I have put myself in an interesting position (which is a euphemism for “I’m trying not to freak out too much”). I’m giving a presentation in a few weeks, and I’m not sure how much I believe one of the main points of my presentation anymore.

On June 1 at the Colorado Academic Library Summit, I’ll be doing a presentation called Web 2.0 and the Digital Library (or, Learning from Flickr). This will be a re-purposed version of the talk I gave last October to a group of librarians, visual resources curators, information technology specialists, and faculty which I blogged briefly here.

Here’s what I told ‘em I was going to tell ‘em:

This presentation will look at the digital library through the lens of the popular web 2.0 site, Flickr , to see how better use of user-created content and metadata (such as tags, comments, and notes) and a more predictable, programmable interface (through feeds, application programming interfaces (APIs), and better URLs) can help us create a more useful and usable digital library. The discussion will be on the conceptual level (i.e., no screens full of code).

The October talk was aimed at people at liberal arts colleges in the early stages of putting together digital image collections. I looked at some aspects of Flickr–specifically tags, human-readable URLs, comments & notes, feeds, and the API–and described how I thought they could be applied in this context, often in contrast to the way things work in the digital image project I’m affiliated with, the Image Database to Enhance Asian Studies (IDEAS).

In many ways that talk was pie-in-the-sky, as it would not be trivial to implement any of those Flickr features, but I thought it was important to look at what Flickr is doing now to get a clue as to what we might be able to do some time from now.

In that presentation, I put a lot of emphasis on the value that users provide through tagging and comments. But I’m starting to lose faith that this kind of user-contributed metadata will ever be that useful in the library context, at least in the way that I had been thinking about it.

The problem with tagging is that people don’t have enough motivation to tag something that isn’t “theirs.” I have often returned over the past year to The Del.icio.us Lesson; here’s the main argument:

The one major idea behind the Del.icio.us Lesson is that personal value precedes network value. What this means is that if we are to build networks of value, then each person on the network needs to find value for themselves before they can contribute value to the network. In the case of Del.icio.us, people find value saving their personal bookmarks first and foremost. All other usage is secondary.

Tim Spalding takes up this same argument in When tags work and when they don’t: Amazon and LibraryThing. I’m afraid that library applications–be they the catalog, a digital image collection, or something else–will feel more like the “Amazon” experience to users than the “LibraryThing” experience.

It is way too early to judge the success of the tagging features of the Ann Arbor District Library Catalog and PennTags, but when I see that most of the “top tags” in those applications have been used fewer than 100 times, I start to wonder if library and academic users have the motivation to tag.

And I wonder how many users a project would need in order to useful. I keep thinking about our new systems librarian, Carol Ou, commenting that she didn’t think Penn’s user community was large enough to create a viable folksonomy. If that’s true, how about a smaller digital library project where the user community might number in the hundreds? (Carol has since said that she thinks smaller groups of taggers can have a greater effect in areas like “clustering” of search results; you can bet I’ll be talking to her some more before this presentation.)

The academic caste system is likely to complicate matters further. One of the examples I used in this talk in October is a Flickr photoset of Soviet-era posters. The comments on that set and on the individual photos go way beyond the typical “wow cool” as Flickr users provide additional information about the posters, including translations and corrections of errors made by the person who originally uploaded the images.

But I wonder if that would happen in an academic context. Would faculty contribute photographs to an image collection if they knew that they could be “corrected” by students? Would students feel comfortable offering up their comments in the first place?

In any case, I’m sure there is still a lot we can learn from Flickr and similar applications–the sections of my talk on URLs, feeds, and the API still hold up basically unchanged, and I’m sure I can find other aspects, too.

So help me out: what can we learn from Flickr? Is user-tagging of the digital library too much to hope for?