Thursday, November 15, 2007

Future of Bibliographic Control,LC, 11/13

Notes from the meeting on Nov. 13 of the Working Group on the Future of Bibliographic Control.

These are my notes and should NOT be taken to represent accurately the thoughts of the working group, only my quick recording of what I understood at the meeting. Also, I must add the disclaimer that I have been engaged as a consultant to the group for the writing of the report. I attempt in that work to be as faithful to the outcomes desired of the group as I can. However, I admit that pure objectivity is a chimera, so my own opinions may come through in the text below.

There was an introduction explaining the creation of the working group (which you can read about on the working group's web site: The group presented an interim report to the Library of Congress. The full report will be available by December 1 for public comment. The comment period will end on December 15, and the final report will be presented on January 8, 2008.

The report was commissioned by the Library of Congress, but it many of its recommendations involve the the library community and other players in its environment. There are over 100 individual recommendations in five general areas.

The working group concluded that there are three major "sea changes" that are needed in the library community:

1. We must redefine bibliographic control broadly to include all materials, a widely diverse community of users, and a multiplicity of venues where information is sought.

2. We must redefine the bibliographic universe to include all stakeholders, including the for-profit organizations that are involved in information delivery and digitization

3. The role of the Library of Congress must be redefined as a partner with other libraries and with non-library institutions, working to achieve the goals of the library community.

The five areas of recommendations are:

1. Increase the efficiency of bibliographic production for all libraries through cooperation and sharing of bibliographic records and through the use of data produced in the overall supply chain.

2. Transfer effort into high value activity. In particular, provide greater value for knowledge creation through leveraging access for unique materials held by libraries, materials that are currently hidden and under-used.

3. Position our technology by recognizing that the Web is our technology platform as well as the appropriate platform for our standards. Recognize that our users are not only people but also applications that interact with library data.

4. Position our community for the future by adding evaluative, qualitative and quantitative analyses of resources. Work to realize the potential provided by the FRBR framework.

5. Strengthen the library and information science profession through education and through development of metrics that will inform decision-making now and in the future.

Under each of these areas there are sets of recommendations. The full set of recommendations is fairly detailed, and the group presented high level groupings of recommendations in the first four areas. (Area five was not presented in detail at the meeting.)

In area 1, the recommendations are grouped:

1.1 Eliminate redundancies in the production of bibliographic metadata. This means making use of data that is created elsewhere in the supply chain, and increasing the sharing of bibliographic records and modifications to records. In particular, the group asks for an examination of barriers to sharing.

1.2 Increase the distribution of responsiblity for bibliographic record production. Increase the number of institutions that participate in shared cataloging activities.

1.3 Collaborate on authority record creation. Similar to 1.2, this recommends that the number of participants in authority record creation be increased, but it also asks that we look at the possibility of sharing across sectors and internationally, to reduce the number of times that an authoritative heading must be created.

Area 2 is called "Enhance Access to Rare and Unique Materials." In this area the group states that any efficiencies gained in other areas should allow the redirection of energy to providing access to unique materials that are held by libraries and other cultural heritage institutions. In particular, the group recommends:

2.2 Integrate access to rare & unique materials with other library materials

2.3. Share bibiographic data relating to these materials. The sharing of bibliographic data must not be limited to those areas where copy cataloging is desired.

2.4 Encourage digitization to allow broad access

Area 3 is about technology and the Web:

3.1.1 Integrate library standards into the Web environment

3.1.2 Extend the use of standard identifiers for bibliographic entities, and include those identifiers in bibliographic records.

3.1.3 Develop a more flexible, extensible metadata carrier that can be readily exchanged with non-library applications.

Area 3 also addresses standards:

3.2.1 Develop standards with a focus on return on investment. Do analysis before beginning the standards process.

3.2.2 Incorporate usage data and lessons from use tests in the standards development process

Area 4 is about positioning the library community toward a more progressive future. In this area there are three main recommendation areas:

4.1 Design for today's and tomorrow's user. This means that we must design into our catalogs and other tools the ability to present evaluative information, and to allow and encourage users to interact with bibliographic data. We must also make use of statistical and other computationally-derived information in our user services.

4.2 Realize FRBR. The framework known as FRBR has great potential but so far is untested. It is being used as the basis for RDA, even though FRBR itself is not clearly understood. The working group recommends that no further work be done on RDA until there has been more investigation of FRBR and the basis it provides for bibliographic metadata. [Note: this recommendation is likely to change such that there will be specific recommendations relating to RDA; FRBR will be treated separately.]

4.3 Optimize LCSH for Use and Re-use. Encourage an analysis of LCSH that would move the system toward a more facetted subject system. Work to create more links between LCSH and other subject heading systems in use. Recognize that with the digitization of works the act of subject assignment may benefit from computational analysis.

In the time that I was at the meeting (I had to leave before the question period ended) there were two questions/comments. The first had to do with the fact that while there are costs to today's methods of bibliographic control, that changes in bibliographic control will have costs as well. (Here it would be good to listen again to the talk given by Rick Lugg at the meeting held at LoC. He spoke of the costs of NOT changing, something that is hard to measure but is very real.) The other comment (from Barbara Tillett) mentioned many of the recommendations and stated that LoC is already engaged in, or has rejected, analogous activities. It was acknowledged, however, that LoC had not made these activities public, so the community is generally unaware of the progress made. To me this points out one of the areas that we all need to work on, which is sharing information about our projects and their progress so that the community as a whole can benefit from work done by a single institution.


Anonymous said...

Karen, thank you for making your notes available. It is much appreciated.

As Chair of the RDA Implementation Task Force (under CCS/ALCTS), I am intrigued to read the WG's recommendation that work on RDA be suspended in order to better understand FRBR (keeping in mind that we do not yet have read the absolute final report to read). I look forward to our future debates and activities concerning this issue.

Anonymous said...

here's my take on FRBR:

The librarian in Robert Musil's “The Man Without Qualities” explains that reading any particular book distracts from what is truly important: the relationship between all books.
from The Economist

João Lima said...

Does anyone in RDA Joint Steering Committee knows about FRBRoo effort? The FRBRer version has “semantic inconsistencies or inaccuracies”.

All logical differences are big differences. — Wittgenstein
All logical mistakes are big mistakes. — Darwen’s corollary

Judith said...

At first glance these recommendations all look reasonable, but recommendation 4.2 is sending me home for the weekend extremely depressed. I don't see how the working group can say that FRBR is untested. Here in Australia alone I can point to Austlit ( and the National Library of Australia's LibraryLabs (

At the National Library of Australia, the FRBR model has informed our thinking in so many areas as we grapple with removing barriers to access for end users, the digitisation of original materials and the collection of born-digital content.

Without applying this model to discovery and delivery workflows users will continue not to be able to request "any manifestation" of a wanted resource. If I as an individual interested in Australian pottery want to link to a resource described in Libraries Australia in a Wikipedia article I'm writing or a blog entry or to bookmark it in, I won't be able to send users just to a single place. The higher education, museum and archive communities here in Australia are all finding FRBR enlightening as a framework to help them understand their own content as they move into a digital age. And how better can one realise FRBR than for an international community to try and align its new Resource Description and Access standard with the framework.

Many individuals will find FRBR conceptually hard to understand but try looking at the CIDOC Conceptual Reference Model (CRM) (ISO 21127:2006) if you want to grapple with complexity. It is how the model assists us to simplify end-user discovery and delivery workflows and to expose relationships between resources that matters. And there's plenty of evidence that it does.

João Lima said...

FRBRoo = CIDOC CRM + FRBRer concepts

I think that the main innovations of FRBRoo are the new Work hierarchy with 9 classes and the inclusion of time dimension (ignored by FRBRer). Apparently complex, this new more detailed modeling allows us to represent each aspect which in FRBRer were under the umbrella of a single entity (the Work Entity) making the model easier to use.

To understand FRBRoo, as pre-requisite, is necessary to understand the ontology constructs (a more powerful notation than Entity Relationship Model).

RDA must address FRBRoo and not FRBRer.

Anonymous said...

Thanks for the great summary, Karen. Like Judith, I can't help but feel a bit deflated by the FRBR/RDA recommendation, especially after being excited by the new FRBR-based RDA organisation yesterday. But I'll reserve judgment until I see the full report.

Anonymous said...

To anonymous:
Whatever my (weakly established, I must reckon — however, I suspect a lot of irony implied from the quotation) respect for Musil, anybody talking about "all" books, makes me cringe back.

Borges' Babylon's library should be considered more seriously than a mere wonderful short story: it raises effective issues if you indescriminatingly gather every expression of every work...