Some interesting and talented people were there.
That's the good news.

The bad news is, there also were a lot of bad talks
(I did not like Rivest's talk about my own work, a fact which
particularly grated) and the press did not show up.
That's a pity sonce there were severla things that really deserved
press.

CA secretary of state and heroine D. Bowen was there
listening to the talks. I sat next to her at lunch and tried
my best to give her the goods. But she's very knowledgeable.
I'm impressed.
She said shortest splitline would be illegal under the voting
rights act. Basically her argument was that gerrymandering
is mandated by the VRA to create "majority minority"
districts and splitline would prevent that hence is illegal, QED.

She says she wants to try to uniformize voting regulations throughout
CA. We'll see. She has a lot of power to tell people in the CA
election system what to do. The voting machine manufacturers
all are ganged up against her trying to get her bad press.

Ka-Ping Yee was there with a nice idea in his talk:
make voting machines have very simple software because most
everything they do wil be precomputed by a different machine that
prepares ballot images, audio, etc. The software on the voting
machine just regurgitates it. So it can be simple enough to prove
valid. That's Ka-Ping Yee's design principle. It's clearly good as
far as it goes.

Also very good talks were those showing how to hack voting machines
and demonstrating the absolutely awesome level of incompetence among
their manufacturers/designers. (It is so hard to be that bad...)

Josh Benaloh from Microsoft had a very interesting simple talk
too about how to make crypto-secure voting schemes work even WITHOUT
voters having their own "trusted digital assistants." This overcomes
a big hurdle, to the extent Benaloh succeeeded.

Essentially it works like this. Voter goes to govt machine
(which does not know who she is). Enters vote.
Machine spits out encrypted version of vote.
The voter now has a CHOICE:
(a) take that vote and use it to vote (gets posted on bulletin
board next to her name)- using standard crypto protocols that have
been around for some time;
(b) take that vote and demand machine now decrypt it, with proof
of decryption validity.

This causes voters to get confident the machine does the right thing.
It causes all of us to be able to tally the election
in encrypted form too.

If the machine does the wrong thing often enough, then people
supposedly get suspicious of it and start video taping it etc.

If machine remembers encryptions, then whoe ver is behind
that evil machine could, by consulting bulletin board, break voter
privacy. So we have to trust govt computerized machine not to do
that. So Benaloh's scheme is defective about trust+vote privacy.

But anyhow, it is an interesting idea.
[There is a little more to it than I said, but that is the core.]