[standards-jig] IBB: Making it all "go"

Dave Smith dizzyd at jabber.org
Wed Apr 9 03:11:01 UTC 2003


----- Original Message -----
>From: "Justin Karneges" <justin-jdev at affinix.com>
To: <standards-jig at jabber.org>
Sent: Tuesday, April 08, 2003 8:44 PM
Subject: Re: [standards-jig] IBB: Making it all "go"


> > 4.) Clarify the error states in the JEP, so that the behavior of the
> > stream in error conditions is well specified. In my mind, this is the
> > only outstanding point of discussion. The clarification should detail
> > what happens specifically when:
> > 4a.) The target goes offline (for any reason)
>
> What should we do about message packets that wind up going to the next
> available resource?  As DW noted, this could be a low-bandwidth resource
that
> does not want to be bombarded.

Well assuming there _were_ low-bandwidth resources actually deployed, I
would say that it is the responsibility of the server to protect said
resources from being bombarded by _any_ large message. This is a much more
fundamental problem that should not hold this JEP up. Let's not
overcomplicate things.

> I hate to wreck the party, but as much as I like these latest ideas, I
think
> we may want to go back to just plain <iq> with wait-for-result.  Sure,
this
> has the downside of extra turn-around time per packet, but at least it is
> issue-free (no counters, no delivery problems).

Don't worry, you're not wrecking the party. From what I can tell, the
community -- at least everyone who's bothered to speak up -- prefers the
<message> to the <iq>.

Is there anyone else who thinks we should use <iq> instead of <message>? If
so, I would suggest you speak up now. Otherwise, I only see one person who
dissents the use of <message> as the base packet type.

Diz




More information about the Standards mailing list