RS Preliminary LMS Migration
Planning Group Minutes
June 28, 2002
The main topic of discussion was the test data load reviews. We also shared information from ALA Annual and various local migration meetings.
We were pleased to have Barbara Gundersen, Gerald Langford, and Cathy Mook join us for the first time! Rich Bennett also took time out of his busy schedule to participate in the meeting - thanks Rich!
Information gathered at ALA Annual
During ALA Annual, Naomi spoke with Ruth Haas, a colleague at Harvard, about their activities during the system downtime, which include:
- Training activities. Staff are required to document that they have completed a certain number of hours of training
- Searching backlogged materials for copy (or improved copy).
- Some new work is being done in the test database as if it were live. While the staff recognizes that it will need to be redone, they decided that doing actual work in the database was a more meaningful activity than making up data
- Serials cataloging work is being done in OCLC; manual worksheets are being used to track the work done so it can be updated when editing is possible
- Check-in is being noted by hand on notebooks with print-outs of the former acquisitions record. This data will be transferred when the system is live
Update from 6/27 Intro to ALEPH session
Resource Services Training Committee organized a session to introduce department staff to ALEPH in preparation for the data load reviews. It was very well attended. Staff were encouraged to load the 15.2 client onto their computers and try some of the searching techniques described.
Update from 6/27 UF ALEPH Steering Committee
Rich explained that the first data load, scheduled for sometime in August, will only include the bibliographic and holdings records.
The SUL/CC OPAC and Indexing Task Forces met in Tallahassee in early June. Their final reports are due in the fall.
The train the trainer sessions are still in the process of being scheduled for January and February 2003.
Update from 6/24 meeting re. Documents unlinked
Jan, Martha, Carol, Lawan, Priscilla, and Betsy met to discuss Documents unlinked item records, which will convert to brief searchable records in ALEPH. Lawan reported that as of January 2002, there were 13,480 Documents unlinked item records. Carol has been hard at work doing clean-up on these records (i.e., linking where records are available, correcting typos). The general feeling is that we are in good shape regarding the Documents unlinked item records. Jan mentioned that her biggest concern right now is how the MARCIVE tapeloads will be handled. Martha said she will bring the tapeload concern to the Technical Services Planning Committee (TSPC) advocating for separate MARCIVE loads for each institution. Jan is going to talk to Jim Noel at MARCIVE re. pricing. Martha will also bring to TSPC's attention the need for other institutions to address the issue of unlinked item records.
Categories of records for test data loads
In addition to the local test record file we've been creating during the last year, Lydia instructed us to identify categories of records to add to the test data loads: "To make up the remainder we can use ranges of NOTIS record keys, or simple boolean searches. You'll need to target some 'problem records', or records from a particular source that might be hard to convert, or older records, serials records, etc. Ranges of NOTIS record keys would help there. FCLA can randomly select records to make up the rest of the 75,000."
The group reviewed and revised a draft list as follows. Some of them will need to be limited by date or some other limiting feature in order to retrieve a smaller pool of records.
- active serial records
fi fmt=per and dt=1990- (8,336 records)
- other serials
fi fmt=per and dt=9999 (27 records)
- records in non-books formats
fi fmt=cf (computer files) (30,139 records)
fi fmt=vid (visual materials) (10,350 records)
fi fmt=mus (music) (27,696 records)
fi fmt=map (maps) (38,533 records)
fi fmt=ser (serials) (144,002 records)
fi fmt=mix (mixed materials) (844 records)
- NOIS records
fi no=source nois (6,714 records)
- Blackwell records
fi no=source bn and dt=1985 (1,027 records)
- Carollton Press records
fi no=carp200101 (2,632 records)
fi no=carp200102 (2,461 records)
fi jx=bible (30,029 records)
fi jx=shakespeare (10,973 records)
- MeSH headings
fi jx=neoplasms (4,456 records)
fi jx=abortion spontaneous (25 records)
- Diacritic blank spaces
fi no=source c81269 and lng=cze (76 records)
fi no=source c81269 and lng=pol (53 records)
fi k856=http (47,103 records)
fi k=cob2s (24 records)
- United States
fi k110=united adj states (231.623 records)
fi no=govdoc20000116 [or some other date] (758 records)
- Music records
k=beethoven symphon? (254 records)
Gerald suggested that we ask FCLA to load the test records into the test database in the following order: 1) smp file (individual records in LTUF production and nercicst) 2) categories based on search strings and 3) additional records chosen by FCLA. In so doing, the FCLA chosen records should fall into an easily retrievable NOTIS record range.
Workflow for data load reviews
Betsy reported that Phek has been working with FCLA and Systems to automatically generate a spreadsheet from our LTUF production and nercicst test record files, which will be used to parcel out the records and note problem reports. Betsy distributed a mock-up of the spreadsheet.
Staff will be asked to review each of their assigned records based on a checklist giving specifics about what kinds of searches to do and what kind of information to look for. This checklist will necessarily vary according to the types of records being reviewed, however, it will be important for staff to view all their assigned records in ALEPH's Public Web view. If staff want to search by LUIS number in WebLUIS to compare to ALEPH's Public Web view, Rich explained that they should use the following search in the command option: rk=[LUIS#]
The group decided to use a Web form for reporting purposes. Staff will be asked to fill out the Web form as soon as problems are discovered. The Web form will allow staff to choose from among different types of problems in order to standardize the reporting language as much as possible, but there will also be a comment area so that staff can elaborate as desired.
The Web form will be submitted to a staffperson who will update the spreadsheet. The spreadsheet will be posted on the Web and refreshed daily (or as needed). A small group of managers (Betsy, Michele, and others to be determined) will review the problems, consolidate, and report them forward to Rich for sharing with Ex Libris.
Rich recommended that the group find out how other institutions have handled the test data load review process.
Review of Tasks as outlined in May 16, 2002
Naomi reported that CAGER did discuss migration related to LTQF, but FCLA responded that they have not given any thought yet to how LTQF will be affected.
Rich said that title level notes will be migrated to the bib record and that this will happen during, not before, migration.
Members of the group have asked FCLA about the possibility of leaving parts of NOTIS active during the system downtime, but there is no word whether or not this will be possible. Barring that possibility, Nancy suggested that we might see if FCLA could create a partition in nercicst for the active serials to allow serial check-in to continue in a familiar interface for later rekeying in ALEPH.
Due to vacations, the next meeting is scheduled for Wednesday, July 31 instead of sooner. The group will try to meet more frequently beginning in August.
E-mail Lydia categories of records for test data loads (Betsy)
Develop spreadsheet in conjunction with FCLA and Systems (Phek)
Determine checklists for record review (All)
Create draft Web form to share with group (Betsy)
Identify staff to receive Web forms, revise spreadsheet, and post to Web (Betsy, Phek, Michele)
Contact other institutions re. data load review process (Betsy and others)