paul +1 415 864 7013