[Members] Summer of Code proposals and applications

[Kevin Smith]

On 17 May 2006, at 16:16, Étienne Lavanant wrote:
> 2006/5/17, Kevin Smith <kevin at kismith.co.uk>:
> > I think a couple of examples of projects of a larger scale would at
> > the least allow me to better understand what it is I'm disagreeing
> > with ;)
>
> What I understand from Lucas' mail is that he would like projects  
> like distributed monitoring of the Jabber Network (which he did  
> propose) :
> An other example of a project non-related to a specific client or  
> server would be a tool to test XMPP compliancy of clients or servers.
Gotcha, in that case I think those types of softwares have an  
important place but that their importance does not overshadow  
development of clients and servers (after all, there's not much point  
knowing a client or server is non-compliant (or broken) if there's no- 
one working on the project to fix it ;)). I think I'd like to see  
these tools though and if qualified students applied would be happy  
to see a couple of such projects sponsored (although the much higher  
qualification and experience required for writing a test-suite than  
adding features to clients/servers means we would be fortunate to get  
many such students).


> I agree that it would be good to have some of these projects but I  
> do not see it as a problem to have projects related to a specific  
> client or server as long as they are innovating or original and not  
> just day to day work on a software.
Well, if I abuse what you're saying somewhat to turn it around to  
making advances in clients and servers used by many people, rather  
than niche situations, I agree. I'd be sad to see someone spend the  
summer writing a new, feature-bare client in python, for example,  
where Gajim fills this role so well, but making genuine advancements  
to Gajim seems like something that would benefit the many users of  
it, whether it's new JEP support, or more run-of the mill (though  
substantial) advancement (see, I didn't even use Psi as the example).

> Does everybody else agree ?
I'm dangerously close to agreeing with you, yes :)

/K

-- 
Kevin Smith
Psi Jabber client developer/project leader (http://psi-im.org/)
Postgraduate Research Student, Computer Science, University Of Exeter