I go to a lot of conferences. Often, there’ll be lots of posters hanging up at the same time as workshops and presentations and it’s difficult to figure out what to go and see at a given moment. Imagine if everyone had an RFID tag in their badges (at their choice) that anonymously logged their movements. Wouldn’t it be useful to know the whereabouts of people who like the same things I do? The schedules tend to be organised in blocks of an hour or two. Feeling at a loose end, I could peer at my laptop, and have the various sessions and posters ranked for me, based on guesses about which things I’d be most likely to enjoy.
Of course, when the session switches, the system’s guesses about what I should see will be basically at chance. Within a short time though, based on the number of people and length of time spent in different places, the system can start to accumulate evidence for its recommendations. And it wouldn’t be too hard to seed its guesses based on the text from the abstracts.
Conferences certainly aren’t the only or even the best example of how to use this information. But if I was a company doing this, I’d make sure the RFID tags are entirely optional, anonymize and noisify the data to assuage privacy concerns, and provide an API to access the data for free, so that enterprising folks could easily build mashups and try out alternative algorithms. Something interesting might come out of it, and it would be a fun way of generating publicity.
Any idea how much 30,000 RFID tags would cost?