There’s a web service for that?!

One of the things that never ceases to amaze me is the amount of data that is available via web services these days. While up until this point I’ve mostly focused on bibliographic data available, I’ve been delving into APIs for newspapers like the New York Times and the Guardian. IMHO these web services can be used to bring valuable evaluative data into library interfaces. I’ve built mashups that add indicators to an pseudo catalog interface which show whether or not a book is a New York Times bestseller. Recently I’ve been trying to see if I could link to movie reviews from DVDs in a library catalog.

The Guardian also has an API and Movie and Book Reviews, but I haven’t figured out the most efficient way to get at those reviews from a title in a library record. I don’t seem to be able to search with the appropriate limits to just get reviews about a particular movie. I always seem to get a cluster of other stuff that also has those keywords but isn’t the review of the movie I want. I’d appreciate any ideas folks who’ve worked with the Guardian API are willing to share in the comments.

I’m very keen on getting metadata about films so another API which I found and have become interested in is the Netflix API.What I really want is to be able to put indicators about what awards a film has won as well as reviews. In addition, I had the idea of adding to an Identities screen, awards which actors, actresses, directors etc have won.

One of the biggest difficulties with films is making sure I get the right film. Title works reasonably well, but since films get remade, it is necessary to check the year to make sure one is linking to the appropriate version. I’m lucky because the New York Times Movie Review API includes a field for DVD Release Date which I can compare to the publication date in the MARC record. It would be SO much easier if the New York Times used a standard identifier of some sort though.

A while ago a colleague brought up the mashing up popular music data like top hit charts with library catalog data. So I did a web search and found the Billboard API. The API provides developer level access to the data about what songs have appeared on the Billboard charts, past and present. It can be searched to see if a particular song or artist was ever on the Billboard charts for a particular date range. One thing that is sort of disappointing is that I don’t see any way in the documentation to search for if a particular album appeared on a particular chart or any chart, which is sort of too bad. Digging around their site I also found that their charts also have RSS feeds available which is nice. Both Rhapsody and Last.fm have web services but Pandora does not.  Another possible source of metadata about music is MusicBrainz.

Its interesting to think about all these information sources as possible ways to enhance library interfaces. It takes time to work through the different APIs and find other sources of data but I think that libraries to think about what other data sources can be used to provide users with a better UI experience.