zben@umd5 (Ben Cranston) (09/13/86)
In article <3945@ut-ngp.UUCP> Werner Uhrig (werner@ut-ngp.UUCP) says: > My argument is that, as long as there is no second site SEISMO on the net > it is an atrocity for people to get slapped in the face by the mailer which > all of a sudden disavows knowing SEISMO. If the intent of this comment is to suggest that the software do automatic disambiguation then it merely postpones the day of reckoning until such time as a second SEISMO appears. Believe me, I maintain a mail system that at one time attempted to do an [Internet, then BitNet] lookup, but was blown out of the water by new Internet sites making existing BitNet sites invisible. The choice would be to bite the bullet NOW or have things be (perceived as) breaking for the next few years. Far better to get it over with NOW. Just in passing, how did Seismo get to be the canonical mail example site? I'll bet Rick just loves it :-) In article <2528@cbosgd.UUCP> Mark Horton (mark@cbosgd.UUCP) says: > ... I would think that on the ARPANET, a mail system that > sees rick@seismo would assume that rick@seismo.ARPA was meant, but > evidently many implementations don't do this. Since my implementation is a fairly new one, and I wanted to look to the future, and perhaps from a bit of local chauvinism as well, I took the "default" domain to be our local one: the UMD.EDU domain. Thus if anybody here types rick@seismo they would be talking about rick@seismo.UMD.EDU which of course does not exist. At the time the idea was "upward compatability" within our local campus, so that people used to typing "mimsy" and "cvl" and such would experience the minimum disruption. My rationale was that the simple case should be the most usual case (intracampus) and if anybody wanted to get off-campus then they should pay the additional cognitive cost of learning domaining. With some experience I can now criticize this decision. Since people on campus do not learn domaining they cannot relay mail addresses to other people off campus. Because "mimsy" works for them, they have never bothered to learn "mimsy.UMD.EDU" and so they cannot communicate mail addresses to their off campus correspondants. On the other (original!) hand, I don't particulary want to have to dial 301 on every local telephone call. If I can learn to add the 301 when I quote my telephone number to out-of-state people maybe there is a chance that the local PHDs can be educated to quote UMD.EDU (but I'm not holding my breath...) -- umd5.UUCP <= {seismo!umcp-cs,ihnp4!rlgvax}!cvl!umd5!zben Ben Cranston zben @ umd2.UMD.EDU Kingdom of Merryland Sperrows 1100/92 umd2.BITNET "via HASP with RSCS"
rick@seismo.CSS.GOV (Rick Adams) (09/15/86)
It appears that "seismo" will be added back to the host tables for the benefit of a few milnet sites who can't handle domained addresses (i.e. they strip everything after the first dot). I fail to understand why they don't require them to update their mailer. (4.1bsd delivermail) I wasn't asked, just informed about it. Another giant step backwards presented by the NIC. Look for NCP coming soon to an IMP near you. Better living through standardization... ---rick
rick@seismo.CSS.GOV (Rick Adams) (09/15/86)
By the way, the machine you get connected to when you ask for seismo.css.gov (10.0.0.25) has been called beno for months now. In a few weeks, seismo.css.gov should return 192.12.141.25 and 10.0.0.25 will point to a dump pdp 11 that doesn't even talk tcp. Those of you who cleverly added seismo as a local alias for 10.0.0.25 will be royally screwed. Those who (correctly) use the distributed hosts.txt or domain servers will be connected to the right machine. Short term kludges often come back to haunt you.... ---rick