Other Directory Sites: SeekWonder | Directory Owners Forum
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:20:11 ADT
Cicero (cicero@redneck.efga.org)
7 Jul 1998 17:54:59 -0000
anonymous wrote:
>Before you object that no hash would have this property, consider a hash
>scheme used occasionally, a concatenation of SHA-1 and MD5. Most people
>would (tentatively) consider SHA-1 stronger than MD5 (even on a per-bit
>basis), hence any truncation would be better done from the MD5 part than
>from the SHA-1 part.
Perhaps this is a example of where truncation would be a good idea.
If the requirement were for a 128 bit hash which was as strong as
possible (so that some additional computation would be permitted),
isn't it correct that you would prefer a truncated SHA-1 to a full
MD5?
Cicero
The following archive was created by hippie-mail 7.98617-22 on Fri Aug 21 1998 - 17:20:11 ADT