query - looking for references

Karen Rosin Sollins <sollins@lcs.mit.edu> Wed, 26 February 1992 03:23 UTC

Received: from nri.reston.va.us by NRI.NRI.Reston.VA.US id aa03669; 25 Feb 92 22:23 EST
Received: from bells.cs.ucl.ac.uk by NRI.NRI.Reston.VA.US id aa03665; 25 Feb 92 22:23 EST
Received: from ZIPPY.LCS.MIT.EDU by bells.cs.ucl.ac.uk with Internet SMTP id <g.00817-0@bells.cs.ucl.ac.uk>; Wed, 26 Feb 1992 02:00:04 +0000
Received: by zippy.lcs.mit.edu id AA02942; Tue, 25 Feb 92 21:01:05 -0500
Date: Tue, 25 Feb 1992 21:01:05 -0500
Message-Id: <9202260201.AA02942@zippy.lcs.mit.edu>
From: Karen Rosin Sollins <sollins@lcs.mit.edu>
Sender: sollins@zippy.lcs.mit.edu
To: osi-ds@cs.ucl.ac.uk
Subject: query - looking for references

I realize this is off the various topics under discussion, but I
thought this might be a good group to answer this question.  A student
came to me looking for one or more references that make the argument
that in information retrieval not only is identification based on
attributes important, but that there is a real need for unique
identifiers for accessing information as well.  Of course I'm not
saying anything about what those unique identifiers are.  The real
issue is their existence.  For this particular student the issue has
come up in the context of Z39.50, which does not deal with unique
identifiers.  Of course there are a number of good reasons for this,
but I couldn't think of any good published references that make that
case.  So, I'm looking for references not the arguments themselves.

Since this may cause a flood of messages, send your responses to me,
and I'll collect them and send them on if there is any interest.
			Karen Sollins