[standards-jig] JEPs and Jabber Adoption

Bart van Bragt jabber at vanbragt.com
Sat Jun 28 22:30:47 UTC 2003


Tijl Houtbeckers wrote:
> All we do now is compete on theoretical assumptions (like you said, 
> experimental JEPs are not at all attractive to implement, let alone 
> ship to your clients) for many many hours, and almost never on 
> implementation and large scale testing of that implementation.

Please keep in mind that you're messing with your userbase if you use 
'large scale testing'. IMO it's _far_ worse to have a feature that 
changes approx every two months (and which only works with half of the 
other clients) than to simply lack the feature. Users can live without 
avatars, they get annoyed when the client says that they have avatar 
support and they can't get it to work. Two years ago you could just 
implement something, test it out and then decide that that JEP sucked, 
IMO you can't do that anymore. Once a JEP gets implemented in a few of 
the larger Jabber clients you can't just depreciate or withdraw it 
without creating large amounts of problems.

IMO a bit more attention should be spend in defining what a JEP should 
really do, a nice example is the recent JEP-70 vs JEP-101 (HTTP 
authentification) discussion. Different views on the purpose of a JEP 
give a LOT of problems that are pretty hard to resolve, simply because 
the involved parties have no idea what the others are talking about.

Furthermore it would be nice if there was some kind of time limit on 
discussions etc but I guess those limits are a bit hard to enforce ;)

Bart





More information about the Standards mailing list