Wednesday, June 12, 2002


"In early April, the Health Sciences OCLC Users Group (HSOCLCUG) conducted an anonymous survey on MeSH authority control practices in health sciences libraries. There were 105 respondents to the survey. Here is a summary of our findings.

Half the respondents (overwhelmingly the largest group) do their authority control manually. Half have brought their MeSH authorities up to date sometime in the last two years, but a third never have. About 70 percent have a mixture in their OPACs of pre-1999, post-1999 and NLM distributed MeSH.

A third of the respondents are using post-1999 MeSH practices in their original cataloging, and almost half are applying post-1999 MeSH practices in their OPACs. The number who would prefer that NLM provide records to bibliographic utilities with post-1999 MeSH nearly equals those who prefer NLM to continue its current practice of providing records with distributed MeSH structure. Thus, respondents appear to be divided on the post-1999 MeSH vs. distributed MeSH structure issue.

The survey also reveals that most vendor systems' global modification functionality are not adequate to convert old records in catalogs to agree with either NLM's current practices (post-1999) or NLM's distributed structure (distributed through utilities like OCLC).

Almost 60 percent enrich OCLC records with NLM classification and MeSH on a regular basis. Almost three-fourths would download MeSH authority records from OCLC if available.

Finally, an unexpectedly large number of comments probably indicate that MeSH authority control is a topic of very high interest to respondents.

For the entire survey results, including graphical displays, please refer to this PDF file, using Adobe Acrobat reader or click the survey link on this webpage

We hope you will enjoy examining the survey results and that you will contact the survey coordinators if you are interested in the work that HSOCLCUG is doing. MeSH authorities, OCLC new directions, changes at NLM, and resource sharing/universal MARC Holdings are four areas of special interest to HSOCLCUG.

The survey coordinators:

Dan Kniesner ( and Judith Dzierba ("

Wireless Web Browsers & the OPAC

Another OPAC I've not seen, but sounds interesting. If you catch it at SLA or ALA let me know how it looks. The III AirPAC is designed for mobile phone, wireless PDA and other mobile devices.

"Patrons can interact with the Millennium system from anywhere accessible by wireless networks, allowing a user to place a hold from a classroom, check due dates and renew items during a long commute, or search the catalog while running errands to find the closest library branch where a desired title is held."

Does the Handheld Librarian know about this one?


I'll be on vacation for the next week and a half. Posts will be very intermittent, if they occur at all. I will be checking my e-mail a few times and if I see anything screaming to be posted I'll get up. Otherwise, I'll be back on June 24.

Cataloging in Other Languages

The other day I mentioned the problem of identifying just what the language is on some materials. Ukrainian and Russian or the Scandinavian languages can appear very similar to someone who does not speak them. Owen Massey brought language guessers to my attention. For example, the Content Analysis Language Identifier has the ability to distinguish between 47 different languages. It uses UNICODE so it can take most character sets. Thanks Owen for the tip.

Preservation Metadata


DUBLIN, Ohio, June 11, 2002--The Working Group on Preservation Metadata, an initiative jointly sponsored by OCLC and RLG, has released 'A Metadata Framework to Support the Preservation of Digital Objects,' a new report available on the OCLC Web site.

The report is a comprehensive guide to preservation metadata that is applicable to a broad range of digital preservation activities. Preservation metadata is the information infrastructure necessary to support processes associated with the long-term retention of digital resources, and is an essential component of most digital preservation systems.

The report represents the consensus of leading experts and practitioners comprising the working group, and is intended for use by organizations and institutions managing, or planning to manage, the long-term retention of digital resources.

The working group based its work on preservation metadata element sets developed by several leading institutions and organizations in the digital preservation community, as well as the Open Archival Information System (OAIS) reference model. OAIS is a conceptual framework that articulates the primary functional components of a digital archive and establishes concepts and terminology for describing and comparing architectures and data models.

The report follows on the working group's earlier white paper, 'Preservation Metadata for Digital Objects: A Review of the State of the Art,' which defined and discussed the concept of preservation metadata, reviewed current thinking and practice in the use of preservation metadata, and identified starting points for consensus-building activity in this area. The white paper is also available on the OCLC Web site.

"The working group's report illustrates the value of cooperatively resolving the challenges of digital preservation," said Brian Lavoie, research scientist in the OCLC Office of Research. "The most effective solutions for the long-term retention of digital resources are likely to emerge from collaboration within the digital preservation community, pooling the expertise of researchers and practitioners from a variety of institutional and geographical backgrounds. Taken together, these consensus-building efforts will establish the necessary infrastructure, in the form of standards and best practices, to support digital preservation activity."

"For institutions creating and preserving digital objects, this new OCLC/RLG report is a good resource for both preservation metadata and its relationship to the OAIS reference model," said Robin Dale, RLG program officer. "It broadens the digital preservation toolset we've been building, which includes such aids as the RLG/OCLC report on the attributes of trusted digital repositories and RLG's OAIS practitioners' Web page."

"The work of the OCLC/RLG Working Group on Preservation Metadata has been an extremely important effort as libraries and related cultural institutions begin to tackle issues concerning digital preservation," said Rebecca S. Guenther, senior networking and standards specialist, Library of Congress, and working group member. "The working group document takes the OAIS information model closer to the goal of actually implementing archival information systems that document preservation requirements and processes and is particularly relevant to the needs of libraries. This document is an important contribution to the development of preservation metadata standards in information technology, an area where much additional work is needed."

The Working Group on Preservation Metadata is part of a cooperative effort between OCLC and RLG to support consensus-building activity in the development of key infrastructure for digital preservation.

Tuesday, June 11, 2002

Dewey in Spanish

A Spanish language version of Dewey to the 1000 level is available. This would be a good guide to have available where some of the patrons are Spanish speaking. I'd not want to use it for classification. The Spanish in Our Libraries (SOL) site has some other valuable resources for libraries serving a Hispanic population.

Open Archives Initiative

my.OAI is a full-featured search interface to a selected list of metadata databases. They have recently made some inprovements

The following changes/improvements were made:

  • Added a recommender system which suggest other documents perhaps worth looking at based on prior search and retrieval patterns of other users.
  • Changed the way in which databases are selected, now they are displayed as a list rather than databases with checkboxes, this is to accommodate their growing numbers.
  • Added the ability to run a meta-search, displaying the total number of results per database before displaying the search results, the user can then pick which database(s) they want to see results for.
  • Added the ability to group the search results by database, the user can then pick which database(s) they want to see results for.
  • Added documents viewed to the search history, now you can see both your previous searches and all the document you viewed in reverse chronological order.

    Thanks to FOS for bringing this to my attention.

  • Monday, June 10, 2002

    FRBR & the OPAC

    Here an announcement of an OPAC that displays records in FRBR format. I've not seen it. If anyone does see it at ALA or SLA I'd be interested in hearing comments.

    "VTLS Inc. Announces FRBR Implementation
    ALA JUNE 2002

    Blacksburg, VA -- Virtua ILS – Integrated Library Systems newest release is currently the only commercially available ILS system in the marketplace to support the IFLA Functional Requirements for Bibliographic Records (FRBR). The FRBR model is the latest development in cataloging practice. It is designed to serve as a conceptual framework for systems that share and use bibliographic data, nationally or internationally. The concept was developed by IFLA with participation from National Libraries like Library of Congress and the National Library of Canada.

    Present cataloging practices do not support "linked hierarchic metadata records" making it difficult to catalog certain types of materials. The FRBR model is designed to address this difficulty. The basic FRBR record consists of three entities - work, expression and manifestation. In addition, item records (holding records) can be attached to the manifestation. For example, Beethoven’s Fifth Symphony represents a work. It can be independently cataloged. The performance of the symphony by, say, the New York Philharmonic Orchestra represents an "expression" of the work. A CD by Columbia records containing the particular performance represents a "manifestation" of the work. Two copies of the CD in the library represent two "items" of this manifestation.

    Using the FRBR model, librarians do not have to re-catalog the entire work again for different performances or for different manifestations of the same performance. Instead, FRBR allows the new performance to be cataloged separately and linked to a previously cataloged "work". This hierarchical parent child relationship in metadata records is the key feature of FRBR. The example above was from music; however the hierarchic relationship can exist in other material types like printed materials with different editions or language versions.

    In implementing FRBR, VTLS has created an environment in which records in the FRBR model and records in the traditional cataloging model can co-exist in the same database. Such an implementation allows libraries to ease into the use of the FRBR model without requiring massive record conversions. It is clear that not every metadata record lends itself to FRBR. For example, many records do not have alternative expressions or manifestations. For this reason, Virtua ILS supports classic cataloging of metadata records along with FRBR records in the same database. The software is "FRBR aware" and automatically changes display formats depending on the type of record (FRBR or regular).

    With a click of a button, the Virtua software converts a regular record to three records of the FRBR model and attaches all items to the manifestation record. This process can also be performed in a batch mode.

    Refinements on the FRBR implementation continue as we receive valuable input from interested parties.

    For more information on FRBR please see:
    IFLA web site
    VTLS web site"

    Cataloging in Languages Other than English

    The Glossary of Bibliographic Information by Language can be a real help when dealing with materials in a language you are not fluent in. It provides translations for common terms used in publishing in Danish, Dutch, French German, Italian Latin, Norwegian, Polish, Portuguese, Spanish Swedish and Turkish. Now the trick is finding out just what the language is. I recently cataloged a map in Czech, Coratian, Polish and Hungarian. It took a good bit of time figuring out that Coratian was one of the languages.

    Thesauri and Web Logs

    On the High Context 'blog this short article shows how a controlled vocabulary could benefit resource discovery in Web logs. Thanks to the Shifted Librarian for pointing this out.