Wednesday, August 23, 2000

Rump Session

I just got back from this conference's rump session. I'm not sure how widespread the "rump session" is in the scientific and technical world, but it consists of all the presentations too trivial, too bad, too new, or just too silly to make it into the regular program of the conference.

In my mind, it has two very important innovations which should give it an edge over the rest of the conference: very short talks and alcohol. The two achieved a very nice synergy in the first talk tonight, "AES Update," a 4-minute talk. The speaker started by giving a history of the AES process (don't ask). Someone shouted out, "We're all cryptographers; we know what AES is." Finally, he was heckled into sticking to the "update" part.

At its best, rump session talks consist of someone madly dashing through a quick, but interesting result -- like breaking a cryptosystem presented at one of yesterday's regular talks. Madly dashing through a boring result also works...the comic effect is still enjoyable. Unfortunately, there was too much plodding through really, really boring stuff. People reading from their slides in a monotone and not being cut off by the moderator when their time is up. Even 5 minutes of that can be deadly.

So I've come up with an idea I think can improve these rump sessions. I think they should be modeled after one of modern entertainment's greatest triumphs -- the Gong Show. Get a panel of noted cryptographers -- and Jaye P. Morgan -- give them gongs, stand back, and watch the fun begin. This would have a number of advantages. One, anything exceptionally boring would be cut off by the dreaded gong. Two, speakers would have even more incentive to entertain the audience for fear of getting cut off. Three, those particularly interesting talks -- the ones that leave the audience wanting more -- could continue on as long as they kept the judges amused. But most importantly, we could all go home earlier.

No comments: