DES-Busting Screen Savers?

At 8:28 PM 7/22/96, aba@atlas.ex.ac.uk wrote:
Hmm, 56 bits is a lot of bits...
Here's some calcuations of my own for your criticism...
So ideally for a break you would like the whole thing to be completed in say 2 weeks wall clock time, which gives rise to the need for ~100,000 machines of similar throughput, full-time for two weeks.
Or several times that number of machines or time for machines with less crunch. Say, 100K Pentium-type machines for a month or two. How might this be gotten? A while back I proposed one approach: a brute force "screen saver" for Windows machines. Other platforms, maybe, but the most cost-effective thing to do is to go after the Windows market only. Instead of bouncing balls around the screen, or whatever screen savers like "After Dark" are doing these days, it could flash messages about "Working on a crack of ...." and perhaps show bar graphs, etc. Maybe some flashy graphics, some Cypherpunkish slogans, etc. That is, an attractive enough screen saver module in its own right that people would be perhaps inclined to leave it running. (I know that "After Dark" publishes the specs on its program and encourages third-party drop-in modules...some have been successful enough to be marketed by the vendor. I presume this is still the case, and with Windows, too.) Acquiring chunks of keyspace remains an issue, but I think we resolved a while back that a probabalistic method works OK: people just pick chunks at random, and the decreased efficiency as compared to perfect scheduling is something like a factor of a couple (I have the numbers I calculated somewhere, and I recall Hal Finney made the same estimate). Some means of communicating results--especially wins!--is still needed. This is where Perry's idea of a Java program is a good one.
As far as cash prizes go how much could cypherpunks and friends generate for such a purpose? I'd guess individuals could come up with a fair bit of money... 1000+ list members x 10$ = 10k (or whatever).
More realistically, 1000+ list members x 10% who make plans to contribute x half of these who actually follow through x $10 = $500. (If that....) Prizes have their place, but are hard to set up properly. --Tim May Boycott "Big Brother Inside" software! We got computers, we're tapping phone lines, we know that that ain't allowed. ---------:---------:---------:---------:---------:---------:---------:---- Timothy C. May | Crypto Anarchy: encryption, digital money, tcmay@got.net 408-728-0152 | anonymous networks, digital pseudonyms, zero W.A.S.T.E.: Corralitos, CA | knowledge, reputations, information markets, Licensed Ontologist | black markets, collapse of governments. "National borders aren't even speed bumps on the information superhighway."

Timothy C. May wrote: | >As far as cash prizes go how much could cypherpunks and friends | >generate for such a purpose? I'd guess individuals could come up with | >a fair bit of money... 1000+ list members x 10$ = 10k (or whatever). | | More realistically, 1000+ list members x 10% who make plans to contribute x | half of these who actually follow through x $10 = $500. (If that....) | | Prizes have their place, but are hard to set up properly. A better way to set up a prize is to find a few big companies willing to sponsor such a demonstration. AT&T, Nortel, RSA, Netscape, Microsoft, Qualcomm, and many other companies have an interest in seeing stronger than DES crypto exportable. Perhaps one of them could set up a prize, similar to netscape's Bugs Bounty, or the RSA-129 challenge. Adam -- "It is seldom that liberty of any kind is lost all at once." -Hume

I have often thought that in spite of the raw numbers of commercial Windows platform, the freely redistributable distributed computation harnesses often end up doing much "more" of the computation. If this is the case, it would seem that some sort of Java VM system--such as that which exists in all Netscape 3.0 with JRI--would be the more natural "target" for development efforts. The recent licensing agreements from JavaSoft for JDK source make me nervous about the use of Java though. Can anybody point to a source which details Hardware/Software combinations used in distributed cracks? Maybe I can put some numbers together. -- "A screaming comes across the sky. It has happened before, but there is nothing to compare it to now."
participants (3)
-
Adam Shostack
-
Mark Evenson
-
tcmay@got.net