[standards-jig] re: Protocol Suites (JEP-0073)

Peter Saint-Andre stpeter at jabber.org
Wed Mar 5 17:06:05 UTC 2003


Thanks for exercising such restraint. :)

I agree completely that we cannot require conformance with informational
protocols, that's just wrong. This document is intended to be a spur to
action on cleaning things up and defining protocol levels to which we can
require conformance. We've been letting things slide for too long, and our
compliance team (not to mention Jabber implementers at large) deserve some
stable specifications. So let the discussion begin!

Peter

--
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.php

On Tue, 4 Mar 2003, David Waite wrote:

> Peter Saint-Andre wrote:
> 
> >This evening I wrote the first draft of a JEP that defines three Jabber
> >protocol suites (Core, Basic, and Advanced). I feel strongly that such
> >definitions are necessary in order to stabilize and clearly define the
> >protocol, to enable efficient compliance testing, and to make it possible
> >for the JSF to effectively market this thing we call Jabber. While this
> >JEP is informational, it has fairly significant implications for many
> >existing protocols as well as for several protocols under development, so
> >I encourage you to read the JEP and sound off on the standards-jig list if
> >you have questions or concerns. Essentially this JEP seeks to answer the
> >perennial question "What is Jabber?", so hopefully the JEP will spark a
> >fruitful conversation in the Jabber community. Join in the fun by reading
> >the following URL (remember this is a first draft) and signing up for the
> >standards-jig list if you're not already subscribed.
> >
> As an aside - how do you want to deal with the comments saying 'this 
> feature should be in this category (instead of this other category)' ? I 
> have several of those sorts of comments, but am trying to keep the 
> emails as objective as possible, at least to start. ;-)
> 
> So to start - I do not believe we should be standardizing on 
> informational JEPs without some sort of review period. In particular:
> JEP-0049: Private XML Storage
> JEP-0054: vcard-temp
> JEP-0055: jabber:iq:search
> JEP-0022: Message Events
> We should either translate these into standards track jeps and go 
> through the standard process, or create new JEPs to accomplish their 
> functionality. Some of the informational drafts put forth by the 
> standards-jig have known issues - they were proposed as informational in 
> order to document the current state of affairs, with the hope that one 
> day the functionality would be reevaluated and then standardized.
> 
> In the same vein, also should not require support for features which 
> have not been through a standardization process by either by the JSF or 
> the IETF, such as
> jabber:iq:time, jabber:iq:version, jabber:x:roster, jabber:x:delay, and 
> jabber:iq:register
> 
> And please keep in mind I've edited and re-edited this email for over an 
> hour, because I really don't want to imply conformance levels are a bad 
> thing, or make value judgements on any of these existing protocols here. 
> I just want to voice a concern over standardizing conformance on 
> nonstandardized features, specifically because the membership and 
> technical council do not get to comment on informational JEPs other than 
> comments on accuracy.
> 
> -David Waite
> 
> _______________________________________________
> Standards-JIG mailing list
> Standards-JIG at jabber.org
> http://mailman.jabber.org/listinfo/standards-jig
> 




More information about the Standards mailing list