I spent yesterday afternoon and this morning at an advisory board meeting for the IMLS Digital Collections and Content project, lead by UIUC. I'm sitting next to Jeremy Frumkin, who was able to blog briefly about the project while we were sitting here, so I took that as a challenge that I should be writing up my thoughts on these issues as well. (Poor lonely neglected blog - if it were a house plant it would be all dried up! I'm not out of ideas by any means, what I am out of right now is energy.)
The IMLS DCC project is starting a new phase concentrating heavily on understanding what collection descriptions really are, and how they could be used to improve retrieval of items within them. It's a researcher-driven project, with most project leaders in the library school at UIUC. There are a few investigators representing digital library practitioners as well, and the advisory board reflects a similar diversity. In the LIS field in general, there's a pretty wide gulf between researchers and practitioners, but I've aways considered UIUC as one of the places where the situation is better than most. This project shows some of that separation - looking at the problem from both the theory and practice perspectives, and hoping to meet in the middle along the way. I see many potential pitfalls in this, but I also see paths that could work.
I'm overall very interested (and concerned, because I don't see a lot of good activity in this area!) about how we move practitioners (such as myself!) towards more consistent and useful work without all of us having to become researchers in the theoretical realm. I'm very interested in the theoretical research in areas like the ones DCC is studying, and see value (and fun!) in figuring things out just for the purpose of figuring them out. But we need better bridges between that theoretical work and how the greater understanding gained from it could be used to build better products. I think the IMLS DCC project is really trying to do that, and hope that the practitioners on the project staff (and advisory board, like me!) are able to help them reach the practitioner community that needs to hear it. I see this with my own work, thinking, "well we published a paper, what more do they want?" but I've found that's not enough. Some combination of publishing, conference papers, informal distribution like listservs and blogs, plus the crucial step of showing a concrete (if test) system that illustrates a research result is necessary. And possibly other mechanisms as well - I don't claim to know exactly how to do this, I'm just muddling through like the rest of us. But it's something I think it's worth our time to work on.