More on the DES II contest.

[Trei Family]

Douglas F. Elznic and Lucky Green have both written saying 
that www.distributed.net will not be devoting their full 
effort to the DES II challenge.

This is incorrect. 

www.distributed.net is a highly centralized effort, in which
clients are almost completely controlled by the central site,
with which they are in frequent communication.

Over the past few weeks, the clients have been upgraded to a
version containing dual search cores - they can search either
rc5-64 or an arbitrary DES key.

On the morning of the 13th, as soon as the DES II challenge 
data becomes available, their servers will stop issuing 
rc5-64 blocks and start issuing DES II blocks. The clients
will all switch over to the DES II contest without any action
by the owners of the client machines.

I too would like to see a serious effort made at a 512 bit
RSA key. However, the infrastructure for this is quite a 
bit more complex than a simple bruteforce against a 
symmetric key, and the organizers of d.n don't seem to be
ready for it.

My greatest fear with DES II is that if d.n finds the key
early in the search, they may decide to sit on it until 
just before the 540 hour deadline. If they are the only
group with a credible chance of finding the key (as far as
I know, this is the case), then their payback is maximized
by by this tactic, since the time limits for the June
contest are determined by the speed with which the key is
found in January.

My personal impression is that d.n seems more motivated
by the money, which they want to further their research
on distributed computing, than they are by ideology.

The above represent my personal, private views, and
should neccesarily be attributed to my employer.

Peter Trei
trei at ziplink.net
ptrei at hotmail.com (more reliable next week)