MARBI report / Diane Hillmann dih1@cornell.edu

ALCTS/LITA/RUSA Machine-Readable Bibliographic Information Committee (MARBI)

ALA ANNUAL CONFERENCE, June 26-28, 1999 New Orleans, LA

Report in agenda order.

Saturday, June 26, 1999

1. Discussion Paper No. 114: Seriality and MARC 21 [http://lcweb.loc.gov/marc/marbi/dp/dp114.html]. This paper presented three areas of MARC 21 that could potentially be affected by the introduction of seriality-related changes to the Anglo-American Cataloguing Rules, 2nd ed. Rev. (being considered by CC:DA at this meeting, and not yet approved) recommended by Jean Hirons in her paper "Revising AACR2 to Accommodate Seriality [http://www.nlc-bnc.ca/jsc/ser-rep0.html]."

The first of these recommendations hinged on the division of the world into two types of publication: 1) Finite: those that are complete or intended to be completed (e.g., monographs, static databases) and 2) Continuing: those having no predetermined conclusion (e.g., serials, loose-leaf for updating publications, additive databases, etc.). The potential effect centers on Leader/07 (Bibliographic level) and potentially 008/21 and 006/04 (Serial type). Current codes m (monograph) and s (serial) exclude a broad category of bibliographic resources, referred to as "integrating resources" in the JSC report. Currently most are being coded as 'm' and are not treated as serials.

Three options were defined:

Option A. Least change approach, which involved retaining codes 'm' and 's' in Leader/07 (Bibliographic level) as currently defined, and dividing up integrating resources based on a determination of which integrating resources are most like monographs and which are most like serials. Additionally, a code might be added to the serial 008/21 (Serial type) for databases and one to 008/18 (Frequency) for "continuously updated".

Option B. Expand code 's' in Leader/07 to cover all continuing resources (any items with no predetermined conclusion, also integrating resources). New codes might be added to 008/21 to include loose-leaf , database, web site or a single code to indicate integrating resources.

Option C: Define a third bibliographic level, using code "i" for integrating resources, regardless of whether they are finite or continuing. This might involve either creating a new 008 to be used with code "i" or using the serials 008.

Responses from MARBI were variable on these options, with no real consensus that major change was necessary. Several people made the point that options B & C could result in serious problems in grouping data for search results, since older data would be done one way and newer data another. Others questioned whether users would understand the new definitions any better than the old, and whether the potential chaos resulting from major changes would justify the costs involved.

The second recommendation involved the 260 field suggesting that it be made repeatable so that the current publisher and place could be added to the record in addition to the original publisher/place, mandated by the serials cataloging rules. Several options were suggested for specifying which 260 were the current and which the original, including the use of $3 or the addition of indicator values (with $3 perhaps used to show the dates of changes).

In general MARBI reacted positively to this recommendation, tending to prefer the use of indicator values rather than $3, and leaving open the question of whether any intervening places or publishers could be included. Some utility representatives expressed concern about matching and de-duping algorithms as well.

A third recommendation involved a change in the coding for bytes 008/34 and 006/07, specifying under what rules a serial was cataloged. Specifically, this would mean that integrating resources and some electronic journals might be cataloged according to conventions similar to latest entry, as applied prior to AACR. Since latest entry records for successively-issued would serials still exist, there was a suggestion for an additional value be added to the 008/34 and 006/07.

Very little discussion occurred around the last recommendation, and no action was taken on the discussion paper, as decisions regarding rule changes have yet to be made.

2. Proposal No. 99-08 [http://lcweb.loc.gov/marc/marbi/1999/99-08.html]: Defining URL/URN Subfields in the MARC 21 Bibliographic Format. This proposal suggested the definition of subfields to contain URLs and URNs in bibliographic format fields other than field 856. These sorts of proposals have met with mixed reaction from MARBI in the past, after an initial flurry of enthusiasm. The fields included in this proposal were: 555, 583, 76X-78X.

The 5XX links were requested by the archival and preservation communities, respectively, and were carried. The 76X-78X links were more problematic, there being fewer open subfields left in those blocks, and no possibility of using the same one across all relevant fields. The CONSER Task Group on Journals in Aggregator Databases, which initiated the proposal, suggested that these links were necessary, particularly in field 773 (Host Item Entry), to link back to the aggregator site, in addition to the 856 $u for the specific journal. Several comments were made concerning possible problems with this approach, considering that most institutions were scripting access to these sites in any case because of the necessity for users to access through proxy servers or because of passwords and other security devices. Including (and potentially displaying) the aggregator URL on the same record would be confusing, and, as it was not necessary at all for the Task Force's proposed batch uses of aggregator records in general, MARBI did not approve that portion of the proposal.

3. Proposal No. 99-09 [http://lcweb.loc.gov/marc/marbi/1999/99-09.html]: Making Field 852 Subfields $k and $m Repeatable in the MARC 21 Holdings Format. This paper proposed making subfields $k (Call number prefix) and $m (Call number suffix) in field 852 (Location) repeatable in the Holdings Format. This would allow for recording more than one call number prefix or suffix and being able to parse them in situations where there are two pieces of information to record about the location of an item within a call number run. As this was discussed rather thoroughly at the DP stage, it was approved quickly with little discussion.

4. Discussion Paper No. 116 [http://lcweb.loc.gov/marc/marbi/dp/dp116.html]: Bound-With Relationships in the MARC 21 Holdings Format. This paper [essentially written by me] suggested a technique for linking together separate bibliographic entities that are bound together. It involved using the Item Information fields 876-878 subfield $p (Piece designation) in the MARC 21 Holdings Format in separate holdings records. This allows for a cleaner presentation and more flexibility than the previously proposed technique of repeating field 004 in a single holdings record, which I felt blurred the distinction between bibliographic and physical relationships.

Discussion on this was reasonably brief, as it had been circulated in draft to MARBI previously, and it required no real changes to the format, just additions to the documentation and examples. A few discussants proposed that either this or the 004 technique could be used, depending on the situation, but as this would create problems with exchange of data, it was judged undesirable. Some comments were made from the DRA representative, with concerns about how the technique would work in their TAOS system which uses the same fields to link to actual item records. Some suggestions were made for alternatives in that situation, and the technique was approved.

Sunday, June 27, 1999

1. Discussion Paper No. 115 [http://lcweb.loc.gov/marc/marbi/dp/dp115.html]: Anonymous Artist Relationships in the MARC 21 Bibliographic Format. Works which cannot with certainty be attributed to a known artist are common in the art world, and art historians express gradations of certainty about the relationship between a work and a known artist or group of artists by coupling a known artist's name with a qualifier. Librarians and visual resources professionals who use the MARC 21 format to catalog art works or surrogates of art works need to be able to record this information in an appropriate place on the MARC bibliographic record. This paper discussed several different options for handling anonymous artist relationships: use subfield $c (qualifier), use field 720 (Uncontrolled Name), or define a new subfield.

This proposal generated perhaps more discussion than any other this meeting, both online prior to the meeting and during it. Much of it centered around whether or not these names should be subject to authority control -- there seemed less controversy in general about how to reflect the names themselves.

LC will write a proposal with option 3.

2. Discussion Paper No. 117 [http://lcweb.loc.gov/marc/marbi/dp/dp117.html]: Coding non-Gregorian Dates in the MARC 21 Bibliographic Format. This paper considered the need to code publication dates from non-Gregorian calendars when they appear on items and suggested various coding requirements and options.

After some desultory discussion, it was determined that more consultation was needed with the affected communities. No action was taken.

3. Discussion Paper No. 118 [http://lcweb.loc.gov/marc/marbi/dp/dp118.html]: Non-filing Characters in MARC 21 Using the Control Character Technique. This paper discussed the use of the control character technique to block off nonfiling characters in MARC records. It considered whether the technique should be allowed in any field, rather than a restricted list of fields, and whether it could be applied anywhere in a field/subfield, rather than only at the beginning.

As usual, this proposal generated a great deal of discussion, though this time there was perhaps more consensus than usual. In general, there was not much support for a restricted list of fields, or restrictions on where in a field it could be used. However, there was support for a restricted set of techniques and uses to avoid use for local problems or solutions that would hinder the standard exchanges of records.

LC will work on a proposal incorporating the comments made.

Monday, June 28, 1999, 2:00-4:00 p.m. Hotel Intercontinental, Pontalba Room

Because of a shorter than usual agenda, the Monday block was devoted to a joint MARBI/CC:DA meeting on the seriality proposals. Regina Reynolds from LC presented some of the proposals and participants from both committees commented. Most of the discussion centered on proposals to change the rules for title changes, and prevent the proliferation of successive entry records for minor changes. The most controversy was aroused by a specific proposal for a new use of uniform titles to control title changes.