Bummer. Apparently Windows Live Writer doesn't automatically save drafts. So here are my foggy recollections of this session.
The session was a bit different this year. In previous years we've mostly discussed providing services (collection development and reference to computer science researchers. This year, we had three speakers on a theme: "Data Curation and Special Libraries: Education, Trends, and Developments"
The first speaker was Bryan Heidorn from UIUC. He teaches bioinformatics and information retrieval stuff, but he primarily works now as a Program Manager in the Division of Biological Infrastructure at NSF. He provided some of the same statistics we've been hearing. Big science - has lots of data, but lots of support for taking care of the data. Little science - if less data, but many more projects, little support for taking care of the data. Ecological stuff is hard because of the scales, etc.
The second speaker was Jane Greenberg who is at the Metadata Research Center at UNC. She talked mostly about work going into Dryad, a repository for digital data from evolutionary biology. Dryad holds the data that goes along with publications and I guess the publications are going to start requiring deposit, if they haven't already. This is sort of a sweet spot for research because there's a real need to re-use data, there are multiple controlled vocabularies because of the various disciplines and purposes for the original data collection, and there's a real need to capture the data and make it available for re-re-use. One cool thing (of many she mentioned), was how the interface supported the author adding keywords from multiple vocabularies. The author enters the citation (I think - not the whole article?) and the system does some entity extraction, then goes to a "HIVE" server, and comes back with suggested terms from various different vocabularies. This is pretty typical, right? But it seemed cooler when she described it, so my words are just not enough.
The third speaker was Richard Huffine from USGS. His talk was a bit different - he talked about what they're doing, and the challenges they're working with. They do have a metadata standard, but they also have a lot of primary data in the form of maps, aerial photographs, and so forth. They're also working on some way to capture articles written by USGS authors and to make them available in a repository (he even threw out NIH-like, but probably not really).
(ok, now I'm seriously annoyed that I don't have what I captured during the session. There were some interesting questions!)
Gotta end by thanking ACM and IEEE for sponsoring lunch. That was really welcome!