Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!ll-xn!ames!scubed!piaget!jc From: jc@piaget.UUCP (John Cornelius) Newsgroups: comp.mail.headers Subject: Re: Trouble with From: lines Message-ID: <112@piaget.UUCP> Date: Thu, 4-Dec-86 12:03:41 EST Article-I.D.: piaget.112 Posted: Thu Dec 4 12:03:41 1986 Date-Received: Fri, 5-Dec-86 04:49:04 EST References: <597@cullvax.UUCP> <1353@munnari.oz> Reply-To: jc@piaget.UUCP (John Cornelius, System Guru) Organization: International Widget Lines: 39 In article <1353@munnari.oz> kre@munnari.oz (Robert Elz) writes: >In article <597@cullvax.UUCP>, news@cullvax.UUCP writes : >> Incoming ARPA messages used to come to us with headers saying things like >> From: arpa-feed!user@arpa.site >> which worked fine--one could reply to it, etc. Recently this has >> changed (the "arpa-feed!" part isn't added) >> .... >> It used to work and now it doesn't. Am I being unreasonable? > >Yes, I think you are. To do what you want properly, your feed site >would have to know the characteristics of every site it feeds (and really, >every site, everywhere) and generate addresses in the form that that >site likes. That's absurd. Much better is to have a standard, and >have everyone comply with that. > >If that means that some sites have to upgrade their software, then >that's just the way it is. Reconcile yourself to this. > >Robert Elz postmaster@munnari.oz While all that you claim may indeed be reasonable to you, please note that many of us do not have the option of implementing a mailer that is not supported by the equipment or software vendor. Even for those of us who do have access to sources and/or considerable support talent, there is considerable time and effort involved in converting from one standard to another. It is also the case that the learning curve is often flatter than we would wish. For the most part, the news system has provided transition tools wherever possible and I have considerable admiration for the manner in which updates and improvements have been handled. On the other hand, for a postmaster to tell a client that he should refer to Figure 1. seems to be out of character for the net in general and the postmaster function in particular. I would think that the postmaster would want to get the problem fixed as promptly as possible. It isn't so much that he should serve the client as it is that the problem is likely consume his time anyway. John Cornelius (...!sdcsvax!piaget!jc)