Re: [Geopriv] What to do about DHCP and the ever changing PIDF aswell as HELD features

From: Hardie, Ted ^lt;hardie@qualcomm.com>
Date: Thu Jul 17 2008 - 02:06:37 EDT

>Hi Ted,

>Just so I am clear then, do you see the following two URIs as being the
>same?

>https://lis.examples.com:9127/dxjediodkjcndjkcvnjkdnjkd
>and
>https://lis.examples.com:9127/3278x893ujz0o4ekdi4eikxd

RFC 3986/STD 66, section 6.2 has a whole lot of text
on the "ladder of comparison". Which type of comparison
are you discussing? For almost all of them, the answer
will be no, but that doesn't seem to really go to point you raised in
your original mail:

"More importantly, when something asks for a reference, that
reference should only be applicable to the thing actually asking for it."

Whether those two URIs *refer to the same location* is
unrelated to the question of whether they string match
and it is equally unrelated to the question of whether they would be
handed to different requestors.

You appear to me to be asking for a property (each requestor
gets a unique URI) which doesn't actually relate to whether the
requestor is at that location. You also appear to be relating one
of the above properties to some flaw you see in DHCP carrying URIs,
but remains really hard to work out where you see the
problem. It is entirely possible with any protocol slot
capable of carrying URIs to do exactly what you want.
It may be recommended for some deployments; it may not
be necessary for others, and those might use URIs which did
not carry uniqueness properties.

It's late here, so I'll likely follow up further replies tomorrow,

regards,

Ted

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www.ietf.org/mailman/listinfo/geopriv
Received on Wed, 16 Jul 2008 23:06:37 -0700

This archive was generated by hypermail 2.1.8 : Thu Jul 17 2008 - 02:06:47 EDT