[xmppwg] FW: Re: ANONYMOUS vs. UNIQUE in XMPP

Peter Saint-Andre stpeter at jabber.org
Wed Mar 2 10:37:17 CST 2005


FYI, from the ietf-sasl list...

----- Forwarded message from Peter Saint-Andre <stpeter at jabber.org> -----

From: Peter Saint-Andre <stpeter at jabber.org>
To: "Kurt D. Zeilenga" <Kurt at OpenLDAP.org>
Cc: ietf-sasl at imc.org
Subject: Re: ANONYMOUS vs. UNIQUE in XMPP


On Tue, Mar 01, 2005 at 11:24:02PM -0800, Kurt D. Zeilenga wrote:
> At 07:08 PM 3/1/2005, Peter Saint-Andre wrote:

<snip/>

> >The question arises: if a server assigns unique JIDs as a result of such
> >a protocol flow, is that overloading the SASL ANONYMOUS mechanism? It has
> >been suggested that it might be necessary to define a new SASL mechanism
> >(perhaps named "UNIQUE"), but as far as I can see the protocol flow for
> >UNIQUE would be indistinguishable from ANONYMOUS, with the only difference
> >being how the XMPP server handles the resource binding process. However,
> >we are open to defining a new SASL mechanism if that is deemed necessary.
> 
> As shown above, I think not, because the assignment of the JID is
> not loaded onto the ANONYMOUS exchange, e.g., its not done as an
> integral part of step 1.

OK, so my understanding is that we will continue to use ANONYMOUS in
XMPP (with some implementations / deployments creating unique JIDs as
a result of authenticating via ANONYMOUS), unless of course there are 
further objections.

Thanks for the review.

Peter


----- End forwarded message -----



More information about the xmppwg mailing list