Re: Identification

[Iljitsch van Beijnum <iljitsch@muada.com>]

On Mon, 17 Mar 2003, Peter Tattam wrote:

> As long as they are unique enough within a site, I can't see a problem.  All
> you want is to not be mistaken for someone else in the quagmire of the internet
> soup.  A site is a pretty good approximation.  The GUID is only a few more bits
> of uniqueness.

What we need here is something that uniquely identifies one end of a
transport session for at least the duration of such a session (for which
there is no clear upper bound). The lower 64 bits of an IP address can't
be such an identifier because several hosts on the internet can use the
same value for the lower 64 bits of their address at the same time and
there is no way to make sure they won't try to communicate with the same
host at the same time.

We have duplicate address detection on local links for a reason...