I just returned from the Sunbelt conference in Riva del Garda, Italy, where I presented my paper with Andrea Knecht on alcohol use as a coordination game, and Michal presented our paper on segregation measures. Here are some random thoughts on the conference:
- This edition was the largest Sunbelt ever. The sheer number of talks and the dense program (up to nine sessions in parallel) make it at the same time more necessary and more difficult to select the really interesting ones. The problem is aggravated by the fact that it is often quite hard to grasp from the title whether a talk will be worth listening to (especially with the tough competition of the great weather, the surroundings, and last but not least, the World Cup). One strategy is to simply pick the authors you can trust to deliver quality, but this would of course favor the established names (another Matthew effect).
- Related to the previous point: I saw a number of talks on ‘social’ recommendation systems. Would it be feasible to come up with a social recommendation system for conference talks? Something like: “other participants who share your research interests plan to see…..”.
- Research on online social networks is hot. This year’s program had no less than three sessions on Twitter networks and four sessions on online social networks, while I also saw talks on online networks in other sessions. Is it just that the data are so easily available, or is there really something there? Maybe I should get into it too ;).