In the air once again, not quite on the way home, but on the way to the Open Library developers meet in San Francisco. A first chance to see the bridge for me.
Where to start? With three and a half packed days of workshops, breakouts, prepared talks and lightning talks as well as the evening socialising there’s just too much to cover. The conference was so well organised in fact that longer standing members of the Code4Lib community wondered if they were at the right conference. Roy Tennant’s absence this year also left some folks confused and disoriented.
My time began with some folks from LibLime who ran a great pre-conference session, as did Equinox (who do a lot of the work on Evergreen); no I wasn’t in two places at once. I spent the morning listening to Josh, Galen and Henri talking about Koha and Richard spent the morning with the Equinox crew talking Evergreen. They’re both great systems and I have to say getting Koha up and running to do some dev work on it was oh so simple. Dan Chudnov also got a pre-un-conference together, reports from that were good, but I didn’t quite catch what was discussed there.
The conference proper started on Tuesday, the organisation mostly done by Jeremy Frumkin of Oregon State. His organisation was so good that when he had to leave unexpectedly on the morning of day 2 it was easy for Ed Corrado, Dan Scott, Dan Chudnov and Ross Singer to step in and hold the reigns. Ross made a fantastic MC introducing each speaker in true Oscar’s style.
But enough name-checking, what about the meat of the sessions?
I spoke this year, and I knew I had a tough act to follow when I saw that I was straight up after Brewster Kahle‘s keynote. Brewster is disarmingly witty and unassuming. Known best for his role at Internet Archive he talked about that a bit and about the Open Library project. Brewster used the occasion to announce that Talis are supporting the effort by donating the millions of records that form our 40 year-old union catalogue to Open Library. This gives Open Library a great boost of UK material including records about older rare and out-of-print stuff.
Brewster’s perspective on life seems to be that things “can’t be that hard” and that if you just make a start then you’ll see how far you can get. This approach has worked well for Internet Archive and appears to be working well for Open Library too. Perhaps we could all do with a bit more of that give-it-a-go approach in library land.
I followed Brewster with a piece about mining MARC data for relationships, a piece of R&D I’ve been working on for some time now. The Code4Lib channel occupants asked me to assume no prior knowledge of RDF, which we’ve been using to describe the relationships, and a load of people told me it was the first time they understood what RDF was about, so I’m chuffed with that. I’ll be presenting a very different side of that work at WWW2008 in April.
In the lightning talks, Andrew Bullen stunned us all with a beautiful piece, set to music, about the robber baron George Pullman, his carriages and the most wonderful music. He’s been scanning the sheet music of Pullman car classics, converting them to midi and releasing them in an archive of midi and MP3 files. Beautiful work that got him huge applause. There is a suggestion that his work should appear in a Code4Lib Journal article soon; that would make a nice sequel to his article on Historical Demographic Data in the town of Pullman.
Jodi Schneider, Ed Corrado and Jonathan Brinley talked us through the Code4Lib Journal, a fully-fledged journal (with its own ISSN, even!). The journal works through a small committee of editors and technical staff to produce a web-based journal once each quarter; they’re looking for new members for the team right now.
David Walker then covered the WorldCat API nicely in his talk about working with it to create a prototype local catalogue, very similar to what we’ve seen of WorldCat Local. It’s great to see OCLC letting their customers get at their data through this kind of API. The Grid services look like a great step forward for those who are members of the OCLC club.
Karen Coyle keynoted day two in her inimitable style, getting cheers, laughs and the occasional grumble of dissatisfied agreement from her audience. If there were ever an audience in almost total appreciation of Karen’s perspective, efforts and humour it would be Code4Lib. Taking us on a whistle-stop tour of RDA, RDF, the efforts of the RDA/DCMI folks and a sideways dig at Michael Gorman (best known in Code4Lib for Revenge of the Blog People!) she had us all hanging on her every word.
Karen’s talk included the introduction of FUQ lists (frequently unanswered questions), pronounced somewhat dangerously (see photo, right). You really have to watch the video (when it’s published) to do Karen’s keynote justice!
Corey Harper also did a piece on RDF/RDA toward the end of day 3; he did a great job at not repeating Karen (or me) and really got a lot of people further interested in why the DCAM work is really important.
Biblios, presented by Chris Catalfo, followed Karen and showed us all how good a web-based cataloguing interface really could be. I saw Biblios in its early stages, when it showed great promise. It now provides a slick, standalone cataloguing UI with plugins at the back to allow records to be stored and retrieved in different stores; that means it can be bolted onto any ILS. Very nice work Chris.
Emily Lynema and Terry Reese presented on the DLF ILS work to establish a common API spec across all ILSs, work we’ve been starting on here with Jangle. Ross did a great job of talking to various folks about how Jangle could fit in with the DLF efforts and Terry and Emily ran a fantastic breakout session for interested people to see how we can all drive forward with the DLF effort. This work follows on nicely from Emily’s Free The Data presentation from last year. Emily and Tito Sierra also did a slot on Talking with Talis last year.
Aaron Swartz did a great slot on the building of Open Library, including an explanation of ThingDB, this had a lot in common with Dan Scott’s talk on CouchDB. Both talks explained a lot about the benefits of extensible schema and both have a lot in common with RDF stores.
The chatter and banter in the Code4Lib IRC channel was, as always, inciteful, witty and good humored. I’m always pleased to see how well the smaller Code4Lib channel community opens up to the larger group of conference attendees – while there were several in-jokes floating around for the three days, everyone was welcomed and everyone supported in presenting.
That’s what really makes Code4Lib a very different style of conference. Around two-thirds of attendees also stood up to present, whether a session, a breakout or a lightning talk and that’s truly extra-ordinary.
To get a more emotional feel for the conference, you should check out the photos of code4lib 2008 on Flickr
Thanks to the efforts of Noel Pedens, videos will appear here sometime soon.
Next year is in Providence, RI, hosted by Brown University – thank goodness it’s on the east coast.