[jdev] jabber.org not conforming to JEP standards

Chris Chen ckchris at idream.net
Sun Aug 7 18:06:25 CDT 2005


Yes, I understand that JEP schemas are informative.  And of course  
nothing in the JEP forbids adding more fields and I am in no way  
stating otherwise, but am merely musing that the whole point to these  
"Jabber Enhancement Protocols" is to extend the XMPP protocol.  There  
would be no end and no point to these JEPs if everyone enhances the  
enhancements.  It would be nice if either another JEP replaces the  
old one OR the current JEP indicates that additional fields can be  
added.

I am a developer  who would like to conform to the specs so that my  
software can communicate with as many different servers as possible.   
And the only way to do that is to follow the JEPs.  After all, these  
JEPs are protocols, and according to the dictionary, protocols are a  
set of rules governing the exchange or transmission of data between  
devices (well, one of the definitions).

My point is that it's not the semantics that I am concerned about.   
It's the communication.  If two devices cannot communicate based on  
one protocol, then there is no point in having such a protocol  
because the protocol is not doing what it should be doing -- bridging  
communication gaps.

Chris

On Aug 7, 2005, at 2:50 PM, Tomasz Sterna wrote:

> 2005/8/7, Chris Chen <ckchris at idream.net>:
>
>> According to JEP-0078's jabber:iq:auth schema, there should only be 4
>> child elements to the query -- username, password, digest, and
>> resource, in that order.
>>
>
> Please notice, that all JEP schemas are informative, not normative.
>
> JEP-0078 3.1 states that the only REQUIRED fields are username and  
> the resource.
> Nothing in the JEP forbids adding more fields there. It would work
> against the eXtensibility of the protocol Enhancement.
>
> -- 
> smk
> _______________________________________________
> jdev mailing list
> jdev at jabber.org
> http://mail.jabber.org/mailman/listinfo/jdev
>




More information about the JDev mailing list