[Standards] RFC 6120 vs. XEP

Dave Cridland dave at cridland.net
Wed Feb 8 21:23:08 UTC 2017


On 8 February 2017 at 08:53, Evgeny Khramtsov <xramtsov at gmail.com> wrote:
> Wed, 8 Feb 2017 08:19:17 +0000
> Dave Cridland <dave at cridland.net> wrote:
>
>> Right, I understand, and largely agree. I might scribble a draft to
>> address this, by clarifying what we really meant here.
>
> I see also two issues here ;)
>

Yup, I understand what you're saying, and I agree that's an entirely
reasonable interpretation of the document. It's also clearly against
the spirit of our understanding, otherwise, as you say, XEP-0198 would
have also run into this problem. It's also not what we want.

Given that, I'm suggesting this is a technical errata in RFC 6120.

> 1. RFC6120, section 7.1 says:
>
>> After a client authenticates with a server, it MUST bind a specific
>> resource to the stream so that the server can properly address the
>> client.
>
> Thus, a client is unable to resume a session in any case.
>
> 2. While almost everybody here argued that "resource binding" is any
> binding mechanism, including Bind2, RFC6120 clearly defines "resource
> binding":
>
> Section 7.3.1:
>
>> The parties to a stream MUST consider resource binding as mandatory-
>> to-negotiate.
>
> And section 7.1 defines:
>
>> The XML namespace name for the resource binding extension is
>> 'urn:ietf:params:xml:ns:xmpp-bind'.
>
> In my book, "resource binding" is exactly something within
> 'urn:ietf:params:xml:ns:xmpp-bind' namespace, unambiguously.
> _______________________________________________
> Standards mailing list
> Info: https://mail.jabber.org/mailman/listinfo/standards
> Unsubscribe: Standards-unsubscribe at xmpp.org
> _______________________________________________


More information about the Standards mailing list