I attended Code4lib 2008 in Portland Oregon at the end of February, and I must say it was the most intense and rewarding and even thrilling conference I've ever been to. Imagine 200 coders in a room, laptop on each lap, programs running, chat flying, boosterism abounding. Lightening talks (5 minutes or less) introduced quick hacks with stunning results (none of which I have been able to replicate with my own very modest skills, but I am sure that others have had more success).
It was an honor to be speaking at that event, and I used the opportunity to present the reasoning behind the RDA Vocabularies project. I have to say that it was an easy audience because a room full of coders understands the difficulty of working with unstructured textual data, which is a lot of what we work with in libraryland. The RDA Vocabularies project is developing a set of elements (or "properties" in RDF-speak) that will facilitate a more machine-friendly approach to bibliographic data, without compromising on user-friendliness. But the main point of the Vocabularies project for me is that we can create better user services with data that plays well on the Web.
The talk was filmed, but the videos haven't been set up for streaming yet. I added text to my slides (PDF), making some of them much too wordy (it seemed to take fewer words when I was speaking it). If I learn to do videocasts (a current goal of minie), I will try to get some of this content available for viewing+listening.
(Note: Jon Phipps, code4lib attendee and programmer on the NSDL registries project, has an interesting post on using RDF triples with DC and RDA. I think I'll get it if I read it 3 or 4 more times.)