CIL2008: User-Generated Content

Roy Tennant
Not an overview of ways users are creating content. If you want that, go buy Social Software in Libraries by Meredith Farkas. Focus will be on user-generated content on library managed sites.
Roy’s tenets for user-generated content: More content is better. More access is better. Can provide more personalized service. Can foster interaction and community. We don’t know everything — we don’t know all we can know about our own collections. Our users can help remedy this. More data trumps better algorithms. (Google learned that the more data you have, the better your algorithms are. Code can’t make up for lack of data.)
Contributions of content. Institutional Repositories are a collecting point for user-created content. (This is often not thought of as a user-generated source.) Even if faculty aren’t doing it themselves, faculty are still getting their content into the library.
Kete.net is an open repository for whatever anyone wants to contribute. (Kete developed by the folks who did Koha.) They’re digitizing the Cyclopedia of New Zealand and are transcribing text. Also enabled software to handle genealogical information well. So a community can start to get a handle on genealogical past.
Descriptive contributions. Example of the Great Lakes Images, where they post photos and get community members to fill in details (names of subjects in photo, places, etc.) Library of Congress’s Flickr project is similar. 5.4 million views of content in first month. Immensely successful.
What has LC accomplished? Higher profile for collections. Enabled community engagement. And corrected metadata. But more importantly, sparked comments and conversation around the images being tagged. People became very involved in the images. And higher visibility for LC blog. Boston Public has done this, too. But they’ve had less traffic than LC.
Exploits knowledge of the masses. Library staff may not be closely connected to the collections they manage. They may not know much about the specific collections being featured. Web offers a feedback loop.
Bookspace at Hennepin County Library — offers community space around books. Has readers’ lists — on wide range of subjects, created by library users. Also guides by librarians; these are likely less specific and focused (not to mention less numerous).
Tags. Uses user terminology. Even if it’s “stupid,” it’s the user’s. There’s a very low barrier to use for users — type and click. It’s useful to the tagger (or else they wouldn’t be doing it anyway). But it is also useful to others. However, tags can be redundant (for example, “blogs,” “blogging,” and “blog” are all, probably, the same). Phrases are often complicated and inconsistent. Steve is a tagging project by several museums. A few tags often get applied by many users.
LibraryThing’s Tagmash brings together tags that are really synonymous. It works “pretty darn well” for bringing together works on a similar topic. The more data you have (the more users), the better the results.
Third-party providers in this general space. SpringShare (LibGuides and LibMarks), LibraryThing for Libraries, ChiliFresh (book reviews by readers).
Things to keep in mind…
Our idea of content might not be our user’s idea. People are going to do weird things. It’s going to be messy, and that’s OK.
Need to know what your goals are. How do you distinguish between user content and library content? Will you need to moderate in some way?
We (libraries) need to do better at inviting our users in. We need to figure out how to get better at using these technologies.