Adam Back wrote:
From what I recall from reading the CR paper a while back they can
tolerate up to some threshold of colluding players. However if you go
over that threshold (and it's not too large) you can remove the mark.
I would think the simplest canonical counter-attack would be to make a
p2p app that compares diffs in the binary output (efficiently rsync
style) accumulates enough bits to strip the disk watermark, p2p rips
and publishes. QED.
If the p2p apps could collude, they could create
a pre-threshold image and share it amongst
themselves only, gradually combining it until
no more differences were detected. When a
post threshold watermark was reached, the
final image could be released. You would need
some way to know that the watermark had
been reached, according to the testing against
a sufficient sized pool or somesuch metric.
Add some reputation nyms to sign and that
should avoid the poisoning attacks as well.
--
News and views on what matters in finance+crypto:
http://financialcryptography.com/
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
