[jdev] Jabber Certification Program

Rachel Blackman rcb at ceruleanstudios.com
Thu Jun 17 17:53:24 CDT 2004

> Clients are "behind" because nearly all of them are hobby projects.  I 
> think I
> can speak for most of us in saying that we are working as fast as we 
> can
> within our available time, and throwing a certification weight on our
> shoulders wouldn't speed anything up.

If you write it as a hobby, though, do you necessarily need the banner 
on your page?  To put it another way, not every webpage author is going 
to strive for absolute W3C HTML4 compliance, but if you achieve it and 
get it verified by one of those test programs, you can put the button 
up for bragging rights.  As a result, there are web authors out there 
who feel driven to achieve that.  Does it mean that a W3C compliance 
badge or similar associated programs have a negative impact on the rest 
of the world?

> In addition, many of the features you mention just plain aren't ready 
> in
> specification form.  Avatars?  XHTML-IM?  Voice chat?  These are all
> Experimental.  Maybe we should start certifying Jabber Council 
> members, to
> motivate them to approve some JEPs?  (and speaking of which, my 
> "jep-secure"
> has beat the record of the longest time from submission to 
> publication, and
> is still counting.)

This is a frustration I can understand, completely.  However, keep in 
mind that JEPs tend to move forward out of 'experimental' faster when 
people are actually working on using them.  XHTML has languished, in 
part, because few clients have added it.

Worse still, it seems as though there is a stigma of some kind involved 
in implementing experimental JEPs.  Your own comment demonstrates this; 
you speak of the Experimental status as reason not to implement them 
yet.  I think this hurts Jabber as a whole.  Experimental JEPs /need/ 
to be implemented, so they can be tried out in the real world; real 
world user experience (benefits /and/ horror stories) do a lot to move 
a JEP forward.

Yes, they might change, which is irritating... but we DO have xmlns as 
our markers, and if for some unavoidable reason a JEP changes enough 
during the experimental phase that it becomes incompatible with its 
previous implementation, we can theoretically change the xmlns to avoid 
breaking lingering experimental versions.

And sure, you can't make an Experimental JEP a 'required' for the 
certification set for one year, but you can definitely make it a 
'recommended,' and encourage implementation of it.  The clients who 
want the nifty little certification badge -- and I think you'd find 
they are out there -- will work to implement these things, and thus 
help move the JEPs forward.  If we had 'avatars' as a recommended for a 
certification list, I bet you there'd have been at least a Draft 
version avatar JEP by now. ;)

There's my $0.02, for whatever it's worth.


More information about the JDev mailing list