Alternative subject classifications

So I was cataloging a book on the theory of integral equations (don’t all great stories start this way?), and I had to read the introduction because I love books about math, and the editor mentioned the Mathematics Subject Classification (MSC) (developed by the American Mathematical Society).

The what now?

So I obviously Googled it. And not only did it lead me to a fascinating system of subject classification for mathematics, it also led me to a sister-classification scheme put out by the ACM, the Computing ClassificationSystem (CCM), and to the Physics and Astronomy Classification Scheme (PACS) (released by the American Institute of Physics).

Now, all of these were developed at different times. The PACS was not developed until 1975, but the CCS was released in 1964. It’s unclear, from just a cursory search through the internet, when the MSC was developed, but someone wrote “It’s been around almost as long as the AMS.” The AMS was begun in 1888, so that would be a LONG time.

So what do they do? Well, fair reader, they are used to classify academic papers so that poor researchers can make heads or tails of what the author wants their work to relate to. I personally think this is genius. Let the author tell you what their work is about! There is only one primary subject class allowed per paper (for the MSC, I don’t know about the others), but there may be several secondary subjects assigned if the author feels that it is pertinent.

I cannot believe I’ve never heard of these systems before, but I am thoroughly fascinated. The whole idea behind allowing, or even *requiring* the author to classify their own research is such a good one. I used to do that, when I was an original cataloger and was doing original cataloging for professors at the university. I would just email them, tell them what I was doing, and ask if they had any special requests for the subjects of their books. First, it generated a ton of good will from the faculty, but it also helped me, because we had a lot of philosophy faculty and no matter how good of a cataloger you are, it's hard to figure out what they're talking about sometimes. 

Anyway, these schemes are something interesting and useful and different.

Searchability

Interoperability is a buzz word. A really, really important buzzword (unlike, say, "paradigm"). And I feel like I'm banging my head against it. I know the old saying: "There's the right way, and then there's our way." I think that this applies to this institution's approach to digitization projects.
Now, don't get me wrong: this place is the most awesomely together place I've ever worked with regards to digitization projects. They have very clear projects and expectations, if perhaps not quite enough staff to go around. But that's a common problem everywhere, and we all know it.

But the more we talk about this new project, the less happy I am with the way we're communicating. The TEI initiative isn't meshing with the metadata initiative, and I feel like, while they're not exactly working at cross-purposes, they're certainly duplicating work and ultimately making things harder for a user. The TEI people have no concept of controlled vocabularies, and the metadata folks are certainly not going to give into the natural language camp, and the more I think about it, the less I like the idea of one side doing their thing and the other side doing their thing, isolated.

So how do we get ourselves out of this predicament?

I'm teaching a class soon on the basics of cataloging for non-librarians. I'm hoping that this helps to clarify, for these natural-language people, just where we metadata folks are coming from in our need for controlling everything, and also how beneficial it can be to control terms and names and places. I think that many users never really understand how much controlled headings help them. Someone the other day asked me why we "bother" with controlling names or subjects, "when Google is right there and you can just let the software do that stuff for you." I think this person didn't really know what he was suggesting. THe beauty of the controlled heading is that I can put in something like "Dostoevsky" and get all the OTHER versions of Dostoevsky's name as well. Or that I can type in "New Amsterdam" and get the references to New York City. These are things that people think "software" can do, but in reality, it can't. Someone still has to map these things out in order for the references to exist.

So when the TEI people say "well, can't we just put Emperor Maximilian" and everyone will know what they're looking at?" I can honestly say "No--because what about the people who just write Maximillian, or the people who are looking for the emperor of the Holy Roman Empire, or the people who are looking for the "emperor" of Mexico? Or the prince of Baden? Or Maximilian the saint?"

 If there's an easy way to solve the problem of searchability...I can't wait to learn about it. But for now, we're going to have to settle for interoperability, and making our metadata and TEI mesh in very concrete ways. And we're not at that point yet, unfortunately.

Thomas Mann's Response to the Working Group

This is big right now...Thomas Mann wrote a paper about the Working Group's paper on RDA that they presented back in November. There's another response to his work from the autocat listserv here.

I’m just going to be taking the things I highlighted from his work and talking about them. Not very systematic, but hopefully it acts as a good guide for me personally when I go back through the work. This is a long post, although I think its worth it for anyone who hasn't read the thing yet.

Please keep in mind that I just love anecdotal evidence, unlike many people who do not believe a thing unless it has a graph. I think that a reference librarian with 30+ years of experience has the right to make observations about users, without doing a study on it first. But I'm very "unscientific" that way. So here goes!

Pg. 11: “the goal of cataloging is not merely to provide researchers with ‘something quickly’…its purpose, first and foremost, is to show ‘what the library has’—i.e. in its own local collections, onsite.”
Amen, Thomas. I don’t understand when or how the idea came about, that libraries are not just responsible for their own holdings, but for the entire scope of human knowledge everywhere. If that was the case, we wouldn’t keep physical books at all; we’d be….um, OCLC? Google? A wish list?

Pg.16: “a major weakness of word clouds is that they cannot show cross-references, scope notes, or further subdivisions of their own terms…we must remain clear about the differences between catalog search environments and Web search environments.”
I think that this is important…we get so excited, as librarians, to see word clouds, that we forget that we are the not the user. The user will see a word cloud and think “oh, more words like what I just typed.” A librarian sees a word cloud and thinks “oh, they took the subdivisions from LCSH that relate to my broad term and made those into a word cloud.” Users don’t see relationships in the way that we do.

Pg. 17: paraphrasing here: the Working Group is not being academically rigorous in its research. They are not using the scholarship that already exists, and are reinventing the wheel when it comes to thinking about subject access. We should probably all read the reports that Mann has put out there in these pages.
Also, he makes the point that OCLC has been funding a lot of the research that results in the support of facetization, “whose own WorldCat cannot display either cross-references or browse-menus of precoordinated terms. Why…should the rest of us naively accept OCLC’s oversimplified software to begin with?” Why indeed.

Pg.18: “the first responsibility of LC is to catalog its own—and the nation’s—unique copyright-deposit collection.” This is like page 11, but it bears repeating.

Pg: 20: ”contrary to the widely touted mantra, facetization does not “make the data work harder”; it makes the user work harder…it is a stunning violation of the Principle of Least Effort in information-seeking behavior. ‘Least effort’ is supposed to refer to the level of work done by the user, not the catalogers.”
Wow. Just….wow. He’s right, he really is. Yes, he ignores the basic funding issues that all libraries have (although he does talk more about the cost of cataloging in other places, and makes good arguments against downsizing at LC), but he’s still right. Our job is not to make ourselves as lazy as possible about cataloging and foist all the effort onto the user. That’s actually supposed to be the opposite of what we do.

Pg 21: “Anyone who has ever done a Google search knows that Google’s search mechanism exacerbate rather than solve…problems of information overload that are now created and aggravated by computer and web-environment retrievals.”
His argument is that LCSH avoids those problems. I agree, at least a little. LCSH is certainly better than Google, with the caveat that you have to learn about LCSH to use it, and with Google...you can get away with never learning about it at all.

Pg. 24: Accuse me of soundbites, I don’t care. This is gold: “it is undeniably true that the LCSH system is complex—but so is the literature of the entire world, on all subjects and in all languages and from all time periods, that is has to categorize, standardize, and inter-relate….the complexity of the world’s book literature is a rock-bottom reality that will not vanish simply because neither the Working Group nor LC management wishes to pay for professional catalogers.”

Pg. 34: He moves on to talk about reference work and the user, to great effect, I think: “Most researchers, when left to their own devices, are quite unsophisticated in doing computer searches…what [the user] prefers [keyword searching]…is based on a serious misunderstanding of what their “preferred” search technique is actually capable of delivering.” I think this is another case of librarians not being users, but some librarians not understanding that. The average user does not understand that a keyword search does not bring up everything. They don’t even understand the difference between a browse search and keyword search. You may think I’m kidding, but I’ve talked to enough college students to know that.

And the last sentence: “If the Library of Congress succeeds in dumbing down its own subject cataloging operations through this reorganization, there will be serious negative consequences for all American scholars who wish to pursue their topics comprehensively and at in-depth research levels, and for libraries in every Congressional District whose financial constraints make them more dependent than ever on the continued supply of quality subject cataloging from the Library of Congress.”

LCSH v. techies

There's a big digital project in the works here at The New Job. They're digitizing something like 400 works or pieces, and then some of us in the cataloging department are charged with creating the metadata. Not from scratch or anything, of course--the works are originally out of the archives here, so there's some basic metadata available. I've been meeting with people about this project a lot in the past few days, since I am the Metadata Librarian.

And I finally think I have a grasp on what it means to the be the Metadata Librarian. My job is to make sure that the catalogers don't feel like they're selling their souls, and that the digital people don't feel like they're being nickel and dimed by the catalogers. Case in point: LCSH.

This new digital project is going to be pretty cool--two institutions working together to create a federated search portal that other libraries/archives will be able to use, as well, in the future, all under one umbrella. It's not the most groundbreaking piece of technology I've seen, but still. It's neat that they're doing it.

They did a pilot metadata creation thing a few weeks back, as I understand it. One cataloger told me that they were given 6 days (really four, since two of the days were a weekend) to create metadata on 35 records. No big deal, right? Wrong. Apparently the metadata includes LCSH. And let's not forget, all the catalogers here have their "real" jobs, where they do all the other cataloging that needs to be done.
So the catalogers are all in a tizzy because they think (perhaps rightly) that the digitization people just don't get how long it takes to do subject analysis, not to mention filling in the other blanks in the metadata record. Oh, and did I mention that the catalogers didn't have anything to look at while they cataloged? The digitization people didn't think that the catalogers needed to see any of the pieces in order to catalog. How does subject analysis get done when all you have is a title?

Now, on the other side of this, the digitization people (this includes the project manager), think that the catalogers are exaggerating how long it takes to do things, and that their time table is going to get screwed up if the catalogers keep insisting on needing more things and more time. I think that the digitization folks believed that the metadata and digitization would be done concurrently, or even that the metadata could be done BEFORE the items were digitized. This is of course possible...but only if, as one cataloger said to me "we go upstairs with a notepad and catalog it by hand in front of the original."

I've already come up with several solutions in my head for this, and I think this is why they hired me. I like creating compromise. But that's not the "biggest" problem.
The biggest problem is that the digitization folks have now started messing with the LCSH field. They have started asking for non-LCSH terms to be used in that field. The catalogers are horrified, of course. I'm kind of horrified, too, but not because LCSH is so inviolate. More because the non-LCSH term they want is not a "subject" at all. It's a type of material. But I think I have an answer for that, too, if I can phrase it correctly. Then everybody wins. We have a meeting today; we'll see how it goes. Considering that I'm totally new, they might not even want me to speak at all. :)

Networking and Subject Headings

I got to meet some new people yesterday. All of them were technical service librarians/digital librarians. And twice I heard the same comment/question: "What do you think about folksomonies?"
Um...?
I think that they're a fad? I think that people only use them because they have no idea that other subject searching is available? I think that LCSH needs to stop being a browsing list?
I got the feeling, though, that the idea of controlled language "death" is very scary for librarians.
And then TODAY, I see this:
University of Chicago Libraries

Which is EXACTLY what I've been thinking about! Do a search in the UC catalog now, and you get not only the list of things the catalog thinks you might want, but the ability to refine that search within the LC classification schema. We have the classification scheme already laid out for us, which roughly corresponds to the LCSH , and why not use it to help make LCSH more hierarchical? I've been gushing over AAT since as long as I can remember, because it takes the headings and makes them hierarchical. I can actually use the headings to help me find more headings! What a concept!
Now, obviously LCSH hasn't always been this way. The books are actually pretty useful when it comes to finding other headings that might be useful. But when everything went online, we really lost that ability. There aren't nearly as many cross-references anymore, or see alsos.
We need to reclaim that heritage, and make our LCSH work FOR us again, instead of against us, and I think that folksonomies will end up following. All people really need is a way to understand a system for them to use it.
I mean, if enough people adopt it, everyone knows what "h8r" means, right? Why not understand subject headings?