Crowdsourcing metadata at the Bodleian

"The Cleopatra Galop" by Charles d'Albert

Tomorrow, as part of my library summer seminar in England, I’m visiting the Bodleian library in Oxford. One thing I’m interested in asking about is their “What’s the score at the Bodleian?” project. The Bodleian has a huge collection of 19th-century sheet music (back when printed was the only transportable medium for music) that they would like to digitize, but they don’t have the resources to do so. “What’s the problem?” you might ask, “Don’t they just need to run it through a scanner?” Well, the scanning part is not the time-and-money crunch with digitization projects—it’s the metadata.

Metadata is the information that is attached to the digital picture. in the case of sheet music, useful metadata might be the title of the work, composer, the lyricist, a description of the cover, and publication information. Metadata makes digitized images searchable and therefore findable and useful. Sheet music metadata, luckily, is relatively easy to locate and enter, even for non-musicians, and so the Bodleian has decided to crowdsource the project. They are putting this digitized sheet music online and letting volunteer digitizers fill in the metadata.

Not really free

With this project, the Bodleian is following the lead of the New York Public Library, whose "What's on the menu?" project has been a huge success. So far, NYPL has digitized almost 17,000 historical menus using volunteer support. These two aren't the only volunteer digitization projects around, either. Trusting non-cataloging experts is a big leap of faith for institutions to make. But if they can figure out how to train volunteers, and a find a crowd of willing people, then they can digitize much, much faster than with only their own paid experts. Still, these projects not free; it takes some dedicated staff to coordinate and curate all the entries, and both NYPL and the Bodleian have some major funders. Even with major volunteer support, these projects would not get off the ground without this extra money.

Of course, the other reason I'm interested in "What's the score at the Bodleian?' project is that part of my job at UNC's music library is the exact same thing—minus the digitizing (If only I could get everything scanned, maybe I could get volunteers to do my job, too…) Our collection is not quite a old as the Bodleian's, mostly from 1900-1950. It is kind of fun, actually, to see what's changed and what hasn't in popular music from years ago.
 
Have any of you participated in a crowdsourced metadata digitization project? What did you think?

Vocab: metadata, lyricist, composer, crowdsource