On January 5th, after one day back at work after Christmas break, I flew out to San Diego for ALA Midwinter. I had to get in a couple days early, because I had to attend a NASIG Executive Board meeting on the 6th. It was very productive all-day meeting, where we talked about NASIG business and set new policies, but confidentiality forbids my discussing it in detail.
From Friday the 7th through Sunday the 9th, I attended a number of sessions at ALA Midwinter, and I can talk about those. Almost every session I attended focused on RDA (Resource Description and Access), the new cataloging code that has been proposed to replace AACR2. This is the biggest thing to hit the cataloging world in over 30 years, since AACR2 was adopted. I’ll try to boil down the useful information I gathered as best I can.
The RDA Update Forum was where I heard the biggest news. A representative from the Library of Congress (I’ll confess I didn’t catch his name) discussed the testing of RDA at LC and the two other major national libraries. The testing period closed on December 30, with about 7,000 RDA records created in OCLC, and they are now analyzing the test data. The first report of analysis is due to the management of the national libraries by March 31, and they plan to issue a joint decision on the adoption of RDA at ALA Annual in New Orleans in June. Their decision could range from refusing to adopt RDA at all, adopting it as is, or adopting it after certain problems have been addressed by the Joint Steering Committee (the body responsible for creating RDA). However, since AACR2 is now dead and will not be updated in the future, it seems entirely out of the question that RDA would not be adopted at least provisionally by the Library of Congress. One way or another, RDA will be the new cataloging code.
Another speaker at the RDA Update Forum, Chris Cronin of the University of Chicago, was quite enthusiastic about the adoption of RDA. The University of Chicago is an RDA test library, and their experience was generally positive. Their general approach was, “It’ll be alright…really it will.” They involved all catalogers, professional and paraprofessional alike, in the test, chose to minimize local exceptions and follow the code as written whenever possible, and gave preference to cataloger’s judgment over broad policy decisions for every scenario. In the end, they found a several things that catalogers disliked (such as the changing of established headings, the recording of copyright dates, and navigating the RDA toolkit), and quite a few things that they liked (such as the way RDA expresses relationships between entities, getting rid of abbreviations, and the treatment of reproductions). The biggest area of concern centered around the training and documentation for copy catalogers (questions to be addressed included: will copy catalogers accept copy records as-is? Will they update poor RDA copy? Will they upgrade AACR2 records to RDA?). At the end of the testing period, the University of Chicago catalogers held a vote on whether or not to adopt RDA and they voted unanimously in favor of adopting the code. So my hope is that when we eventually move to adopting RDA at Wake Forest, we’ll find, like Chicago did, that “it’ll be alright…really it will.”
I also attended the FRBR Interest Group Meeting, where I heard an interesting presentation by Yin Zhang and Athena Salaba of Kent State University. They discussed their research project to use existing MARC records to create FRBR-compliant records at the expression and manifestation levels (FRBR is a conceptual model for the description of bibliographic entities that underlies RDA). I realize that’s probably clear as mud to most of you, but the important thing is, in their research, they used algorithms to convert existing bib records into smaller records that describe the work in more abstract terms (that is, a record for all English-language versions of the work, a record for all Spanish-language versions of the work, and a record for the work at its purest form, regardless of language). They found it difficult to use the algorithm to split the existing MARC records into these more finely granulated FRBR-compliant records without losing data or incorrectly converting records. Undoubtedly, some type of automation will be necessary for creating FRBR-ized records, but they will also require a great deal of human intervention to clean up errors, because moving from less clearly-defined data to more clearly-defined data is very difficult to accomplish using only computers.
I saw a presentation at the Cataloging Form & Function Interest Group, which further discussed the issue of FRBR and RDA compliant records from an ILS perspective. John (something, again, I failed to catch his full name, even though I’ve seen him speak before) from VLTS talked about the implementation of RDA in their catalog. They appear to be the company that is most aggressively pursuing a full implementation of RDA, with an underlying relational/object-oriented database structure. Another presentation at this meeting discussed a joint experiment by four libraries involved in the RDA test: North Carolina State, University of Chicago, Columbia, and University of Illinois at Chicago. They experimented with encoding RDA records using formats other than MARC, namely, MODS, EAD and Dublin Core. The results were mixed, finding that EAD worked pretty well overall, but that MODS had too many fields with inferred data to be useful in machine processing, while Dublin Core did not allow for the fine granularity required by RDA.
I was quite interested to hear this discussion of using alternative formats to record RDA data, because I’m currently chair of the ALCTS/LITA MARC Formats Interest Group, and I had to recruit speakers to discuss a topic at Midwinter (and will have to do it again at Annual). The topic I gave my presenters was, “Will RDA Mean the Death of MARC?” People have been saying for years that MARC is inadequate and needs to be replaced, but there has been no serious movement toward adopting a new format, so I wondered if the adoption of RDA might be a big enough event that it might start movement toward a new format. The speakers I invited all thought that MARC should be replaced, but all approached the issue from different directions. Chris Cronin of the University of Chicago, suggested that RDA won’t kill MARC, MARC will kill MARC, by which he meant that the inadequacies of the format will make it untenable in the future. Cronin didn’t know what format would rise up to replace MARC, but he strongly urged that we begin the conversation in the cataloging community. However, the questions of who will finance this shift, and who will do the work loom large. Also, we must be prepared to deal with the fact, that any wholesale move from MARC to another format, will inevitably result in the loss of some data. Jacquie Samples, newly of Duke University, spoke about the need to develop a successor format. In an apt metaphor, she said it was as if MARC were a very old king who had done great things in his youth, but had been lingering on his deathbed for many years, and there was no clear line of succession, and not even a decent contender for the crown available. And Kelley McGrath, of the University of Oregon, spoke of how the more richly detailed data required by RDA will quickly hit up against the structural limitations of the MARC formats, and how, if we wish to take advantage of the possibilities offered by this new, richer data, we will need to find an alternative to MARC. The session was quite well attended and the audience included representatives from the US MARC Office and the Library of Congress. At least one attendee told me afterwards that I had given my speakers a controversial set of questions and that they were brave. So this may either generate some interest in the problem, or make me a pariah in the cataloging community. I guess we’ll see which one at the next meeting.
I’m already running long, but I just want to say quickly that I also managed to have some fun in San Diego. I saw a number of former ZSRers. I roomed with Jim Galbraith, who is now at DePaul University in Chicago, and I had lunch with Jennifer Roper and Emily Stambaugh (with whom I did some impromptu consulting on a data management problem). As Erik posted, the two of us had a brief, but intense conversation about RDA in the halls of the San Diego Convention Center, and I hope we’ll continue that conversation on this side of the continental divide. And, on my last night in San Diego, I went to dinner with Susan, Roz, Carolyn, Molly, and Bill Kane, where Bill announced the ACRL Award to the table with a champagne toast. After that, we went for a reception and tour at Petco Park, where the San Diego Padres play. It was a pleasant way to end my stay in San Diego, even if my travels home were somewhat difficult (I’ll spare you that story).