On Sun, May 06, 2001 at 05:01:55PM -0500, Scott Gregory Miller wrote: > > > > It turns out we aren't though. With the current mode, the IV cannot be > > retrieved (since you need to previous block to decrypt anything). What we > > are in fact doing is running an all zero IV. Since being able to restore > > the key to check it is not that important anyways, I think we should move > > to actually using the key as the IV (or rather the key as a single block > > encrypted by itself). > > Your right, but the reason we are able to do this is because we are > assured that roughly the first block of data was random, *because* it > contained the key. We can use the key as the IV as well (or the hash of > the key, which would make me feel better).
So at the moment, the data to be encrypted always begins with a 2-byte length field followed by the bytes of the crypto key, and we leave the IV initialized to all zeroes. Is there any actual advantage in initializing the IV to some hash function of the crypto key? -- # tavin cole # # "The process of scientific discovery is, in effect, # a continual flight from wonder." # - Albert Einstein _______________________________________________ Devl mailing list Devl at freenetproject.org http://lists.freenetproject.org/mailman/listinfo/devl
