[viff-devel] [PATCH 0 of 4] Insecure ElGamal based two player runtime

[Martin Geisler]

Ivan Damgard <ivan at daimi.au.dk> writes:

Hi everybody!

> About Claudio's latest proposal: I agree that this seems to be the
> natural solution given two parties and a homomorphic encryption
> scheme.

Okay, great -- I've started implementing the improved version while I
was back in Denmark, but I got away from it again. I hope it will be
ready this week.

Mikkel: can you make some comments and unit tests for the Paillier
code you wrote?

You can simply save and do 'hg import' on this mail (including headers
and all):

  http://article.gmane.org/gmane.comp.cryptography.viff.patches/11/raw

Then make a new commit and mail it to viff-patches. I'll then
integrate it in the patch series. I think that is the suggested work
flow, but I have no real experience yet, so I guess I'll have to ask
the Mercurial guys if there is a better way...

> As the homomorphic encryption scheme it would be interesting to try
> the scheme from mine, Martin's and Mikkel's paper. This would have
> to be the variant we propose for MPC where the order of one of the
> elements in the public key is a 2-power times a prime. It should be
> significantly faster than Paillier, although it is not clear how to
> make it actively secure in a really efficient way.

Yeah, that would be an interesting next step!

-- 
Martin Geisler