It looks it's shaping up to be a very good workshop. In particular, the keynote is by Dan Russell from Google, who is one of my favorite speakers.
One quite interesting component is the data challenge:
Challenge participants will have no-cost access to a large collection of almost two million newspaper articles with rich metadata generously provided for use in this challenge by The New York Times (NY Times Annotated Corpus). The focus of participation is building systems (or using existing ones) to help people search the collection interactively.