[Standards] FW: pubsub question related to notification events

Peter Saint-Andre stpeter at stpeter.im
Fri Aug 3 15:58:56 UTC 2012

On 8/3/12 9:44 AM, Tuomas Koski wrote:
> Hi Todd,
> On 3 August 2012 17:16, Todd Herman <todd at apx-labs.com> wrote:
>>> I have been reviewing XEP-0060 (pubsub) and have come across something that
>>> is bothering me.
>>> Throughout the document, notification events are highlighted and discussed.
>>> The events come as a Message element with a child Event element that is in
>>> the jabber.org/protocol/pubsub#event >namespace.  This is the case for
>>> nearly every event notification until you get to the very end of section 8
>>> in the document.
>>> 8.9.4 describes the notification event for an affiliation change.  However,
>>> this event notification is a Message stanza with a Pubsub child element
>>> (rather than an Event element).  I would like to >understand why it is this
>>> way.  The section above it, 8.8.4, covers subscription notifications (very
>>> similar in nature to the affiliations) and it is under an Event element.
>>> Could someone tell me if there is a specific reason for it being setup this
>>> way?  If their isn’t, I would like to submit that “affiliation” be added
>>> under the Event namespace (similar to subscription) and >section 8.9.4 be
>>> updated accordingly.
>> I am resending this message from yesterday because we really need a
>> response.  I don’t want to move away from the specification but we want to
>> code this in a manner that makes sense.  We currently use OpenFire and
>> OpenFire doesn’t event seem to support this event at all so it shouldn’t be
>> an issue with OpenFire.  We are writing our own server so we can code it to
>> behave the way we want.  Does anyone have any thoughts here?
> I think this issue has had some previous discussion already here
> http://mail.jabber.org/pipermail/pubsub/2011-August/000719.html
> (Yep, some time ago).
> I have always implemented the "affiliation -notifications" in
> http://jabber.org/protocol/pubsub#event namespace (where I think it's
> more consistent with the rest of the events).

Well, I said I would flag this as an open issue, and I suppose it still
is, because I haven't gotten around to working on revisions to XEP-0060. :(

However, I think we have agreement that this is an error in the spec.


Peter Saint-Andre

More information about the Standards mailing list