American Library Association
Midwinter Meeting (2005 Jan. 14-17 : Boston, Mass.)
Report of cataloging, etc. meetings
Technical Services Directors of Large Research Libraries Discussion Group (“Big Heads”) (Friday morning)
The new website for Big Heads is at LC. cf. http://www.loc.gov/library/bigheads/bigheads.html Lee Leighton (UC Berkeley) reported on their planning, in conjunction with WESS, for cooperative cataloging. The focus will be on updating vendor records. Bob Wolven (Columbia) reported on browsable access to e-journals outside the opac. Several libraries are using vendor products for controlling titles, including TDNet, SFX, and Innovative’s ERM. Columbia is mapping LCC call numbers to broad subjects. The University of Washington is using III acquisitions data to create subject and alphabetical lists for access to e-journals. Cornell is finding about equal use of browse and keyword indexes. The A-Z list of e-journals is the most used link on their homepage. Catalogers at the University of Virginia add data to cataloging records to produce subject lists, while those at NYPL are created by public services and web development staff. A standard solution would be nice.
Tim Jewell and Joyce Ogburn reported on the DLF initiative to develop electronic resource management software (ERMI). They are describing the appropriate architecture, building a data dictionary, considering XML issues, and involving librarians and vendors as reactors. The following products were mentioned: III, Verdi (Ex Libris), Verify (VTLS), Meridian (Endeavor), Dynix, SIRSI, Serials Solutions, Gold Rush (Colorado alliance), Hermes (Hopkins), ERDB (UCLA), Boston College, and NC State. The University of Wisconsin is using an Access database and will share their data structure. A program now entitled “Bringing up ERMI” is being planned for a future ALA. Questions: consortium application of ERM, institutional application of ERM with different campuses, cross-institution needs (tech services, public services, IT), interoperability, ERMI Lite, how to combine link resolver and opac and subject lists and ERM, standard license terms, license terms for digitization projects.
Beacher Wiggins reported on catalog enrichment efforts at LC. cf. LC round robin on bigheads webpage They are using ONIX data to add contents notes to catalog records. The contents page is scanned and input, with the data including page numbers and other potentially extraneous information. They are adding links to review articles, with public service staff being worried about unmediated reviews being linked. They are building analytic records from the bib records for technical reports from the World Bank. Columbia is scanning the TOCs for East Asian journals going into remote storage.
Some of these developments are having an effect on organization charts with the separation between technical and public services being less pronounced than formerly. UC Berkeley is becoming more team- than unit-organized. Initiatives and projects seem to work better than traditional organizations, with staff drawn from units as relevant. The Big Heads of public services have developed a list of the top ten trends in public services. Changes in management of resources are coming faster than staff is changing.
Karen Calhoun (Cornell) reported on a study of cross-collection searching. They analyzed seven systems, including RLG Cultural Materials, U Washington, New Zealand Digital Library, Oaister, and the European Library. Google Scholar is also part of this picture. The goal is to avoid a hodgepodge.
RLG Strategy Focus Group (Friday afternoon)
Harvard had been loading changed holdings when they were on NOTIS but may stop, with decision by University Library Council in process. About one million records have been loaded from the Australian National Library. Records from Keio University have been loaded. The Z39.50 connection to the CURL catalog is being dropped.
Export from RLIN continues to be in MARC-8 character set; they would like to offer Unicode export too but it may not be simultaneous with the database migration. Libraries which change their local database architecture will need to do a test file. The migrated database will be stored in XML in UTF-8.
Casalini wants to charge for their MARC records but RLG is trying to convince them that the records are advantageous and that no library would export records from RLIN that they could get in conjunction with orders.
RedLightGreen is fed by data mining which is done several times a year. Searches are not against live database and are presently a bit stale but will be up-to-date after migration.
The account authentication server will be ready on February 1st but the crossover from WinRLIN to RLIN21 won’t take place until March 1st so that authorities and bibliographic can be simultaneous. RLG will continue to do duplicate record detection for the authority file (LC/NAF). Copies of mock-up clusters with new clustering algorithm were distributed. Values to be used for determining primary cluster member and/or which will display for record selection include Encoding Level, classification numbers, scripts, authentication code (e.g. pcc, dlr), last update date, and other factors such as language of cataloging and master microform.
RLG Cultural Materials has released new description guidelines.
Machine-Readable Bibliographic Information Committee (MARBI) (Saturday morning, Sunday afternoon)
The Midwinter agenda -- http://www.loc.gov/marc/marbi/mw2005_age.html -- includes a link to the proposals and discussion papers. The report below is listed in number order rather than agenda order.
Proposal 2005-01: Definition of Field 766 in the MARC 21 Classification Format -- This proposal to add a secondary table for breaking down classification number ranges was approved.
Proposal 2005-02: Definition of Subfield $y in Field 020 (ISBN) and Field 010 (LCCN) in the MARC 21 Formats -- This paper generated the most complicated discussion both on the list before the meetings and at the meetings. ISSN is strictly controlled by international bodies. ISBNs are much more likely to appear in the “wrong” book which causes problems for systems that try to do record replacement based on 020. For example, reprint editions might repeat the ISBN of the original edition, typos may appear, paperbacks may use the hardback ISBN, and duplication may not be known at the time of cataloging. Because of the ambiguity of invalidity, it was determined that $y would not be added. The proposal was rejected.
Proposal 2005-03: Definition of Subfield $2 and Second Indicator value 7 in Fields 866-868 (Textual Holdings) of the MARC 21 Holdings Format -- This proposal to expedite the encoding of holdings information from the U.S. Newspaper Project was approved with minor emendations. There was concern about the use of angle brackets though recognition that the USNP standard uses them.
Proposal 2005-04: Hierarchical Geographic Names in the MARC 21 Bibliographic Format -- This built on earlier discussion papers with options to expand hierarchical geographic access from mere production information to subject access. There are still several issues to resolve and the proposal will come back next time. Issues: indicator for geographic features; more examples; the effect of repeatability on indexing; whether extraterrestrial objects should be $a or $g; LC-centricity; whether 752 and 662 subfields should be parallel; $w for link to thesaurus.
Proposal 2005-05: Change of Unicode Mapping for the Extended Roman “alif” Character -- This paper to bring MARC encoding in line with Unicode was approved.
Discussion Paper 2005-01: Subject Access to Images -- This paper was sponsored by ARLIS/NA and VRA, and was written by Liz O’Keefe with my input. It analyzes the need to distinguish access ABOUT a subject and depictions OF a subject. A simple division by text versus image does not always allow distinction, nor does it solve the problem of emblematic images which could be, for example, OF a lion and ABOUT Venice. The paper lays out several options, i.e. limit by location, limit by material type, different indexing vocabulary, new fields for depictions. Little support was expressed for a new range of fields for subject access to images. No problem was determined in adding codes to existing fields to allow for relator codes where those codes are not now possible. Liz and I will work on the paper for another round. LC may work on some simple proposals to expand the relator codes. Input has already been sought on CAAH, VRA-L, and ARLIS-L. Other cataloging communities have promised us more fodder for this topic. The difference between still- and moving-images was pointed out by Martha Yee of the UCLA Archive of Film and Television Archive. SAC expressed support for the idea of distinguishing. LC Prints & Photographs uses material type for distinguishing. Many catalogs and databases only include one type of material which serves to distinguish. Issues for next round of the paper: distinguishing image, sound and text OF or ABOUT a subject; resource or part of the resource ($3?); relation to thesauri and varying vocabulary; greater granularity in form subdivisions. Liz will initiate list discussion of the functional requirements.
Business meeting and LC report: a motion from CC:AAM will be answered by a letter from the chair indicating that MARBI recognizes the eventual goal of full UTF-8 as the character set but that it is an evolutionary process; MARBI might hold a joint meeting with CC:DA about AACR3 at Annual 2005 or Midwinter 2006, with a discussion paper on MARC implications of revised cataloging rules; Understanding MARC authority records is now available on the web; update 5 is in press; the 583 preservation document is up; some new 042 codes including “natgaz” for the U.S. BGN feature names; CDS is adding 505s to bib records with simple coding (not enhanced $g, $r, and $t) in addition to linked TOCs; a Unicode paper is probable this summer.
Committee on Cataloging: Description and Access
The big topic on CC:DA’s agenda this time was the draft of AACR3. A group of five CC:DA members presented a document with some “significant issues that remain unresolved in AACR3”: representation of multiple entities in one record; automated and user expectations; features common in many metadata standards beyond the library community; semi-hierarchical rules; FRBR; and the nature of the catalog. There is an overall sense that the Joint Steering Committee is rushing the revision and merely rearranging rather than moving forward in compelling ways. Again, as in the development of AACR2, the needs of special cataloging communities seem to be getting short shrift. Programs on both AACR3 and Cataloguing cultural objects are being planned for Annual in Chicago this summer. More info, including some of the discussion papers, may be found at http://www.libraries.psu.edu/tas/jca/ccda/
MARC Formats Interest Group (Saturday afternoon)
Diane Hillmann (Cornell) gave an overview of the classification format. Adam Chandler and Jim LeBlanc, also from Cornell, described their use of the Hierarchical Interface to Library of Congress Classification (HILCC) to build an interface for undergraduate research that will provide reasonable amounts of hits on various topics. They built on work done at Columbia. They divided the LCC classes into ranges that would result in hits between 10 and 1000 in an attempt to “using the LC classification numbers provided in standard catalog records to generate a structured menuing system for subject access on the web.” More info at http://www.library.cornell.edu/cts/backstory/v1n1/hilccfeature.htm
RLG Technical Services Client Transition Forum (Sunday morning)
The CJK Thesaurus from WinRLIN will become a web page for searching. “Cop id” and store/copy keys were added in version 1.2 of the client. A Z39.50 test database of 5 million records is available. The crossover date was postponed to the end of February in order to do cataloging and authority input/update at the same time. Selection of the primary cluster member will be somewhat different in RLIN21 with added weight given to the presence of an 050 field, pcc coding, and digital library registration (part of preservation programs). The RLIN-defined Cataloging Category (CC) will disappear. The authentication server for new accounts was to be available as of February 1st though authorities were still being developed. Special features of authories will include: duplicate record detection (LC asked RLG to continue doing dup detection), 1XX/4XX form conflict detection, “gen” will be developed but not necessarily on Day One, old saved records (more than 60 days) will not be migrated to the new database. It will not be possible to enable authority record input/update until the migration is complete.
PCC Participants Meeting (Sunday evening)
Business items: Mark Watson is chair elect; Carlen Ruschoff will chair a group looking at the mission of PCC which will be looking at how to deal with electronic resources and other issues. Following the reports and business, Dale Flecker (Harvard) gave a keynote address on “OPACs and our changing environment.” He addressed the invaders in our information domain, e.g. Google (Print and Scholar), Amazon, Yahoo. Change is the point, and therefore adaptability is critical. We need to develop search systems that are complementary. Google is fast and can find relevant hits; Endeca and Grokker are new products that can help with large result sets. Endeca works on result set; Grokker works on patterns. Flecker’s talk led to the most vigorous discussion at PCC Participants in a long time. cf. http://www.loc.gov/catdir/pcc/
Subject Analysis Committee (parts of Sunday morning and Monday afternoon)
Catalogers in LC Prints and Photographs use the normal rules of specificity in applying form terminology. Dewey is looking at their segmentation practice, with a draft of the revised practice to come in February. Canadian Subject Headings are being mapped to Dewey. A draft of 741.5 for graphic novels was up for discussion in the fall, after considering whether graphic novels belonged in art (700s) or literature (800s). The NISO standard for thesauri (Z39.19) is still being reviewed. Work is underway on classification training materials, building on the model of earlier training documentation. The intent is for the materials to be modular so they could be expanded as appropriate for special subject workshops, e.g. law, art. The III Users Group is looking at the recommendations of the subcommittee on subject relationships. A paper on that work will appear in the next LRTS. The FAST thesaurus of deconstructed LCSH includes more than 1.3 million records.
Subject Analysis Committee - Subcommittee on FAST (Friday evening)
The FAST (Faceted Application of Subject Terminology) project grew out of the CORC Project at OCLC. It is a deconstruction of LCSH for use by non-catalogers. Each of the elements of a subject heading is broken into its topical, geographic, chronological, and form components. So far, thesaurus development has been built from existing headings. Chronological subdivisions are perhaps the most challenging. The subcommittee discussed the project in general terms. Ed O’Neill of OCLC agreed to supply a test database of 5000 records for analysis by subcommittee members. We will be looking at: how well FAST covers the ABOUTness of a resource; are topical headings well-constructed; hierarchy in geographic headings; personal and corporate names; whether deconstructed headings lose their meaning; chronological headings from scratch versus those created from existing chronological subdivisions. There is quite a bit of documentation on the FAST website at http://www.oclc.org/research/projects/fast and the thesaurus can be found at http://fast.oclc.org
Subject Analysis Committee - Task Force on Named Buildings and Other Structures (Monday morning)
Full minutes at http://artcataloging.net/ala/mw05/sacbldg.html by Liz O’Keefe. The task force is preparing its final report with the following recommendations: explore Art SACO funnel or separate buildings funnel; continue to collect examples; publish examples in such venues as ARLIS/NA update or SCM:SH; encourage more training and workshops such as SACO pre-conference workshops at ALA or subject cataloging workshop at ARLIS/NA in Baltimore. While the current task force is working on its final report (and was extended to Annual 2005), the report and its ancillary documentation will be fed to the Cataloging Advisory Committee of ARLIS/NA and perhaps a subcommittee of SAC to be established. A liaison from ARLIS to SAC was also discussed.
Compiled by Sherman Clarke, New York University Libraries
... go to more ALA reports ...