[Standards-JIG] XEP-0136 Thoughts and Musings

Jean-Louis Seguineau jean-louis.seguineau at laposte.net
Tue Nov 7 08:18:12 UTC 2006

Chris, I agree on all counts with you on compliance. I just needed an
example to hammer out the idea that <body/> was not enough ;)

Peter also made a good point with non IM applications.

But more generally, XEP-0136 exhibits a very common pattern for an XMPP
extension: it combines management protocol (storing, retrieving, deleting,
etc...) and target content. It may be opportune to draw XEPs authors
attention to separate the two aspects in their proposals. I do not have a
precise application example in mind except what you said about compliance
needing IQ and presence logging, and when you think of it, the management
protocol of XEP-0136 could also apply to them. It is somewhat independent of
the content, isn't it?

That said, we are left to decide how we would include the complete message
content in the appropriate wrappers inside XEP-0136 stanzas. Do we leave it
unchanged; do we encode it or zip it?


-----Original Message-----
Message: 5
Date: Mon, 6 Nov 2006 14:35:08 -0800
From: "Chris Mullins" <chris.mullins at coversant.net>
Subject: RE: [Standards-JIG] XEP-0136 Thoughts and Musings
To: "Jabber protocol discussion list" <standards-jig at jabber.org>
	<97B71C0C860DEC40A993AB9F7F0D4335702A63 at fattire.winfessor.com>
Content-Type: text/plain;	charset="us-ascii"

JLS Wrote:
[Storing the whole message]

> Do we really want to see proprietary 
> extensions flourish to compensate
> easily corrected shortcomings?

I'm not sure if this would comply even if we did store everything. 

I know in the customer cases we often see (and I assume other commercial
entities see), logging messages isn't enough. We need to log...
everything. This mechanism doesn't log IQ or Presence packets, so I keep
thinking that a compliance mechanism on the back end will still be

I completely agree that all of the various types of bodies, as well as
the languages, need to be stored. Logging only "<body/>" elements is not
at all sufficient.  

Chris Mullins

More information about the Standards mailing list