On Fri, Dec 30, 2005 at 03:55:58PM +0100, Michael Gerz wrote: > Martin Vermeer wrote: > > >Attached, finally, the corrected version of the patch using crc32. Turns > >out y isn't needed here after all. > > > > > Just wondering: Is crc32 a suitable algorithm to build a signature (aka > hash value). Why don't we use one of famous hash algorithms? > > Michael
I think it is. The requirements are 1) identical rows give always identical hashes 2) different rows give _in practice_ always different hashes. CRC32 meets these requirements, and seems to be OK on speed. In my understanding, stronger hashes are used if you have a third requirement: 3) One should not be able to "doctor" a row to produce a pre-desired hash, AKA a trapdoor algorithm. You see, I ended up researching the thing ;-) - Martin
pgpw3O1De7FBd.pgp
Description: PGP signature
