GSOC proposal advice

[Toby]

If you are applying for a Google Summer of Code grant for this (or
other) projects, here is some general advice:

Applications are reviewed by essentially all of the Python Software
Foundation mentors, and they are generally professional programmers
who want to see a professional-quality application, as if you were
submitting a resume for a job. Plus it is a competition: there are not
enough grants for all applicants, and so you will get ranked in
comparison to everyone else who submitted and application for the
Python Software Foundation.

Thus:

- Ensure your application has *perfect* spelling and grammar. If two
otherwise equal applications are being compared, this could make the
difference.

- List previous coding experience, especially if it is related to the
project. Links to actual code or projects you have worked on are
ideal.

- Show that you are aware of relevant tools and software packages
(e.g. if you say you are going to implement a GUI, say GUI tools you
think would be appropriate).

- Be realistic: don't promise too much or too little. This is often a
judgment call, but the important thing is to think about what your
proposing and present it in as practical a way as you can. Remember,
you are being judged by mainly by the standards of professional
programmers.

- Make some estimate of the time you expect to spend working on the
various tasks. Again, be realistic (hint: if you are new to making
such estimates, a good rule of thumb is to --- at least --- double
your final estimates!).

- Include learning, researching, coding, documenting, and testing in
your time estimates.

- Yet at the same time, don't be unrealistically precise in your time
estimates --- show us that you are aware of the inherent uncertainty
in predicting how long it takes to do implement features on a project
that is new to you. Blocks of 5 hrs is perhaps a reasonably
granularity --- although different programmers have different
opinions, skills, and experiences that could can affect whether or not
they feel these estimates are realistic.

- Details are important in this application, although not too many
details. Don't just say that you want to implement feature X, but say
*how* you plan to go about, and show that you have thought about some
of the issues that might arise. There should be enough detail such
that a professional programmers (such as the ones reviewing your
application) can understand what you are doing, and can see that it is
a workable approach. It is hard to say in general how much detail is
necessary, but, of course, the more detailed or complicated the
feature, the more detail you should provide. In general, it is better
to err on the side of *too many* details.

- If mentors give you advice in the Google application system, then
take it seriously and think about what they are saying. Sometimes
students write a quick, argumentative reply that shows they are not
really thinking about the comment; this looks very bad.

- Include your time zone, and when you are free to meet (on-line).

Toby