On Wed, 11 Oct 2000, Chris Lonvick wrote:
> >> - Is this going to be limited to md5 or are other hash algorithms going
> >> to be accepted?
> >
> >Again, that's up to the implementor. This is just an indication of a
> >general method.
>
> I think that there needs to be at least one defined algorithm along with
> its identifier. Alternatively, you could propose defined fields where the
> value is the MD5 hash exclusively. Otherwise, each implementor will
> define their own hash and we won't have any interoperability.
Agreed. If we know exactly the set of keys we are going to be hashing,
it's possible to design a minimal perfect hash function.
Minimal perfect hashing guarantees that n keys will map to 0..n-1 with
no collitions at all.
(In my opinion interoperability is preferable)
ciao
alfonso
--
Alfonso De Gregorio, [EMAIL PROTECTED]
