Am I missing something simple here? You pad pre-encryption and unless the encryption algorithm is completely flawed then slightly different input files will give completely different output.
In a similar (but entirely unrelated way!) look at the MD5/SHA256/your-fav-hash-function of the numbers 11 & 12. Do you see any similarities in the hash? Olaf On 12/07/2013 5:58 PM, "Iraklis ." <[email protected]> wrote: > But still someone can identify the common pattern at the end of two or > more similar files so the pad can be detected and deleted, thus length is > not hidden. > On 07/11/2013 09:01 PM, Tony Arcieri wrote: > > On Wed, Jul 10, 2013 at 7:31 AM, Ed Kapitein <[email protected]> wrote: > >> Hi Zooko, >> >> Be careful with zero padding, it might leave the data open for a >> watermarking attack. >> > > How about padding with the output of a PRNG seeded with the content > hash? That way it'd still be deterministic > > -- > Tony Arcieri > > > _______________________________________________ > tahoe-dev mailing > [email protected]https://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev > > > > _______________________________________________ > tahoe-dev mailing list > [email protected] > https://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev > >
_______________________________________________ tahoe-dev mailing list [email protected] https://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
