Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles; site hpcnof.UUCP
Path: utzoo!linus!decvax!decwrl!Glacier!oliveb!hplabs!hpfcdc!hpfcla!hpcnof!dat
From: dat@hpcnof.UUCP
Newsgroups: net.news.group
Subject: Re: Orphaned Response
Message-ID: <53200001@hpcnof.UUCP>
Date: Sat, 26-Oct-85 18:32:00 EST
Article-I.D.: hpcnof.53200001
Posted: Sat Oct 26 18:32:00 1985
Date-Received: Fri, 1-Nov-85 00:15:25 EST
References: <10523@ucbvax.UUCP>
Organization: 26 Oct 85 14:32:00 MST
Lines: 42


	So...what happens if you do something like this:

	% ln /notes/mod.laser-lovers /notes/fa.laser-lovers

	(or whatever the exact syntax is)

The idea being that since the notes are kept merely as a set of files
each with a number indicating it's ''unique'' message id on the local
system, you could have the 'old dogs' reading the fa.* groups, and the
new people coming onto the system reading the mod.* groups.

	This would also allow a very smooth transition, because you
could keep the directory links for infinite time without ANY overhead
at all (except perhaps a single inode).

	What would this do to the notes / readnews systems??

	Also, how hard is it to have a 'translation' table on the 
gateway of a machine:

	other
	machines	---->      translation 	   ---->	local
	in the	        <----	      table        <----	machine
	world

so that entries like "translate fa.sun-spots to mod.comp.sun" would
work both incoming and outgoing.  (note that this would also allow
systems with greater than 16 (!!) character filenames to have 
MNEMONIC names instead of CRYPTIC names.  They'd just pity their
neighbours..)

	This is perhaps what was refered to as the 'alias' table??


				Leaping into the fray!

					-- Dave Taylor
					Hewlett Packard, CNO

					...ihnp4!hpfcla!d_taylor
				     or ...hplabs!hpcnou!dat