I noticed that my local track, New Egypt Speedway, has booked a National Sprint Tour race for Tuesday, August 1. Talking about the NST, they actually had their first two events ever – with Steve Kinser winning both of them. I guess it doesn’t matter where Steve Kinser runs a sprint car, if he shows up, he has a great shot at winning.
The RSS feeds of new books project I am working on with Eric Thul and Heather Moulaison made it into a TCNJ press release. The press release was mostly about Eric, a student at TCNJ that worked on this project. I was even quoted!
I went to the future of code4lib breakout session. It was a good session in which we talked about the pros and cons of formalizing code4lib. We also talked about what we want to do under the code4lib umbrella in the future. Some of the ideas included a week long institute/school/workshop, a journal, receiving and funneling grants (a grant clearinghouse) and provide blueprints. There was also some feeling that a formal structure may lead to more sustainability. At the end of the session most of us seemed to agree that we may want to formalize more in the future, when the pot is at the boiling point, but it is not quite there yet. However, this was still really constructive and it is good to think about where we want to be not only tomorrow, but months and years down the road as a group.
After a wonderful lunch that included discussions about Coffee Luwak and the price of goats, Raymond Yee talked about “Teaching the Library and Information Science community how to remix information.” The class he is teaching seems really interesting. I wish I had a class like this when I was going for my Master of Library Service degree. Raymond pointed out some of the challenges with teaching this type of class. Some of these challenges include the challenges of teaching in a wireless classroom with constant connectivity and also the challenge of teaching this type of class to students with a divers technical background.
The second afternoon talk was by Devon Smith. Devon’s talk was titled “Two Paths to Interoperable Metadata.” Devon pointed out a number of issues with using XSLT. Since I have never used XLST, I’m not sure how valid they were, but his arguments seemed to make some sense, but even so I’m sure in many cases XSLT is the right tool for the right job. Devon went on to discuss using Seel instead.
The next session was “The Case for Code4Lib 501(c)3″ by Roy Tennant. Roy really talked about the future of code4lib – not just advocating that it was must be a 501(c)3. The best thing about it for me was that Roy mentioned me more then once 
. Roy brought up that a more formal code4lib organization may bring more libraries into the fray as using some of the software code4libers create, advocate, and support. Somethings code4lib already is include a mailing list, chat room, and conference. But it can be much more if we want it to be. It can be a support organization, and advocacy/lobby group, or anything else.
Some options for the structure include being a 501(c)3, a private corporation, an or just what it is right now (no formal structure). Some key advantages of a 501(c)3 include ability to get grants, personal liability protection, and more. There are also drawbacks such as “enough paperwork to choke a horse.” This work is not just when forming. Roy also talked about the advantages/disadvantages of other options including a private corporation, cooperatives, partnerships, limited liability companies, joining another 501(c)3, or keeping the status quo.
Some of the fundamental questions are what do we want to be, and how do we get there?
The first regular morning talk was given by Colleen Whitney. She talked about “Generating Recommendations in OPACs.” Some things they are looking at are using circulation data to relevance rankings and an index-based spell check. The recommendations stuff is interesting, but I’m not so sure I like the limiting of the relevance of circulation records to call number range. I’m afraid that doing so will hide the many multi-disciplinary titles that are of use — which may be the hardest to find using more traditional relevance ranking methods.
The second talk was by Robert Sanderson who discussed “Library Text Mining” of bibliographic data using Chesire3. He mostly focused on the way Chesire3 works.
The third talk of the morning, “Antanomy of aDORe,” was given by Ryan Chute. aDORe “is a write-once/read-many storage approach for Digital Objects and their constituent datastreams.”
Like Chesire3 talk earlier in this session, aDORe Archive is an interesting project – but one that I really don’t have any use for in my environment. I guess this is one of the advantages of many short 20-minute sessions, we can learn a little bit about stuff we wouldn’t know about and have very little reason to use. I would probably never go to a session on aDORe or Chesire3 if they were longer because I know I’m not going to use it anytime soon. However, it is nice to see what projects people that have different needs then someone like me are working on and get an overview of what is out there.
Day 2 of code4lib started out with a keynote talk by Thom Hickey of OCLC. One thing he pointed out is that you can make a lot of small programs in the time you can do one large project. I think his point was that you should try this small projects so you can get some successful applications running. In other words, try lots of small projects and some will work! Don’t be afraid of failure. I plan on trying to take this advice and start looking into more smaller programs I can work on that will give more more immediate results.
To continue my review of code4lib day 1, the next session was Jim Robertson’s Lipstick on a Pig presentation. Jim talked about 7 ways to make your OPAC sexier. Some of the cool things he did included adding book reviews, book covers, links to area catalogs (via ISBN), virtual new book shelfs, usage stats, links to searches to Google and Red Light Green and much, much more. One of the things pointed out is that we should “put the complexity on the back-end! Don’t torture the user!” One of the ideas I got out of Jim’s presentation was to figure out how to include table of contents from journals into the journal record views.
The next session was given by Casey Durfee. He talked about some of the advantages of OPAC add-ons being de-coupled. While the de-coupling is normally done because the OPAC is a black box that you can’t get into and is often seen as a disadvantage, it also has some advantages. These include: better scaling, ends finger pointing, upgrade proof, reusable, sets the stage for web services, and is easily testable.
The third session of the afternoon was given by Robby Robson. Robby’s talk was about “Standards, Reusability, and the Mating Habits of Learning Content.” Like all of the talks, it was well done. The thing I got most out of this presentation was some of the characteristics of (and problems with) learning content are.
After these three talks we had break out sessions where people discussed a particular topic. I went to the one about building a better user interface for federated searching and for OPACs. Basically what we agreed on was that we need more data (such as indexes) from vendors so that we can build what we want/need.
Next was the lightning talks. They went really well and it looks like we have a lot of them scheduled for today (Thursday). Besides my talk which was hampered by a broken projector, for the most part that went smoothly. What I did find out though is that 5 minutes can be more then enough time to talk about something — if you are only talking about one thing, are prepared, and have no technical glitches.
After a little housekeeping the conference part of the day was over. Many of the attendees went to a bar/pizza place downtown and shared a few pictures of local Oregon beer and some pizzas. Unsurprisingly, the beer and company was great, while the pizza didn’t live up to New Jersey standards (although it was still edible and not nearly as bad as I have had elsewhere around the country).
Looks like CART has their own version of Danica now. It will be interesting to see who (if either) gets a win first. I’d put my money on Katherine if I was going to bet on it.
* The recently married female IRL driver who has never won a race in an automobile.
Here are some of my notes from Wed, Feb 15 at code4lib:
The first session of the day was the keynote from the Georgia Pines people who are working on Evergreen – an OSS Library Management System. Their design philosophy is a lot different then the other systems out there because they started to design it to scale well from the ground up. They expect to have 300 libraries, 10 million holdings, 2 million patrons, and 25 million circulation transactions a year. That’s a big library system!They also went into the reasons why they decided to go OSS instead of purchasing a commercial system (or just making a home-grown non-OSS system in-house). Basically they did it this way because no commercial system could meet their needs and the support they would be paying for was nowhere near adequate. They went OSS in hopes that other people outside of Georgia could get involved. I really, really hope the succeed. They plan to go live over Labor Day weekend 2006. I’ll keep my fingers crossed for them.
The next session was by Art Rhyno who talked about ERPs. Basically an ERP spends a lot of time up front figuring out the work process and work-flow to help people work more efficiently — something not done in traditional library acquisition system designs (unfortunately). The key thing I got out of Art’s session though was when he talked about a toy called “Beep Beep Buggy” he used to sell. It was a simple and easy to use toy that looked nice. He said this is what users want. Another key point he made was that if we fixed the backend of the library system (acquisitions, cataloging, etc., the front end will work better too.
Dan Chudnov hen went on stage and talked about unAPI and OPA. Basically what Dan is working on is a way to copy data between web apps easily (his example was to think of it like the clipboard on a computer desktop). Seems like an interesting idea and the methodology he is using to create the OPA standard is very streamlined.
After Dan, Jeff Young talked about WikiD. WikiD is a tool to build collections using a wiki-based method. While it is an interesting concept, their is probably not much I could do with it in my current environment but I do want to look at it more.
At this point we broke for lunch. I’ll catch up on the post-lunch talks latter.
I’ll have more to write latter, but I figured I’d jot some stuff down before going off for dinner. Code4lib the conference has been excellent. All of the talks went well – even the lightning talks. Well, that’s not completely true, my lightning talk about Adding content from the OPAC into other systems wasn’t all that good since the projector stopped working — but I think I was able to make my main point anyway even if I couldn’t show off the things I wanted to. But I digress. All the other ones were. I think that the format for code4lib is a great format for the conference and the turn out has been good. I think this conference will be here for years to come.