Other Directory Sites: SeekWonder | Directory Owners Forum
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:15:20
lcs Mixmaster Remailer (mix@anon.lcs.mit.edu)
7 Oct 1998 00:20:02 -0000
The point is not whether googolplex is a "scientific number", or whether
there are other such numbers greater than 2^4096. The point is that
4096 bits is a very poor choice for the key size of a symmetric cipher.
(The comparisons by Jaws management between 4096 and the widely use 128
bit key size implies that they are referring to symmetric cipher sizes.)
There is no reason to choose key sizes anywhere near 4096 bits. Contrary
to the statement in the article, 128 bit key sizes have not been broken
and are nowhere near being broken. Such keys are 2^72 times harder to
break than 56 bit DES keys, which have only been brute forced within the
last couple of years. The computer technology necessary to search a
128 bit keyspace would be so advanced that it would remake the world.
The 256 bit keys required for the future AES standard are so far beyond
128 bits that it is almost inconceivable that they could ever be broken
in the entire history of the universe. Maybe if you could create a
perfect quantum computer the size of the galaxy you might have a problem,
but then moving to 512 bit keys would make you safe even from that.
In short, fundamental limits on the physics of computation make key
sizes of 4096 bits completely inappropriate. No reasonable evaluation
of attacks on the cipher would justify such a choice of key sizes.
It's like putting a 10 inch thick hardened steel door on a tarpaper shack,
although even that doesn't begin to capture the degree of mismatch.
The fact that Jaws uses such a key size, and is now exploring the use of
16,384 bit keys, implies that there is something fundamentally wrong with
their approach to cryptography.
The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:15:20