Summit Catalog Committee Duplicate Record Reduction Working Group (SCC-DRWG) Notes from teleconference call, 2006-12-01
Attending: Marcia Bianchi, Reed; Michael Boock, OSU; Diana Brooking, UW; Carol Drost, Willamette; Tom Larsen, PSU; Nancy Nathanson, Alliance; Friday Valentine, OHSU
Absent: Daniel CannCasciato, CWU
Admin notes;
> Next mtg (teleconference), Friday 12/15 1-2pm [Susie S. will send
> email; Friday V. will start as Nancy N. is out] Our summit-cat
> committee home page link is:
> http://www.orbiscascade.org/staffhome/SCC-DRWG.htm
> Our wiki home page link:
> http://wiki.library.oregonstate.edu/confluence/x/NC0
Homework/Action Items:
> Discuss how we want to use wiki (we forgot to do this 12/1).
> Each person should post a word doc to the wiki with examples & catalog record [or enough fields to be useful], especially if those aren't already on the examples document. (post to wiki by 12/11 so we get a chance to read others examples). List each set and our recommendations for each type.
> The goal of the group was decided: Develop general principles document
> and generic overview/process for future packages & for loading
> non-sets. (like contact Alliance staff to set up the prefix and others
> need to follow it)
General Discussion (I left my notes in their fairly freeform, yet sequential state. FKV):
Various packages folks have:
--Books24x7: OHSU
--Evans early american imprints (Carol)
--18th c. ecco from UW is going to summit (Diana) --History ebook project (reed) --Netlibrary( Willamette )
Maybe do a larger survey of who has what packages? no, Currently go with what we (the members) have.
What do you need to tell your vendor(s)? Costs of customization?
Non-OCLC records are the bigger prob.
Collection level records & part (item) level records (PSU digital collection) What about merging brief records? (and overlaying) Larger issues of who is contributing what (PSU reports not submitting everything, and stripping some 001 fields from vendors)
Category: records where records don't have 001 in the first place or where 001 has been stripped and can't be recovered
Looked at http://www.orbiscascade.org/staffhome/ebooks-discussion1.htm
Summit looks first at 001 field, if one is not found it creates one with prefix 2nd check is 020 z9tes0 b2734758 institution code followed by bib#
001 bks00004336 vendor pattern
We are trying to fix records coming in as well as current dupes
Load table training, forcing folks to make load table changes that will cost them money
What about putting up training materials online regarding standards we adopt to help those w/o load table twitching (and if you can't comply don't submit those to Summit)
Perhaps we need to talk to III about having a central person/consortium person getting load table access for those institutions without a person who has load table training (with constraints on it should benefit summit)
Folks using MarcEdit to pre-massage records. Create training for a MarcEdit loader.
Everyone needs to have a loader that will handle alphabetical prefixes
First concentrate on full marc records, then look at ERM brief records. (Janet Crum has written a description for ERM fix: press III to construct 001 field based on ISSN with prefix, III should take care of this [part of report from last year] http://www.orbiscascade.org/staffhome/SCC_recommendations-06jun-final.htm)
Tom suggested Draft a report with examples and see
Decision/homework: Post a word doc to the wiki with examples & catalog record, especially if those aren't already on the examples document. (post to wiki by 12/11 so we get a chance to read others examples). List each set and our recommendations for each type.
Decision/homework: Develop general principles document and generic overview/process for future packages & for loading non-sets. (like contact Alliance staff to set up the prefix and others need to follow it)
Each institution needs to ask vendors to create consistent unique numbers (not accession numbers).
*----------------------
Please send correction/addtions to Friday V. at valentif@ohsu.edu.