One Big Library.

Now that the final report is out, I'm ready to say - it doesn't go anywhere near far enough. Presently it reads:

3.2.5 Suspend Work on RDA
3.2.5.1 Suspend further new developmental work on RDA until a) the use and business cases for moving to RDA have been satisfactorily articulated, b) the presumed benefits of RDA have been convincingly demonstrated, and c) more, large-scale, comprehensive testing of FRBR as it relates to proposed provisions of RDA has been carried out against real cataloging data, and the results of those tests have been analyzed (see 4.2.1 below)

I'd propose adding the following:

3.2.5.1.1 Also suspend further new developmental work on OAI-ORE, SRU@OASIS, HTML5, RDF/*, SKOS, MODS, OOXML, USB3, Android, OpenSocial, the Facebook API, 802.11N, Python 3000, Universal Wiki Markup 1.0, and actually all of Wikipedia, just stop updating that for a bit, yes, that would be lovely, thanks, IMDB, all Google Calendar accounts, the Writers' Strike, Chandler, DukeNukemForever, everybody's del.icio.us bookmarks, recordings on the Tzadik, Anti, and Cantaloupe Records labels, Stevey's Blog Rants, the Whole Wireless Spectrum, OSX, Vista, CSS3, Rails, ngc4lib, the Fed Funds rate, the Billboard Top 100, HTTPbis, DLF Aquifer, ECMAScript 2, IndieTorrents, TED Talks, The TWiT Network, XKCD, mongrel 1.2, Koha 3, dchud's hairline and belly, all Debian package repositories, Led Zeppelin V, the Django trunk, the new Jandek record, NTP, Archivists' Toolkit 1.1, OpenSRF 1.0, and Perl 6, until such time as we have been able to catch up on all of our other obligations and professional responsibilities and newly devote the full and complete community attention each of these critical developments vital to the future of our entire profession deserves. And testing, lots of testing, too, we can't forget that (see 4.2.1 below).

When not otherwise dangling toes in the surf, I've been caught up, like many others, in thinking about things WoGroFuBiCo and DCAMish lately. I'm tending to think that DCAM would be awfully cool if it were just an AM (sans the "DC"), and if it simplified the description set model into looking a lot more like the resource model, and if it dropped the primacy of notions of "record" and "resource".

You can't drop the "legacy" of centuries of dependence on catalog cards without dropping the importance of the "record", and maybe it's time that we did. It just seems to assume too much about the "item" vs. "surrogate" dichotomy, and it places too much importance on us librarians and our work. Similarly trying to differentiate "resource" and the work/manifestation dichotomy just seems confusing. Lately all I want is just a bare bones, fully decomposed model for passing around descriptive statements which might or might not reference externally defined elements and vocabularies for properties or values. MARC, RDF, these formats are essentially all the same to me (give or take 40 years), and it's all a distraction. They'll all (continue to, in the case of MARC, or prove to, in the case of RDF) work, but that's not the point, so let's not focus on that. If I understand correctly, based on what lots of colleagues I trust and depend upon tell me, a lot of other people seem to be leaning the same way, but there's no way I can penetrate the RDA docs or discussion lists deeply enough to confirm that myself.

In some reading on all this tonight I returned to William Denton's "FRBR and Fundamental Cataloguing Rules", inside of which may be found several definitions of "the purpose of a catalog" through the ages. These are fascinating, and informative, but like my above claims about records, resources, and surrogates, I don't think we can base new designs for new systems or formats on these same premises, however valid they remain in the specific (i.e. they're great if you're interested in authors and books, but otherwise...). I doubt I'm alone in thinking that, too, but I'm not sure if I've seen a problem statement broad enough for 2008 and beyond yet.

At the most general level, the purpose of a user-facing library system today can only be:

to find and navigate information in ways appropriate and convenient for the members of the community a library serves.

Maybe that's so broad it's useless. But how, when speaking of matters broad and foundational, can we justify any specific talk of "books" or "author-title" or even "items" when we pretty much can state only that the variety of intents driving users in our libraries is too diverse and fluid to ever enumerate usefully?

Authors and titles and works are important because they're convenient for a lot of functions in user-facing library systems, but they're not the whole game anymore, so we shouldn't pass library data around in complicated, hard-to-understand, impenetrable-to-outsider, non-degradable-to-basic-indexing formats that presume that they are. That's not a controversial statement any more, is it? It doesn't seem to be so, and I sure hope it isn't.