Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!triceratops.cis.ohio-state.edu!karl From: karl@triceratops.cis.ohio-state.edu (Karl Kleinpaste) Newsgroups: comp.mail.uucp Subject: Re: Domain Registration Message-ID: Date: 6 Jun 89 19:50:04 GMT References: <7589@bsu-cs.bsu.edu> Sender: news@tut.cis.ohio-state.edu Distribution: na Organization: OSU Lines: 31 In-reply-to: dhesi@bsu-cs.bsu.edu's message of 5 Jun 89 15:23:52 GMT dhesi@bsu-cs.bsu.edu writes: I was essentially saying, "if you want to get mail to a BITNET host, just give it to a gateway and let BITNET be 100% responsible for delivering it." That is exactly what is wrong: Asking mail users to know the network connectivity underlying their correspondence. I expect a mailer to hand a BITNET-bound piece of mail to some BITNET gateway, and for that gateway to be 100% responsible for safe delivery. But I, as a mail user, don't expect to have to know that such a gateway was needed. If you send mail to me, you address it in geographic hierarchy. You don't know how the planes, trucks, and humans push it around, and you don't care. The system knows its own internal connectivity, and does the work for you without intervention by either the sending or receiving human. If you send email to me, you address it in organizational hierarchy. You don't know how the networks, routing gateways, and phone lines push it around, and you don't care. The system knows its own internal connectivity, and does the work for you without intervention by either the sending or receiving human. I don't have to know the name of the mail delivery person who goes down your street every day. But for some reason, you expect me to know the name of the mail delivery network which connects your system to the rest of the world. The two are exactly analogous. --Karl