FRBR in the Jewish Public Library

Imageby Eddie Paul

On a warm spring evening in Paris on May 29, 1913, an opulent crowd at the Champs Élysées Theatre rioted at the premiere performance of Stravinsky’s Rite of Spring. One shrewd musicologist wrote that "the pagans on stage made pagans of the audience." The piece is a celebration of pagan rituals that eventually leads to a sacrifice of a chosen young woman to propitiate the gods of spring.

No such harsh reaction greeted the introduction of RDA or FRBR to the library community back in 1997 when the new replacement for AACR2 was introduced by IFLA, although it certainly hasn’t been without resistance from some elements, possibly because paradigms (MARC is a little over 40 years old) and the people who conceive them shift just a bit more slowly than the tectonic plates. The gods of spring are also not that easily propitiated, especially when the propitiators are limited by consensus and cataloguing subcommittees.

At a conference I attended many years ago, a former colleague of mine who once worked as the systems librarian of a university in Tennessee and later as a designer of catalogue portals for our ILS vendor kept whispering the mantra “MARC must die,” during the breakout sessions. When I first started reading about RDA/FRBR. I thought perhaps this prophecy was about to come true: the equivalent of the rapture as it were. But this was hardly the case. If anything, MARC has become the undead cousin in a binary catacomb of propitiation.

The JPL transitioned directly from catalogue cards to an integrated library system in 1994. The system was eponymously called “VTLS” after the company in Blacksburg, Virginia that created it. In 1999, the second-generation system (Virtua) was developed, and the JPL migrated a year or so later. In the first decade of the 21st century, it became obvious that ILS companies had hit a critical threshold in the systems and layers they were developing to dress up the ways bibliographic information and metadata could be made accessible to users: consider the ILS company as a health spa that was marketing the equivalent of waxing, exfoliation, and Botox treatments.

But the national libraries that comprise IFLA already knew that a new conceptual model was needed to address the problem of accessibility: if ILS companies were becoming cosmeticians in a capitalist war over how best to make metadata and access enticing and seductive, it was inevitable someone was going to determine that it was the organism that needed internal bodywork.

The JPL is an independent public library with two broad mandates: lending and research. The content of its collections, 75% Judaica covering a vast spectrum of humanities, social sciences, fine arts, science, language and literature, is coupled with a popular collection (25%) of bestsellers, cookbooks, popular biographies, and for the most part, works one will find in any public library in Montreal. The five active languages (English, French, Yiddish, Hebrew, and Russian) are supplemented by works in languages that run parallel to the history of Jews for the last 4000 years. Inasmuch as collections budgets have permitted, we’ve transitioned from audiocassettes to compact discs, VHS to DVD to Bluray, and if the gods are sufficiently propitiated soon, e-books. It’s a big soup.

But the spoons with which users are equipped to scoop out the parts of the soup they like are slotted. I’m sorry to say, that the larger the electronic landscape grows, the slots in the spoon grow larger. The best parts of the soup are sometimes hard to find because they settle at the bottom of the pot.

ImageVTLS is not the largest ILS company on the planet, but it has been known for a few radical moves to adopt standards before their competitors did. They were the first to implement Unicode, the first to implement web “skins” technology in its catalogue portal, the first to use an off-the-shelf relational database system, and the first ILS to implement linked authority control and support for the US MARC format. It is also the first ILS vendor to offer a full implementation for FRBR.

Several weeks ago, a patron at the reference desk asked for a well-known epic Holocaust poem Le chant du peuple juif assassiné by Yitzchak Katznelson. The poem was originally written in Yiddish, and the JPL has the original, and the translated editions in French and English. But unless you are somewhat versed in Holocaust poetry, you may not know the exact titles in all languages. Not really an issue since they should all be linked to the author authority, but what happens when you have two authorities for the author (one in English and the other in Yiddish) and the two are not necessarily linked by a “see also” reference. You could blame the cataloguer (as I do when I’m on the reference desk, except I’m also usually the cataloguer and blaming myself in absentia is dishonest albeit not inconvenient), or you could petition your director to spend some additional funds on a FRBR implementation. Honesty is never without a cost.

Obviously this is not a new dilemma. Years ago, a similar situation occurred when someone had asked us for the original version of Elie Wiesel’s Night. Again, unless you know the genealogy of a work, you may not be in a position to intuit the best way to search this on a catalogue. Night was the only work Wiesel ever composed in Yiddish and the title (Un di velt hot geshvign [“And the world remained silent”]) was not rendered into English in a way that was remotely similar.

The children’s library may have 5-6 different editions of a popular story rendered into a picture book, a board book, a graphic novel, a kit (audio and text), and then in 2-3 different languages. More soup anyone?

At this point, the rationale becomes evident. RDA/FRBR will allow us to link new and existing works, formats, editions, languages, into the WEMI (work-expression-manifestation-item) conceptual model. Consider also that the JPL’s collection of Jewish canonical texts (Pentateuch [and commentaries], Midrash, Mishnah, Talmud, Zohar, Mishneh Torah) span editions, translations, and lectionary sections. Reference librarians become heavily reliant on the competency of the cataloguers (see above attribution in absentia), but ostensibly the motivation to move ahead of the pack with RDA/FRBR is to empower users, many of whom expect the library catalogue to behave like Google or Amazon.

The JPL cataloguers are scheduled to be trained in Virtua’s FRBR application in mid-March. Specifically, the cataloguing tools don’t change with the transition, but their functionality is expanded by virtue of modifications that are made on the Oracle relational database management system on which the Virtua layer operates. The linking 004 tag is used to create correspondences between the work, expression, and manifestation records based on criteria that we select. VTLS has already determined that a little less than 1000 records have automatically qualified for FRBRization based on existing MARC tags, but now the work at hand is to revisit many bibliographic records that don’t have these necessary tags so that we can FRBRize them manually. To this end, the gods of accessibility will be propitiated through our own rite of spring, this despite the fact that vast majority of books in our Judaica collection kind of see pagan rituals as anathema to their existence here.