[tahoe-dev] 100 year crypto notes

[coderman]

On Tue, May 25, 2010 at 5:50 AM, Jack Lloyd <lloyd at randombit.net> wrote:
> http://tahoe-lafs.org/trac/tahoe-lafs/wiki/OneHundredYearCryptography
>
> ... intended to both record our design decisions thus far and also
> track what has yet to be decided.


regarding AES CTR with XSalsa20, was GCM considered?

as for key sizes, AES 256 might be better but only if you could
increase rounds. padlock engine offload can do 32 rounds per
instruction, though this might be a biased bound :)

future resistant signatures / asymmetric crypto are Hard (tm). will be
curious to see how that's addressed...

best regards,