Re: [9fans] GSoC '14

[erik quanstrom <quanstro@quanstro.net>]

> ·        A device driver for Plan 9 for an unsupported device
> 
> ·        A GUI builder for the Plan 9 control graphics library
> 
> ·        Porting of the Tcl tool-kit Tk and the Tk GUI builder Xf
> 
> ·        VRML support for the Httpd
> 
> ·        Writing of an editor similar to Vi for Plan 9
> 
> ·        An Html front end to the PQ database system. This would involve
> adding PQ support to the Httpd
> 
> ·        Java support for Plan 9
> 
> ·        A lockable screen saver for Plan 9
> 
> ·        Mpeg support for Plan 9
> 
> ·        Addition of load balancing support into Plan 9

without addressing these particular ideas, these are the things that i've found
helpful as a mentor spotting projects likely to be finished in a summer.  

1.  the project should be of personal interest to the student.  it helps guide
decisions, it helps motivation, etc.  build something for yourself that happens
to be of interest to the community.

2.  the project needs to be fully scoped out.  some projects are conceptually
straightforward, but require mountains of code to implement.  some projects
are conceptually interesting and take less code.  projects that require lots of
code often run into trouble.

3.  planning that allow time for a false start.

4.  a stepwise plan.

- erik