Re: KeyID as left vs right substring of fingerprint

["L. Sassaman" <rabbi@xxxxxxxxxxx>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Fri, 23 Jun 2000, Bodo Moeller wrote:

> On Thu, Jun 22, 2000 at 01:37:49PM -0700, L. Sassaman wrote:
> > On Mon, 19 Jun 2000, Bodo Moeller wrote:
> >> L. Sassaman <rabbi@xxxxxxxxxxx>:
> 
> >>> [...]  Expanding from the right is less natural to users whose language
> >>> reads from left to right, in my opinion.
> 
> >> Have you used a telephone lately?
> 
> > That's a non sequitor if I have ever seen one...
> 
> Dialing certain postfixes of a given phone number can work, dialing
> any prefix won't.  That is, phone numbers expand from the right.
> I haven't yet heard of anyone complaing about this on the grounds that
> thear language reads from left to right.

That's an invalid analogy. Telephone prefixes exist so that phone numbers
can be grouped seperately. There is no such grouping with OpenPGP key
ideas -- all key ids are members of one group. In a left to right
most-significant-bit key id expansion, if the 32 or 64 bits you provided
weren't enough, you could keep adding more until a match was found. In a
left to right least-signification-bit expansion, you would have to start
over with the new information.

This is all academic, however, since I think we decided that changing this
now would be worse than leaving it the way it is.


__

L. Sassaman

System Administrator                |  "If you chose not to decide, 
Technology Consultant               |  you still have made a choice" 
icq.. 10735603                      |  
pgp.. finger://ns.quickie.net/rabbi |                    --Rush







-----BEGIN PGP SIGNATURE-----
Comment: OpenPGP Encrypted Email Preferred.

iD8DBQE5UwzZPYrxsgmsCmoRAiE9AKDQSEgF2hMJlE1Niwdb3oL+X1Q5VQCg/AA3
rMeJBPdynBM7rI6CrWSB7TU=
=W8Or
-----END PGP SIGNATURE-----