[Foundation] JIGs and JEEPs

[Iain Shigeoka]

At 05:43 PM 5/24/2001 -0600, you wrote:

>>Also, other than the server2server and client2server protocols, the 
>>foundation doesn't dictate a particular architecture. Sure, both existing 
>>implementations I've seen  have a very similar architecture, and I can 
>>seem some informational notes pushed through to help compatibility for 
>>external components between implementations. However, as long as I 
>>support the 'standard' c2s and s2s protocols, an alternate implementation 
>>should be able created with any architecture desired, even if it means 
>>some of the existing code written for components will not work with it.
>
>True. Right now, however, we have identified a need for some 
>standards/guidelines regarding clients, servers, and components (if for no 
>reason other than our standards-conformance activities), so I would see 
>those JIGs working on that. Perhaps we need to emulate Python and make 
>sure that every JIG has a start-date and an end-date, because perhaps 
>these architecture-specific JIGs will wither away like the Marxist state.

It's starting to sound to me like too much administration overhead.  As I 
understand it (and what I hope for) is that the Foundation is here to help 
in well focused "support" functionality and standards 
creation/maintenance/testing/enforcement.  As such, it seems to me we 
really only need to worry about 3 things right now:

Jabber standards (standards & protocols, performance metering, and 
compliance testing)

Jabber PR (press releases, press relations, promotion, branding, and "legal 
enforcement" of standards through compliance testing and logo/branding)

Jabber Biz (legal, helping businesses (other than Jabber.com) use/exploit 
Jabber, creating partnerships,  negotiating on behalf of the Jabber 
community, $$$ to do fund raising, paying for conferences, scholarships)

>>I don't see why there would need to be platform/language forums within 
>>the Jabber Foundation. I do see these being useful on Jabelin.
>
>Well, Jabelin is just C. That's why I'm struggling to find the right place 
>for these community forums.
>
>>Does the Jabber Foundation need development tools or CVS? :-)
>
>Yes, if it provides organizational assistance to the projects within the 
>Jabber community, which is part of the purpose of the Jabber Foundation as 
>I understand it. Plus it might be good to have those JEEPs under version 
>control. :)

I'm a bit unsure we need this.  Why not just defer implementation support 
to someone like SourceForge.  I mean why recreate the wheel?  Isn't 
everyone's time and effort better spent elsewhere?  I am sure if we can 
organize some critical mass, we can have SourceForge create a JabberForge 
focus group (they have some name for it) like they have for other large 
groups like Java, Linux, and the like.

I don't see why the Foundation needs to provide services to particular 
implementations.  What need is this serving beyond replicating what 
SourceForge already provides?

>>>3. docs-jig (Documentation) -- kinda speaks for itself. Lots of work to 
>>>be done here. :) May develop JEEPs related to documentation standards, 
>>>and will definitely work with the other JEEP-creating JIGs.
>>
>>What would the goal for this JIG be? To document incomplete/ poorly 
>>understood ("legacy") protocol, to exist as a resource for all other JIGs 
>>in maintaining documentation, maintain all the documents produced by the 
>>foundation, and/or to propose the rules on how documentation should be 
>>written and what is required for a formalized proposal and final specification?
>
>That all sounds good. :P

:)  Yeah.  I'm hoping this is priority #1 for the Foundation.

>>>6. client-jig (Client Development) -- creates standards and guidelines
>
>>>7. server-jig (Server Development) -- creates standards and guidelines
>
>>Which of these last two jigs would define a specification as a whole?
>
>That's a good question. I think we need perhaps something more general. 
>So, as hinted above, the client-jig and server-jig might develop these 
>conformance standards, but perhaps we need an over-arching protocol-jig.

Yes.  We need standards and tests for standards.  But I'm not sure why we 
need a foundation to help people write clients.  What benefits does the 
Foundation provide for an OSS jabber client writer beyond say a sourceforge 
project?  Are we trying to standardize how people write clients?  I was 
under the impression Jabber shouldn't care how the client is written as 
long as it meets the standards (and can therefore interoperate).

I see things like server and client implementations being more of a user 
group/community type thing.  Jabelin may form a separate Jabber community 
group that concentrates on server, server transport standards, component 
standards, etc.  But that should be separate from the Foundation which is 
supporting Jabber... and in the final analysis, Jabber is the standard, 
marketing, and the "biz" around it.  Jabelin is not Jabber.  WinJab is not 
Jabber.  The foundation should try and monitor, link to, and provide 
information on particular implementations, but the details surrounding them 
seems to me to be outside of the Foundation's "jurisdiction".

I see the "Jabber Foundation" as dealing strictly with the Jabber standard 
(and compliance testing) and the surrounding Jabber issues 
(logo/trademarks/marketing).  In my mind, this is akin to forming an "HTTP 
Foundation".  If we extend the meaning of Jabber to include other IM or JAM 
type applications on/around the current Jabber core and still call it 
"Jabber Foundation" then it becomes like the "W3C": standards and 
"implementation" free.  You define standards and compliance tests, and 
perhaps provide some reference implementations and focus on education, 
community, building consensus and dealing with businesses on behalf of the 
standard (like taking 800 lb gorillas to task for "extend and enhancing" 
standards into proprietary systems).

IMHO, if we include implementation specific concerns it would be more 
appropriate for us to form another "Jabelin Foundation" to mimic the 
"Apache Foundation" both of which focus on a particular implementation to 
support some set of standards.

I personally have great interest in the "Jabber Foundation"; no interest at 
all in the "Jabelin Foundation", and no desire to mix the two.

-iain

ps - I just realized how smiley-free this email was.  So here are a few :) 
:) :)...
I'm enjoying this conversation and don't want to come across as 
angry/confrontational/difficult.  :)  Just want to make my point with 
relative clarity... Yeah, I know, I try but it still comes out like mud.


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com