[tahoe-dev] Google Summer of Code 2010 -- Ideas Needed!

[Zooko Wilcox-O'Hearn]

Folks:

What could one smart student do in one summer, with the mentorship of  
an experienced Tahoe-LAFS hacker?

We need ideas which are valuable and useful but not a core feature  
which would harm the whole Tahoe-LAFS project if it didn't get  
finished on time. We need to think of things that are not too hard to  
finish in one summer but hard enough to take up a whole summer. :-)  
It helps if the project can be specified in two or three consecutive  
parts so that you get "bonus points" by doing more than the initial  
project.

Most importantly the project should be SWEET and OSSM. :-)

Our Road Map (http://allmydata.org/trac/tahoe-lafs/roadmap )  
indicates that we'll already have released Tahoe-LAFS v1.7.0 before  
the Google Summer of Code students start, so the projects we suggest  
could be from the v1.8 era or from the v2 era.

I think it is nice for the project to be a new feature rather than  
optimization, refactoring, documentation, etc., so that then the  
student can point to that feature afterward and say "I added that!".

Things that are not part of the Tahoe-LAFS core but are auxiliary  
features would be fine, too, such as a new front-end written in Caja,  
pretty visualizations, integrating Tahoe-LAFS with (another)  
decentralized revision control system such as darcs or git, etc..

Here is an example of the ideas generated by a successful Google  
Summer of Code mentoring organization:

http://www.netbsd.org/contrib/soc-projects.html

Please post your ideas to this thread! Do it right away, because we  
want to show the GSoC organizers our ideas in order to convince them  
that they should sponsor Tahoe-LAFS this year.

Regards,

Zooko