[standards-jig] Consumer Jabber Service (was Re: [JDEV] Poll about who will use Jabber when Trillian comes out)

Iain Shigeoka iain at jivesoftware.com
Wed Jun 11 18:18:47 UTC 2003


On Wednesday, Jun 11, 2003, at 09:50 US/Pacific, Bart van Bragt wrote:

>>  I guess the question keeps boiling down to, is this
>> a major JSF/jabber.org problem?  Are we supposed to foster the end 
>> user
>> community and develop the protocol/system?  I think that's just the 
>> tip
>> of the iceberg for my thoughts, so if you want some more, just ask.
> I think it is.. As matt matt put it; we're going to have to 'push' 
> Jabber with a bottom-up approach.
>
> Companies like MS/IBM are able to push SIP/SIMPLE/MSN/Whatever through 
> everyone's throaths, inclusing network managers and managers. Jabber 
> won't be able to do that, not without major backing from MS I'm > afraid.
>
> So IMO; yes, there needs to be a good starting point for end-users. Of 
> course it's great if this could be a complete site like icq.com (well, 
> icq.com is horrible but you get the idea ;) but IMO it's already a 
> good start if there is something like a quickstart guide. Another 
> point are the clients. Most of them are created by hackers for 
> hackers. They do great stuff have all the latest and greatest but they 
> almost all lack end-user documentation and quite a few us a lot of 
> techno-babble and counter-intuitive (but efficient) UI designs.

I completely agree. I don't want to bring the whole name change issue 
up again but this is part of what I have been discussing with people 
during the development of the name change proposal. As I see it, the 
Jabber community is bigger than what the JSF is doing. Ideally, we need 
three separate pillars to support the Jabber community. they need to be 
separate because they will have diametrically opposed charters.

1) jabber.net - mainstream IM service (a.k.a. Jabber Open IM Service, 
a.k.a. 'Jabber'). This is the AIM killer. It is run like a consumer IM 
business presenting a slick, easy to use website portal into one Jabber 
service. It's sole charter is to gain maximum marketshare (e.g. Jabber 
users) expanding the Jabber network. It's inherently unfair to the 
general community because:

   a) It competes with other Jabber IM service providers.
   b) It will choose and run one 'best' server (be that jabberd, or a 
commercial one if needed).
   c) It will choose, and push users to one 'best' jabber client. one of 
the most overwhelming things with new users and Jabber is there's too 
many choices when it comes to clients. There should just be a 'if 
you're on windows, download this' one-click process.

However, we need this service to show that jabber is prime time, and a 
real contender. It also will help to gain market share for jabber users 
which is critical to legitimizing Jabber as a mainstream IM, and also 
provides more value to everyone on the jabber network. Hopefully this 
is something we can fund through at least in part ads but I suspect 
we'll need to contribute money to it from the Jabber community to run 
it.

2) jabber.org - open source jabber (a.k.a. Jabber Software Foundation). 
Essentially jabber studio, plus all the things that are useful for open 
source developers such as the technical documentation, mailing lists, 
jdev chatrooms, etc. that are currently on jabber.org. I'm not sure if 
this should be more like the apache model (managed) or the sourceforge 
model (free for all) but the essential goal is to encourage and support 
open source jabber development. Once again, this should be at least 
partly supported by advertising. However, this is where a lot more 
sponsor dollars should go. the primary goal is to create and maintain 
open source Jabber software.

3) xmpp.org - XMPP standards and compliance (a.k.a. ?? XMPP Group, XMPP 
Standards Foundation, Open IM Group, etc). This is what the current JSF 
is doing Re: JEPs, JIGs, etc. This organization needs to be fair to 
commercial and open source projects, and serve as a neutral place for 
standards to be developed and compliance to be tested.

-iain




More information about the Standards mailing list