In the five years I’ve been a part of libraryland, one conversation seems never to die: How can we improve our online catalogs? There are so many ways to approach this question, from a user-interface perspective or a metadata perspective, a software architecture perspective or a task-based perspective. But there is an assumption at work that not enough people have raised, and that I think deserves more consideration. Do we even need our online catalogs anymore?
The primary issue raised around our catalogs is their general user unfriendliness, from the terminology that we use to the overload of hard-to-decipher information that displays. Because of how library metadata is organized and silo-ed, our catalogs don’t provide access to the entirety of the resources available, and while modern discovery systems go some way toward alleviating that problem, they aren’t perfect and often just add to the chaos in the discovery landscape (especially since we’re all so keen to brand them and give them unique names).
The fact is that when a person is looking for information, they aren’t likely to go to the library first. We might have just the resource they’re looking for, but if that resource doesn’t surface in a Google search, they aren’t going to know about it. Students might be more likely to turn to the library for information, largely because they’re encouraged to do so by their professors, but they’re still much more likely to start online, outside of the library’s web presence. Clearly, what we need to do is to get our information out of our little corners of the web and into the wider web.
This is the promise of linked data. Yes, it’s The Buzzword of 2013 already. Every conference I attend has double the number of linked data presentations as the one before. And a lot of people still seem kind of fuzzy on what a linked data system would look like for libraries. I think part of the problem is that we’re still thinking about our System, and the promise of linked data is that the closed library discovery system could finally disappear.
I’m not saying our internal management systems would go away. Obviously, we still need those. But our users don’t need those, at least not for discovery. If we use linked data principles to describe our holdings, ALL of our collections and not just the items that are in our catalogs, that information will finally be able to be surfaced through web searches. Imagine searching for a book title in Google, and seeing not only the familiar Amazon link that pops up, but also a link to your local library, with a call number, availability information, and maybe even a link to Request an item, right there from the search results screen. Imagine searching on a subject topic using your favored search engine, whatever it might be, and finding a handful of resources at your local library right near the top of the list, including books, available journal articles, and archival material.
So far the focus of discussions around linked data has been on putting bibliographic metadata on the web. And that’s great, but once that’s done, it’s done. We don’t all need to do that. What we need to do is put our ownership information on the web. We need to link our ownership and availability information to a centralized bibliographic database (or a few), and make it available on the web for indexing by search engines.
Why do we still force our users to come to our special web sites to find what they are looking for? Why do we still keep thousands of copies of the same bibliographic metadata in thousands of databases around the world? Our data isn’t findable or usable right now by the people we’re ostensibly collecting it for. Our primary concern, as we talk about updating our metadata and our systems and our bibliographic framework, is how can we get all of this wonderful stuff that we own, and provide to people for free, out on the web where they are already living and working. Let’s stop trying to reinvent our own little corner of the web, and join in the game where everyone else is playing.