[jdev] The future of Jabber/XMPP?

Matthew Wild mwild1 at gmail.com
Fri Aug 27 11:03:01 CDT 2010

Hi Mathias,

On 27 August 2010 16:48, Mathias Ertl <mati at fsinf.at> wrote:
> On 08/27/2010 05:18 PM, Matthew Wild wrote:
>> Do you have a better solution that doesn't have the issues your
>> implementation has? All we want are working specifications, and that's
>> what we're aiming to develop.
>> The only cries I've heard that PEP doesn't scale seem to be coming
>> from folk involved in ejabberd. I'm not sure why that is.
> To be fair, Stephen Pendleton claimed earlier in this thread there is a
> memory leak in Openfire. Jonathan Schleifer claimed that PEP is not
> supported in jabberd1 and jabberd2, the latter still seems to be
> maintained?

A memory leak is a bug, and not an issue of protocol scalability.
jabberd14 is hardly developed nowadays, PEP is not the only feature it
is lacking. jabberd2 is relatively more developed still, but as I
understand it "scalability" is not the reason it doesn't implement PEP
(more that PEP is relatively hard to slot into the jabberd2
architecture, which I understand).

> I don't think the current tone is bringing us anywhere, in any case.

I don't know about "tone" - this is a technical issue, and deserves a
good technical argument. If at the end of the day we can improve
either the specs or implementations (or better, both) then it's
certainly worth it, and it's quite normal and healthy for such debates
to take place in a standards organisation (that said, this thread
started to be on the wrong list a while back).

I will agree that "scalability" means different things to different
people, and it is better to focus on more concrete issues, but it
seems to be the reason used here for pushing the issue downstream
(it's affecting me as a user), and I see no evidence yet that it is a
valid reason, which is why I mentioned it in my post.


More information about the JDev mailing list