[jdev] Re: The State of Our Code-bases

[Bart van Bragt]

Tijl Houtbeckers wrote:
> The next question would be: "How can we fix what is rotten?"
> It puzzles me why *this* client/server project would succeed in solving  
> those issues, except for the very last one perhaps.
This is my point exactly. There are a _lot_ of things that can and need 
to be improved, creating 'Yet Another' client/server is not one of them 
if you ask me. When I'm creating something new I need 
proper/clear/complete documentation, not a reference implementation in a 
language that I can't even use directly. If I want to know how to 
implement filetransfer then I'll read the JEPs. It would help a lot if 
there was some extra documentation to go with those JEPs with extra 
examples, some hints, tips and common pitfalls and user comments of 
other developers that have already implemented the JEPs. What I won't 
need is a reference implementation that is implementing the JEPs in a 
way that's probably not compatible with the way my application works.

Don't get me wrong, there are quite a few advantages to having some kind 
of reference server/client. Especially if we want to quickly test if a 
JEP is implementable, if we want to test compliance, etc, etc. But 
creating this client/server pair will require a _huge_ amount of effort. 
Just look at how long it took Rob to come up with jabberd2 (Kudos for 
that!) and just look at how much time has been invested in all the 
different Jabber clients. Allocating all those resources to creating 
this client/server wouldn't be my first choice :D I'd _much_ rather see 
a decent certification program, more documentation, more marketing, a 
more stable jabberd2, Avatar support, easier transport integration, much 
better web integration, etc, etc, etc. IMO those things would help 
Jabber as a whole quite a bit more.

In short: A reference client/server is a very good idea if we have the 
resources to burn but IMO we don't.

But still looking forward to your proposal, maybe I'm not getting your 
intentions correctly.

Bart