Bibliographic control does have a future. Though, after spending 30 minutes last night describing the problem to my wife, I sure wish we had called it “the future of finding stuff online.”
I’m headed out this morning for a meeting convened by the Library of Congress’s working group on the Future of Bibliographic Control. It is an austere group that I think has real potential to make some serious recommendations. Boy, that sounded like I’m running for office, didn’t it?
Anyway, I am giving a short presentation on “New Services,” which is really just a euphemism for the fact that MPOW has a pretty cool faceted navigation catalog and a homegrown electronic resources management system. So, I’ve been thinking a lot about data, relevance, and subject access to metadata records. It seems that everyone is after a less expensive way to do what libraries have always done. Not a bad goal.
As a systems guy, I am confident that computational analysis of data will be able to accomplish some measure of what has traditionally required human intervention. (Politician-speak again, eh?…I think I just said that computers should replace people). As a librarian, I am stuck on the paradox that better analysis (even computerized analysis) is only accomplished with the aid of human intervention.
Let me put that another way: Some have suggested that “subject keywords” should replace Library of Congress Subject Headings (LCSH). The paradox is that those very keywords are most logically derived from LCSH. “Tagging” (or “social tagging”) would be another option, but that is human intervention writ large, and while the effectiveness of tagging is not shrouded in great doubt, I think the jury is still out on its comparative effectiveness and precision in controlling bibliographic data. I will say, however, that I am very, very impressed by the work of Tim Spalding at LibraryThing. Libraries should be licensing his database’s social tags. LibraryThing’s interface design is also very, very good. I’m sure Amazon.com has taken notice.
Of course, one more option would be to just scan all the books. Every last one of them. And then rely on a really good algorithm to sort things out. Did I mention that the meeting I’m going to is at Google Headquarters?
I don’t disagree that we should always be in pursuit of faster, better, cheaper. I remain skeptical that all three will be accomplished at the same time. And there’s the rub. One thing’s for sure: I’m happy to be on the ride while it lasts.
[This post originally appeared as part of American Libraries’ Hectic Pace Blog and is archived here.]
You say: “Some have suggested that ‘subject keywords’ should replace Library of Congress Subject Headings (LCSH). The paradox is that those very keywords are most logically derived from LCSH.”
First, who are these “some,” and where have they suggested this? Also, this argument is built around a false dichotomy–no LC, no other road to access. (Or even, LC versus “let the world tag us!”)
What I did learn in my last job (and validated through careful analysis) is that valuable non-LCSH “subjects” (complete with divisions, hierarchies, and all the richness those provide–what’s missing from tagging) can be quickly and inexpensively assigned by non-cataloger professionals–inexpensively, as in, 1/4 the cost of LCSH, and we didn’t over-spend on that. Not only that, but the language of these “subjects” (we called them topics) was far more closely aligned with user behavior.
Furthermore, valuable terms can be gleaned from a variety of sources: tables of contents; machine-generated abstracts (see Infomine’s work for excellent examples in this regard); and… dare we say it… the content of the item itself, which for most new items is already available in digital form, as that is how the book is produced.
So do not limit this discussion to LCSH “or “chaos is come again.” Didn’t work for Othello… shouldn’t work for us. 🙂
You’ve hit it. Very well stated. Maybe you should be in politics. And do tell about the meeting once it has happened. Very exciting stuff.
Okay, KGS, I see your point. But, even at 1/4 of the cost of assigning LCSH, who is going to assign this new list of “topics” (where is that list and who maintains it?) to the bibliographic corpus that currently exists.
As for this statement:
“and… dare we say it… the content of the item itself, which for most new items is already available in digital form, as that is how the book is produced.”
That reminds me of exactly what everyone said when predicting that e-books would explode upon the scene. It’s already digital, how hard can it be? Hard enough that it would seem only the power of Google and Microsoft can put a dent in digitizing books.
As for who advocates subject keywords…I think you just did, no? It’s just that you called them ‘topics’ and ‘tagging’. Actually, that came from a recommendation in the Calhoun report to the Library of Congress (http://www.loc.gov/catdir/calhoun-report-final.pdf).
I’m not defending LCSH, just pointing out that a real viable alternative has not been thoroughly researched and proposed. I certainly don’t have the answer. Hence these meetings, I guess.
Ah yes, thorough research and proposals! That’s what we need. 😉 Titanic… deck chairs… you know the drill…
There’s a difference between expensively assigning LCSH and letting librarians assign ad-hoc topic terms on an as-needed basis. Like everything else, we took something important and made it so convoluted that no one but us is insane enough to do it.
What list, you ask? Well, there is a sort of a list, actually a number of them, but basically, you let librarians assign terms as they come up. It’s a cost model between tagging and LCSH.
I wish that the working group on the Future of Bibliographic Control had an RSS feed to alert me to new content on their website.
Andrew,
Good morning. I just discovered your blog this morning. I appreciate your reflections and all the good work that you are doing.
I am wondering if you are familiar with Thomas Mann and his work? I have come across some of his stuff lately – he is a rather accomplished (skilled, practical) reference librarian and the library of Congress who has written some good books on how to get the most out of libraries when it comes to doing hard scholarly work (research – not just finding information fast). He seems to think at quite a deep level that I don’t detect elsewhere.
If you are not familiar with him, let me know – if you are up for it, I can try to send you some snippets re: what I think are some of his more penetrating insights.
Thanks again,
Nate Rinne