Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!umich!samsung!brutus.cs.uiuc.edu!psuvax1!psuvm!RICEVM1!SCHAFER From: SCHAFER@RICEVM1.RICE.EDU (Richard A. Schafer) Newsgroups: bit.listserv.policy-l Subject: Re: User interface for list postings Message-ID: <900131.094628.CST.SCHAFER@RICEVM1.RICE.EDU> Date: 31 Jan 90 15:46:28 GMT Sender: Discussion about BITNET policies Reply-To: Discussion about BITNET policies Organization: Networking and Computing Systems, Rice University Lines: 43 Approved: NETNEWS@PSUVM Gateway In-Reply-To: Message of Tue, 30 Jan 90 13:55:24 EST from On Tue, 30 Jan 90 13:55:24 EST Michael Johnson said: >Software that might fall under the auspices of BITNIC would include such >things as Mail Transfer Agents, Mail User Agents, file servers, NEWS servers >and NEWS interfaces. When BITNET II hits the horizon, there will be other >things that need to be standardized across the net. Some of this software >would be required (such as the MTA) and some would be optional (the MUA). >As it is now, we have a horrible hodge-podge of (for example) up-to-date >MAILER servers, back-leveled MAILER servers, and no MAILER servers at all. >Sites that choose not to run MAILER can justify it on the basis that MAILER >support may not continue in the future. If support of the BITNET MTA was >assured for as long as there existed a BITNET, then that justification would >go away. Support of the MTA and other software could be a full-time job for >whoever is contracted/hired to do it, rather than being fit in among the other >duties of busy systems programmers. I'm glad to see someone raise this issue again. While I'm certainly personally committed to supporting the code I write and distribute *now*, there's no way I can possibly guarantee that I won't (heaven forbid) walk in front of a cement truck on my way out to the car tonight, in which case, who would take care of the RiceMail code? I have recently had at least one site say that they were leery of installing anything like the Mail command on precisely those grounds, and I encourage people with those kinds of valid concerns to bring them up before the board. As long as BITNET depends in a major way on programs written (and owned) by individuals without institutional support committments, then there is a major risk involved that at the very least needs to be recognized, if not dealt with. The only way that anything like BITNET-supported software is going to exist is for the membership to demand it and for the board to agree that this is a good idea. You'd need negotiations with current software owners should BITNET decide to try to obtain ownership of the existing widely used software. You'd need some network-wide funding committment, I would think, to provide for the software support structure you're talking about wanting. You'd need a *lot* of decisions about what software was needed, and we all *know* how hard that is to achieve. Am I pessimistic? Perhaps. But I think the only way to proceed is to start talking about the subject. Richard