On Mon, 22 Nov 2004, Georgi Guninski wrote: > would prefer to keep my secrets encrypted with algorithm whose breaking > requires *provable* average runtime x^4242 or even x^42 instead of > *suspected runtime* 2^(x/4). (due to lameness the previous statement may be > incorrect but hope the idea is clear). afaik crypto algorithms don't exists > with provable average breaking time in suitable P.
Provable complexity is a rather scarce commodity in the area of cryptography. Yes, there are tons of proofs out there but most of them are based on *unproven* conjectures about the complexity of certain basic problems (RSA problem, discrete logarithm etc.), therefore the best thing we get is provable *relative* complexity. Most of the cryptography is black magic (I wouldn't say that if I haven't heard similar claims from true cryptologists...<g>). Of course, you can always use the Vernam cipher when you need something provably secure. :) --Pavel Kankovsky aka Peak [ Boycott Microsoft--http://www.vcnet.com/bms ] "Resistance is futile. Open your source code and prepare for assimilation." _______________________________________________ Full-Disclosure - We believe in it. Charter: http://lists.netsys.com/full-disclosure-charter.html
