[standards-jig] JEPs and Jabber Adoption

Tijl Houtbeckers thoutbeckers at splendo.com
Sat Jun 28 23:11:47 UTC 2003


Bart van Bragt <jabber at vanbragt.com> wrote on 29-6-2003 1:07:54:
>
>Tijl Houtbeckers wrote:
>> You normally don't do that kind of testing with your entire userbase.
>>  There's the well knowns process of ALPHAs and BETAs that a lot of 
>> opensource  (and closedsource too) projects use for these things. 
>> Most jabberclients do too, as far as I know. 
>This means that your development cycle will have to coincide with the 
>development of a number of (experimental) JEPs :\ So you'll just 
>retract a feature that has been through beta testing because it still 
>has an 'experimental' status? I you do; that's great :D But I'm afraid 
>that's not how it works in most case :? Could be wrong ofcourse.

Unless it's stable, don't put it in your stable releases. That 
shouldn't be to hard. You could also brach your client for testing such 
a feature, or use a plugin structure if your clients has one. Ofcourse 
some clients/projects are more suited for testing/prototyping than 
others. 

>All I wanted to say is that IMO you have to be very careful with 
>testing implementations in this way. As soon as a feature is 
>implemented in a public version or Trillian or in a version of Psi 
>(and/or other 'big' clients) then we're just stuck with that JEP for 
>years to come :D 

So the JEP's technically ok, it makes it possible to implement the idea 
that was at the base of the JEP. It does the job well enough for client 
makers to want to implement it, and it's gone through some testing. I'd 
love to be stuck with that :) 

-- 
Tijl Houtbeckers
Software Engineer @ Splendo
The Netherlands




More information about the Standards mailing list