Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!mstar!mstar.morningstar.com!bob From: bob@MorningStar.Com (Bob Sutterfield) Newsgroups: comp.mail.misc Subject: Re: mail servers Message-ID: Date: 24 Aug 90 21:30:21 GMT References: <1990Aug17.021921.1863@chinet.chi.il.us> <1594@i2ack.sublink.org> <1990Aug23.200043.17259@chinet.chi.il.us> Sender: usenet@MorningStar.COM (USENET Administrator) Reply-To: bob@MorningStar.Com (Bob Sutterfield) Organization: Morning Star Technologies Lines: 59 In-Reply-To: les@chinet.chi.il.us's message of 23 Aug 90 20:00:43 GMT In article <1990Aug23.200043.17259@chinet.chi.il.us> les@chinet.chi.il.us (Leslie Mikesell) writes: In article bob@MorningStar.Com (Bob Sutterfield) writes: A mail-based archive server is an impolite and destructive way to distribute large amounts of information, and should be discouraged in such applications. ...if this message is in response to my request for information about mail servers, it doen't really apply. Yours prompted it, but it's really in response to the general trend. What I want to do is to set up something resembling a library service within our organization to help avoid duplication of effort by various branches. The transport is already established and would be unlikely to send anything beyond our own offices. You're right - this is unrelated to the subject of my diatribe (:-) since it doesn't involve other people's money. In fact, it sounds like a pretty good idea! It seems like this is something that every group with an established electronic mail service would need, so I'm kind of surprised that the current crop of programs seem pretty much oriented towards distributing program sources. People just haven't generalized the idea yet. I've toyed with the idea of using RFC822 envelopes to carry, e.g., SQL queries and responses... Any number of schemes are possible, usually based around a reliable file transfer protocol. How would you like to train a few hundred people to adapt to "any number of schemes"? ... they already know how to use a mail program for other reasons. They shouldn't need to adapt to anything they don't already know how to use. I was saying that, no matter what users are already using, a user-compatible scheme could probably be hacked together. Virtually every other question of human time vs. computer resources these days is answered with "computer resources are cheap". They're particularly cheap if someone else is paying for it :-) In fact, Steven Wolff said about NSFnet bandwidth, something along the lines of `crunch all you want, we'll make more'. If mail transport of bulky items is a problem, then perhaps we should fix the problem rather than avoiding it. Telebit Trailblazers and other high-speed modems; advanced compression techniques; the increasing pervasiveness of the Internet, BITnet, SPAN, and friends; and other factors are rapidly bringing transport costs down. But only one of those types of factors provides zero-incremental-cost data transfer. There are still a lot of folks out there who pay Ma Bell (or whomever) for every minute they use slinging bits hither and yon. And if those bits belong to someone else, they are slung out of either ignorance or the goodness of their hearts. In either case, let's not take advantage of the hospitality.