[Standards] stream errors in 206

Ian Paterson ian.paterson at clientside.co.uk
Sat May 19 12:55:10 UTC 2007


Stefan Strigler wrote:
> Am Freitag, den 04.05.2007, 11:40 +0100 schrieb Ian Paterson:
>   
>>> Alex Wenckus wrote:
>>>       
>>>>> For example, only if the agreed 'ver' is 1.7 or greater then the 
>>>>> points 1 and 2 MUST be allowed.
>>>>>           
>>>> For each subsequent version of the XEP it will get progressively more
>>>> difficult to maintain server/connection manager code if particular
>>>> things are added to each and every version. It gets to the point,
>>>> depending on how the server is handling errors where you have to check
>>>> versions throughout the code. I thought it was acceptable as a one time
>>>> thing for the 1.6 release but it maybe better to accept that the newer
>>>> version will break older clients then having the server distinguish
>>>> between each version and know what it can and can't send to the client.
>>>>         
>> I tend to agree. Although I think we need feedback from as many XEP-0124 
>> developers as possible before I'm confident to proceed and recommend 
>> this change to the council.
>>     
>
> I totally agree: We should keep checks on version numbers as low as
> possible or avoid them completely. Things start getting confusing
> otherwise as the number of implementations start to increase. You always
> should take into account that implementations contain bugs. A fact that
> breaks interoperability. So keep things simple please!
>   

I've changed the CVS copy of XEP-0206 to reflect the new content of 
remote-stream-error responses.

Peter, can you please add acceptance of this to the agenda of the May 
30th council meeting.

Until then you can see the latest version here:
http://www.clientside.co.uk/xeps/xep-0206/xep-0206.html

- Ian




More information about the Standards mailing list