LITA MARC HOLDINGS INTEREST GROUP
                    Saturday, 12 January 1991

     A handout containing the proposed agenda for the
Preconference on the MARC Holdings Format, scheduled for Atlanta
on Friday, 28 June 1991, was distributed.  (Appendix B)  The IG
is also planning a joint program with the LITA/ALCTS
Retrospective Conversion Interest Group on the recon of holdings
on Sunday, 30 June 1991.

     Much of the remainder of the meeting was devoted to
consideration of proposals relating to the MARC Format for
Holdings and Location, e.g., Should the subfield 'o' of the
854/855 and 864/865 fields be made obsolete now that there is an
844 field which contains the designation of the unit being
described; the consensus was No.  Another question considered was
the current limitation in subfield 'y' of 853-855 fields; as
currently designed it cannot handle items issued on a particular
month and day, e.g., an item issued on the 1st and 15th of the
month will have 24 issues a year; this is not the same as a bi-
weekly frequency which results in 26 issues a year.

     Sally McCallum (LC) said that NISO will work this year to
clear up the discrepancies between the MFHL and NISO Standard
Z39.57 on holdings.

     The second half of this meeting was devoted to a January
1991 proposal for a shared pattern database that Linda Miller of
LC and Bonnie Postlethwaite of Tufts had drafted. (Appendix C)
Opinion diverges on the question of building a database of
current publication patterns vs. including a pattern history.
Someone asked if the Holdings format can be used for prediction
and claiming purposes; the answer was sometimes Yes and sometimes
No.

     Among the decisions to be made are the means by which data
on publication patterns could be transmitted.  Timeliness is
essential.  Tape is probably too slow.  Online, perhaps over the
Internet?  Velma Veneziano of Northwestern University suggested
that publishers should incorporate this information in the issues
of their journals.  Perhaps multiple claims of non-existent (but
expected) issues might give them an incentive to provide the
pattern information and to keep it both up-to-date and accurate.

     CONSER has decided to set up a Task Force (chaired by Barry
Baker of Georgia) to see what role CONSER might play in a shared
patterns project.  Text of a survey of possible participants in
such a database was included as part of the draft proposal;
perhaps the various local system vendors or their user groups
could distribute it to the users of the various systems.