ELAG Library Mashup Workshop

So I taught a workshop on library mashups at ELAG this week. Since I haven’t been to ELAG before, I really didn’t know what I was getting info. ELAG workshops are one part preconference, one part hackfest, one part Birds of a Feather. So what I tried to do was explain mashups to the group, show some examples of library mashups, provide a list of possible data sources for library mashups, and then have the group break up and work on building mashups with Yahoo Pipes. I chose Pipes because when you’re working with such a diverse group of people its very difficult to know what their programming skill level is going to be or what languages they are comfortable with.

Overall, I thought the workshop went really well. The lecture portion went the best. I’ve got lots of experience teaching this part and have a wealth of examples to draw upon. Also a couple of people were able to show examples that they knew of. The niftiest of these was an example where CDs in the library catalog had recommendations based on Last.fm. I think people really appreciated seeing the possibilities and how their UIs could be enhanced.

The “hands-on” part of the workshop didn’t go as well as I would have liked. Some of this is because I’m really not a Yahoo Pipes expert, yet. While I could answer some questions about Pipes, others were mystifying. Also, even though Pipes is (IMHO) easier in some ways than programming from scratch, it still has a learning curve and I really needed to do better demos of Pipes to show how it works. Also, it would have been much better if I had more pretty pre-built examples that people could work from and tweak. This was in some ways my intention, but as always happens work concerns kept me from having time to play with Pipes and prepare examples.

For me the experience was quite good. I learned a lot from leading the workshop and will definately be able to take the feedback and make the next workshop I teach in this area better. It was also interesting to see what possible data sources people at ELAG had at their disposal.

Probably the most helpful part of the workshop for me was asking people who had open APIs for their catalogs and if they could provide folks with their urls and documentation. This was very helpful because there is a limited set of sources for bibliographic metadata in traditional library formats (MARCXML, MODS, DC). I also liked it because it gives me different sets of data for people to work with. All of these are posted on the workshop webpage so others can use them.

All in all, I think the group learned a lot, including what works, what doesn’t and the potential pitfalls. So in that sense the workshop was worthwhile.