Re: [fab] Google's Summer of Code 2006

[seth vidal <skvidal linux duke edu>]

On Sun, 2006-04-16 at 12:00 -0400, Jesse Keating wrote:
> On Sat, 2006-04-15 at 18:11 -0500, Patrick W. Barnes wrote:
> > Google has just announced that they will sponsor the event again this
> > year.  
> > Will we participate again? 
> 
> The question is how are we doing now with the projects we SoC'd last
> year?  Where are the developers we paid to do work?  How
> maintainable/sustainable is the code and the projects?  One of the
> problems I heard expressed from last year was that once the money dried
> up the developers vanished, leaving what they had done at whatever state
> it was at.  While this might be acceptable for improvements to existing
> projects, it wasn't very good for brand new projects.
> 
> I think we seriously need to look at how to use these folks in the best
> way possible, and I honestly thing starting a bunch of projects from
> scratch is a bad use.  Have them work on improving some project with
> code that can be maintainable after they leave.  Basically approach it
> with the idea that as soon as the summer is over, the developer will be
> vanished, and we have to take what they've done and continue it somehow.
> 

I agree.

I can think of a number of enhancements to moinmoin or applications in
turbo gears (what Elliot has been doing a fair bit of coding in) that
would be useful for fedora and wouldn't take a herculean effort to
maintain.

Targeting web apps might be a better way.

Hey, I know - what if we had someone write us a mirror database system
since it seems like bouncer is just about never going to be in the state
we need it in. :)

or maybe just see if we can SoC someone into fixing bouncer as we need
it.

-sv