We live in interesting times. The Anglophone library associations are well underway in their effort to remake AACR2 under the umbrella of Resource Description and Access (RDA). Meanwhile, the Library of Congress Working Group on the Future of Bibliographic Control recently released it’s draft report, and no less of an AACR2 luminary than Michael Gorman has let loose with both barrels on the RDA effort, as if it were expressly designed to annoy his editorial sensibilities. Of course we can’t overlook the fact that Gorman nails down one end of a spectrum that is marked on the opposite end by such well-known colleagues as Diane Hillmann and Karen Coyle. It is, in other words, virtually anyone’s game at the moment. So perhaps I can be forgiven if I think I have something to contribute to this debate.
On my way to the airport (a trip of which I’m all too familiar these days) it hits me — the nub of the problem for me is that we even call this issue "bibliographic control". There is no part of that is nothing but just plain wrong in today’s environment. Let me break it down.
"Bibliographic" – This term has a lot to do with published literature — mainly book literature but to some extent also journal — but not much at all to do with many of the forms of communication now being used on the Internet. When we harvest web sites, what is the part that we can call "bibliographic"? How bibliographic is, for example, a collaborative blog? We desperately need a more general and generic term for this, which for lack of a better idea I’ve decided to select "descriptive", since no matter what the resource we are basically talking about describing a resource.
"Control" – What part of this do I need to explain? Google is mass digitizing the contents of our largest libraries and making the output available for full-text searching. LibraryThing.com and others are being quite successful at letting anyone and everyone assign whatever terms pop into their heads to the books they own. If this was all garbage, and easily ignored, it would be one thing, but it isn’t and we can’t.
So here’s the thing: you may or may not have noticed it, but we just went from a world where we were the gatekeepers to information to one in which we are hanging on for dear life. We can either wise up or get out of the game. I prefer to wise up. For me this means forgetting about "control" and getting good about "enrichment". NOW WAIT. I’m not talking about ditching controlled vocabularies. But neither am I talking about ignoring uncontrolled "tags" that users assign. I’m talking about being intelligent about using them both. That is, why not use all the descriptive information about an item we have at our disposal? Sure, we can give a greater weight to cataloger assigned controlled vocabulary headings than we do to user assigned keywords, and index and display them differently, but we can still use them both. I mean, why wouldn’t we want to? Isn’t more descriptive information better than less?
My vision of our future is one in which for any item we care to describe we begin by taking whatever information about the item already exists. This could be the author’s own metadata, or someone’s tags, or the publisher’s ONIX record. We then enrich this initial record as we can with controlled headings, authority work, and additional descriptive elements. But we don’t ignore records that we’ve been unable to enrich in this way — we simply treat them differently. Perhaps we don’t rank retrievals based solely on this information as highly, or display uncontrolled topic and authority headings as prominently. The point is to neither ignore them nor raise them to the level of professionally assigned descriptive elements.
There is a great deal of value in the work that catalogers do. But this does not mean that we should ignore that which others can provide. In at least some cases, these individuals will be more knowledgeable and experienced in their topic area than any cataloger. In other cases they will be rank amateurs. So be it. Welcome to a world without control. Or perhaps I should say with measured control — control of some parts and not of others.
The world that I envision is one in which, as I’ve said before in other venues, we like any metadata we see. We should strive to acquire, use, and appropriately display both user input as well as cataloger-assigned metadata elements. Not one or the other alone, but both.
I no longer believe in the future of bibliographic control. I no longer believe that the term "bibliographic" encompasses the universe in which we should be interested, and I no longer think "control" is either achievable or even desirable. We have entered the age of "descriptive enrichment" and we’d better get bloody well good at it.
About Roy Tennant
This is a great post for many reasons, but one angle I particularly like is the idea that expert-supplied data and user-supplied data are not mutually exclusive but in fact complementary. You don’t succumb to hive-mind fabulism but you also aren’t back there with Gorman, insisting on the glory of the Olden Dayes.
One thing I haven’t seen anyone raise is the preoccupation in cataloging with bibliographic *records* to the omission of any discussion of bibliographic *data.* We’re stuck with literalisms based on the old catalog card. If we’re seeing metadata as a kind of service layer within an object, it should be viewed as more flexible than a mini-me record for the item. We can do a lot more with data than records.
I’ll echo K G Schneider: the two are not mutually exclusive. Libraries are no longer the sole gatekeepers to information but still have a role by offering value-added services and access. Enrichment, as you suggested, is the way to go.
Good post, Roy. I, too, prefer to wise up, and I like to think I’m working on it. Your focus on description is a good one, but I think it’s probably broader than that, even. But at bottom, it IS the data rather than our old notion of a bounded record, created perfectly at one point in the lifecycle and meeting all needs of the present and future.
I’ve long been a proponent of getting down to “statements” and not focusing on “records” and I hope we can convince others that this is part of the shift from the fantasy of “control” to the reality of effective use and re-use of all the data we can find.
P.S. Thanks for spelling my name right! :-)
Tennant makes some valid points (especially the prospect of dual enrichment between formal and informal mechanisms), but I also think there is a misguided smugness to the bleeding edge proposition that traditional bibliographic control is effectively out of date. The analogy that comes quickest to mind is how “new media” is negatively effecting the role of reportage and journalism in society today. Relying on something like user produced tags for proper bibliographic access is really not that far from getting all your news from the blogosphere. If you choose this path just don’t be surprise if you end up down some rabbit hole headed in the wrong direction. To Tennant’s credit he does hedge his bet and calls for a mixed solution to the dilemma before us, but I also think the assumption that this external whirlwind of metadata is a gold mine is a tad naive at best.
This is a straw man argument. I don’t think many have suggested that there is no value in things like user tags and author created metadata. I think everyone agrees that they can be useful and we should find ways to incorporate them. Those of us on Gorman’s side of the argument are fighting the proposition that things like controlled vocabulary are too expensive, should be eliminated and replaced entirely with keyword searching or some other new method that hasn’t really been figured out yet. That proposition has already been implemented with the lack of series authority at Library of Congress. We’re jumping away from the old systems before there really is a functional replacement in place.