[RC5] D.net goals (Was: Global Stats)

Tom Cramer drakino at usa.net
Fri Dec 24 12:41:45 EST 1999


Let me start by saying that I picked this message out of the few posted in 
response to my original message due to the part Abigail picked at.

At 03:59 PM 12/21/1999 -0500, you wrote:
> > Date: Sat, 18 Dec 1999 01:53:44 -0700
> > From: Tom Cramer <drakino at usa.net>
> > Subject: RE: [RC5] Global Stats
> >
> > The people who insist on not running CSC for stats reasons are only 
> hurting
> > the people trying to prove something.  Maybe it's time for statbox ii to
> > break down for a short time so people might realize what the project is
> > really about.
>
>
>Actually, the RC5 stats *were* down for a long time just as the CSC
>project started. That did have an effect on some people on my team - they
>started dividing their CPU cycles between distributed.net and dcypher.net,
>and the suggestion was raised that after RC5-64 was completed, we drop
>distributed.net because they couldn't maintain their stats programs.

The comment about the stats going down was more of a comment then it was a 
wish.  As far as the stats go, they are useful, but shouldn't be a reason 
people don't participate in a certain project.  Also, with the dcypher.net, 
keep in mind that they just started a distributed system, so they don't 
have the problems d.net has with stats.  I find it amazing that d.net can 
even provide the stats they do after seeing the processing power required 
to churn out stats on my personal proxy machine.

>I don't think CSC is proving anything. What are you trying to prove? That
>56 bit keys are insecure? If you are still trying to prove that, then
>what was the point of the other contests that did 56 bit keys?

By what I understand, CSC uses some extra layers of encryption to try to 
make it secure.  By us cracking it in the quickest amount of time, we prove 
that the extra layer doesn't help as much as simply moving to a higher 
bitrate encryption.

>Furthermore, why aren't you making a case that the people who do run
>CSC are hurting the people trying to prove something with RC5-64?


I had a feeling that this would come back at me.  The reason I am not 
making a case about RC5-64 is due to the fact that the RC5 contest is not 
time limited.  If we crack RC5 one or five years down the road, the reward 
is the same.  With CSC, if we can't beat the deadline, the reward is lost, 
and all the work done has no benefit.  Even if we cracked CSC one day after 
the deadline, the publicity wouldn't be as great, and the awareness of the 
weakness of CSC wouldn't be as high.  Once D.net starts OGR, I won't 
personally care what the majority of people run.  When OGR starts, it 
introduces a new goal for D.net participants to pick from.  But for now, 
all the contests D.net has run are all encryption contests, so the force 
should be put behind the one that is time limited.  The last DES contest 
has a very short time limit, and would have been possible to loose out 
on.  If the top one hundred teams has said "Stats are important, so we have 
to keep cracking RC5 instead of DES", we might have missed the deadline.

Tom

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list