On Fri, 1 Aug 2003, Orna Agmon wrote:
> On Fri, 1 Aug 2003, Shlomi Fish wrote:
>
> > there is a limited number of key IDs and supposedly a greater number of
> > circulating keys, then there must be two keys which share the same ID.
>
> It is not SUPPOSEDLY a greater number of keys. If your key is1024 bit
> long, you will need 64 hexa signs to express it. The fingerprint only
> contains 40 hexa signs, and the key ID only contains 8. So there are more
> keys to choose from than key IDs or fingerprints.
>
Hmmm... I was talking about the keys that were already allocated. Out of
all the possible keys not all keys were yet allocated. So even though
there's a 1024-bit-space of keys, if there's only a million of them
around, it is possible none of them would have the same key ID.
Regards,
Shlomi Fish
> With central coordination, a database maybe, and probably not making it
> possible to compute the key ID and fingerprint from the key itself, it
> might be possible to assign the first 16**8 keys a different key ID, but
> since the gpg can work in a distributed manner (i.e. the keyserver just
> holds the key, it does not compute its key ID), then we must resolve to
> the natural way of doing it:
>
> Simply compute the fingerprint (and hence the ID) from the key itself.
>
> Orna.
>
----------------------------------------------------------------------
Shlomi Fish [EMAIL PROTECTED]
Home Page: http://t2.technion.ac.il/~shlomif/
There's no point in keeping an idea to yourself since there's a 10 to 1
chance that somebody already has it and will share it before you.
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
