Category: Search and Discovery

For the last year and a half, the five liberal arts colleges known as MnObe (pronounced min-OH-bee and derived from Minnesota Oberlin Group Libraries) have been thinking collectively about our library catalogs. We started at very different points (ranging from “what we’ve got works pretty well” to “what is this thing we’re using and where are the punch-card computers it seems to match?”). Then Roy Tennant came and talked to us and everybody got pretty excited about the idea of a next generation catalog. And I was pretty sure we’d stay that way… excited about the idea of change.

But then we held a “debate” between five teams of librarians each proposing an alternative to our current systems. Our biggest surprise that day was the amount of agreement between groups and audience members about a) the need for change and b) the types of things that need to change. (Here’s a quick list of features and functions we liked, none of which are surprising.) At the conclusion of that meeting, the directors of our libraries charged a group of volunteers to actually map out our movement forward.

Well, a couple of weeks ago, this group of volunteers met, thought, discussed, and drafted a report. You can read it on the wiki if you want, or the PDF (which is prettier). It’s definitely a first step, but the group felt that we needed to clear up the first step first. You’ll notice, though, that we clearly and explicitly state that we want some kind of drastic change in the near future, and we map out a few key actions that will start us moving toward that change.

Now it’s been delivered to the MnObe directors. We’ll see what happens. I’m having to work a lot harder to keep from getting too excited now that we’ve actually formally recommended action.

I spent all day today at a school two hours northwest of my school meeting with a bunch of librarians of all kinds (including library directors) who work at private colleges in Minnesota. On the agenda: contemplating the future of our catalogs.

Four groups of librarians presented their visions of what this future might look like using a combination of examples already extant (WorldCat in many paid and free forms, Koha, NCSU’s Endeca interface, Google, LibraryThing for Libraries, etc.) and “what if” ideas drawing from the world of Web 2.0 and the hope that search algorithms will continue to evolve. Then we broke into groups to discuss what we’d heard and add our thoughts. And finally we reported to the group any ideas, concerns, questions, or random thoughts our discussion groups had generated.

Even more interesting than these presentations, though were two revelations. First, only one of us indicated she liked our current catalog (which is not a smear against any particular catalog, since we all used different ILSs, but against the ILS landscape in general). And second, we learned that when it comes to thinking up feature and function lists, there was a large degree of consensus, regardless of school affiliation or job description. We wanted all the usual social web features, web 2.0 discovery aids, and recommendations based on user behavior… all of that. Plus we wanted the option to search the full-text of our books’ tables of contents, introductions, bibliographies, and indices.

One group even decided that we need to ditch the name “Catalog.” In fact, we were so busy ditching names that all we could come up with was “up-pushing thing” (based on the idea that whatever tool or system of tools we come up with should be able to push relevant/important results up from a pool of articles, books, archival material, images, recordings, web pages, and anything else the library might care to point people toward). Oh, and this tool’s uber-descriptive name is only proper if accompanied by a hand motion that would take far too long to describe, but involves moving upward…

We still aren’t quite sure about a couple of things (the role of metadata, privacy, how long some of the transitional technologies and solutions we identified might need to be kept around, etc.). But we’ve only met once…

At the end of the day, the directors of our respective libraries stood up and reported where each library stands in respect to its ILS. These ranged from “thinking about thinking about changing” to “actively exploring alternatives,” but it was nice to get that up front. It was especially nice to get that cleared up because after that they called for volunteers to work on behalf of all of these schools to explore future options. And somehow (not quite sure what possessed me…) I ended up on this Think Group with 7 or 8 other librarians from around the state.

So here I am… thinking about thinking about the future of OUR catalog. I’m a little worried that this will be a lot of work, and that it’ll turn out we don’t actually do anything. (And I DEFINITELY don’t believe what one director said: “think as if money is no object.” We tried that for another and much smaller decision and got burned. Money is always an object.) But at the very least, I can contribute my year’s worth of accumulated information on this topic, and maybe set up a wiki for us…

Ex Libris vendors showed me Primo while I was at CIL. It workes on top of Metalib and your catalog to function as a search of everything available through the library. Shiny. Probably the flashiest thing they showed was opening a video and a text version of the same title in two side-by-side windows. If it weren’t for one small problem, I’d say I’d seen the future of library discovery systems.

The problem? After 15 or 20 minutes of talking to these vendors, I finally got them to understand the problem of searching subject and generic databases at the same time. Subject databases never index their primary subject, right? So “psychology” would return gazillions of results from catalogs, Academic Search Premier, ProQuest, etc. But it wouldn’t return anything from PsychInfo. That’s a problem.

When I’d finally gotten them to understand what I was saying, they agreed. But here are their answers: create custom views or scopes for each subject (so yes, make students chose their subjects first and then do searches, which still won’t compute if they search for “psychology”), and tell PsychInfo to include “psychology” as a subject term. What? That’s about as likely as catalog vendors moving their products to wiki platforms.

Why doesn’t Metalib or any other federated search system allow us to tag databases and then add those tags to the results that get fed from the database to the search system? That way we could customize for our communities AND get our patrons the most relevant results possible.