Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site bu-cs.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!qantel!dual!lll-crg!gymble!umcp-cs!seismo!uwvax!harvard!bu-cs!root
From: root@bu-cs.UUCP (Barry Shein)
Newsgroups: net.mail
Subject: Mail Addressing Solved
Message-ID: <695@bu-cs.UUCP>
Date: Wed, 2-Oct-85 22:27:07 EDT
Article-I.D.: bu-cs.695
Posted: Wed Oct  2 22:27:07 1985
Date-Received: Sun, 6-Oct-85 05:29:50 EDT
References: <134@mck-csc.UUCP>, <2843@sun.uucp>
Organization: Boston Univ Comp. Sci.
Lines: 117

I sent something like this to Human-nets, I'll try here also for reactions:

		     A SOLUTION TO MAIL ADDRESSING

Ahem, I propose we adopt a system that seems to have worked quite well
in the past, I suspect this is not original, but that has never stopped
anyone from posting an opinion on this net before, I refuse to be the
first. Why don't we just use PHONE NUMBERS? Yes PHONE NUMBERS, you heard
me. They don't have to be modem numbers, but they should be real phone
numbers (tho I guess there should be allowance for anything.) Before I
try to justify this, an example:

	"Barry Shein"@617-353-2780

That's the # for the front desk at the computing center here, I don't
care if I use my full name quoted or 'bzs' or 'root', not the issue
here.

If I am in the Boston area I can use:

	bzs@353-2780

and if I am at B.U. I can probably get away with:

	root@3-2780

Of course, my mailer has a simple list processor for aliasing hosts,
so I have created a file with a line:

	host	bu-cs	617-353-2780

And now if I like I can type:

	bzs@bu-cs

KEEP READING!

Ok, why?

	a) There is already a naming authority: TPC

	b) If I choose a real phone for my site it
	is guaranteed to be unique (I know, some phone
	phreak out there is already typing in a case
	that happened in Missoula in 1914 where two people had
	the same phone number)

	c) The defaults are well known (leave out an
	area code, it is local etc.)

	d) I CAN PROBABLY FIND OUT YOUR HOST ADDRESS BY
	DIALING (XYZ) 555-1212, especially if some reasonable
	convention for which phone in your organization should
	be used in general (at worst it's the same as now, you
	have to ask someone.)

	e) Even children (the lowest known form of life :-) can
	manage to use the phone system.

	f) By and large, through use of international codes, even
	international addressing has largely been solved.

	g) Large organizations can (and often do) have mnemonic
	phone numbers, if I want some info on new unix stuff I
	can probably try:

		unix@800-ATT-UNIX

	for example (or 800-ATT-UNIX!unix, that's not the issue!)

I'M SERIOUS (get back here!)

Anticipated objections:

	Strings of digits are not user-friendly:
		tell it to AT&T, neither is root@a.cs.bu.edu
	Do you mean use the phone system exclusively?
		no, just as a way to form names
	What about typos and 'wrong numbers'?
		what happens now that's so wonderful?
	What if AT&T changes it's numbering conventions,
	like going to 718 in the Queens/Brooklyn Area?
		you'll change also, you'll survive
	Who's going to act as the 'switch' for exchanges?
		who does now? this makes no attempt to propose
		how mail gets delivered or routed, just that it
		might provide a sane base on which to solve those
		problems
	You mean if I have a micro at home I just use my home #?
		why not?
	what if I don't want people to know my home number?
		I can't answer everything, stay off the net
		or arrange for only your feed to recognize
		some other number as a 'special' for you to
		publish, c'mon, this aint that tough. Or get a
		phone installed just for this and shut the ringer
		off, who cares really, privacy always costs.

	Does that mean an organization in the phone book could have
	something like:

	BOSTON UNIVERSITY
		...
		Computer Science		353-8919
			E-MAIL			353-8919

		Why not?

	Won't the phone company object?
		ask the phone company, I doubt it.

	Ok, what's the catch?

		It's too obvious, that's not what mail addressing
		is all about.

			-Barry Shein, Boston University