John Miedema

Many library people called attention to an article in the Guardian this past Thursday, Why you can’t find a library book in your search engine. The writer observes that finding books in your library should be a simple matter of a web search. This is not the case. You can look up books in OCLC’s WorldCat but most of these records are not shared with Google Books. Instead of trying to find new ways to share library records, OCLC recently introduced more restrictive licensing on them. Set to be implemented in February, the library community roared, and the date was deferred. It is a time for OCLC to pause and reflect.

Open Library represents a different way of doing business. Instead of concealing their book records, Open Library has APIs that let webpages freely consume their data, which then gets indexed by search engines. The Guardian article makes what must be an indirect reference to my OpenBook plugin as an example. OpenBook inserts book covers, titles, authors and publishers on web pages, using data from Open Library. OpenBook also inserts a link to the WorldCat record, though I have never been satisfied with it. You see, my hometown public libraries are not in WorldCat. Jessamyn West has observed that tiny rural libraries do not have the budget for the annual OCLC membership fees, and so they too are excluded. Other libraries opt out for other reasons.

I have begun work on a software project that uses web design patterns and “fuzzy logic” to perform a similar function to WorldCat, hence “FuzzyCat”, except this critter goes where WorldCat does not. It is intended to work with any number of library catalogues, membership not required. You can read more about it on this new FuzzyCat page on my website. You can try the early prototype with three libraries that are not in WorldCat — two are public libraries near me, and one is West’s library in Vermont. And of course, you can download the code — it’s open source. Partnered with OpenBook or any other similar tool, it will create widely distributed links from webpages to library catalogues, available to be indexed by search engines.

Look, I am just one web developer writing this code on the side. My expectations for FuzzyCat are small. My real goal is to build software that makes a good point. OCLC, on the other hand, is large, well-resourced, and can make a real difference. I think OCLC is a good idea. I think WorldCat is a good idea. I just think they have their business model wrong. I’m kicking off FuzzyCat with a grin and a wink, but also as a serious invitation to OCLC to openness. Libraries are meant to share information as widely and generously as possible. In this time of reflection, think a bit about FuzzyCat, and how openness helps libraries.

Update Feb 3, 2009: Corrections and clarifications in the Guardian. But once a library shares its records with WorldCat, are they not then WorldCat records and so affected by the more restrictive licensing? And if Open Library uses those records, its growth is in fact obstructed. No claim of expertise in this area on my end, but it sounds like PR to me.