I gave a presentation on my thesis today to the entire Media Lab (eek!) I think it went well. There was lots of good questions from my critics and readers (thank you!). I'll try to summarize:
1) David Reed seemed to think that the financial data would be self-reported by the users. Not true! Expensify grabs the transaction data directly from the banks and we do the filtering. We get direct access to the whole raw feed.
2) Sandy Pentland wanted to know why we didn't just ask for all of the data then parse out what we need after the fact. Maybe some campaigns will ask for this, but I think it would be a major hurdle for most people for adoption. My mom would be very uncomfortable giving access to all her data, especially without any constraints on what I could look at or do with it.
3) Henry Holtzmann suggested we try to aggregate as much data together as possible, even across campaigns that were not related. I tried to defend why I thought that would also be an inappropriate invasion of privacy, but outlined some scenarios were campaigns that made their data public could do something like that.