Z39.50 Report from the Distributed Indexing/Searching Workshop
Reported by Ray Denenberg (Library of Congress)

      At the Distributed Indexing/Searching Workshop there was considerable
interest and discussion about the role Z39.50 could or should play as a web
protocol for information retrieval. There were dissenting views, but most felt
there should be some standard information retrieval protocol deployed, at
least for queries, and most thought that Z39.50 was the proper choice, since
it exists and there has been no alternative protocol proposed; if Z39.50 were
rejected it would have to be re-invented. 
      Several attendees felt, however, that Z39.50 is too large and complex,
and it addresses a much wider scope of information retrieval functionality
than is needed for web applications. 
      The workshop Chair issued a challenge to the Z39.50 community to develop
a rendering of Z39.50 that is acceptable to the web community; that is,
develop a Z39.50 document that web developers will be willing to read -- a
self-contained document specifying Z39.50 profiled to some agreed-upon,
relevant subset of the full standard. The Library of Congress representative
accepted this challenge on behalf of the Z39.50 Maintenance Agency and on
behalf of the Z39.50 implementor community.
      In the near-term (1-3 months) there will be an effort to agree upon a
relevant set of information retrieval requirements appropriate for
standardization within a web information retrieval protocol. Subsequently a
Z39.50 profile will be developed, which will be a rendering of Z39.50 as
described above. Currently, the code name for this effort is "Z39.50 lite".