Re: [boost] Summer of Code 2007

From: Jeff Garland (jeff_at_[hidden])
Date: 2007-03-03 11:54:53

Gail Banaszkiewicz wrote:
> Jeff wrote:
>> You have -- we should be making a more official announcement shortly, but 
>> since you already have a project idea I would suggest you go ahead and pursue 
>> it.  If it's anything like last year the application process is highly 
>> competitive -- the best applications really went above and beyond.  I think 
>> some of those might be on the web somewhere.
> Thanks for the encouragement! I read over the report on last year's 
> edition of SoC and indeed it looked very competitive.  I did not see any 
> of the proposals themselves, but if they are online, I will definitely 
> have to take a look,

I believe you might have missed the post from one of last years students: 
Giovanni Deretta.  You can find some (some aren't complete) of last years 
proposals here on the Google site:

http://code.google.com/soc/boost/about.html

But truth be told, the best proposals escaped from the confines of the Google 
tool and were setup on there own website.  For example:

Giovanni was able to impress with this work done prior to SoC:

http://libstream.sourceforge.net/index_c.html

And Matt Calabrese had these -- again as part of the initial proposal:

http://2006.planet-soc.com/node/201
http://members.gamedev.net/Rivorus/surge/html/index.html

And Matias Capeletto with this entry which he didn't start until after the SoC 
announcement.

http://h1.ripway.com/mcape/boost/libs/misc/index.html

As you can see, these folks had some fairly advanced ideas about what they 
wanted to attack.

Some of the student proposals used 'quickbook' -- a boost derived Wiki-syntax 
like C++ library documentation tool that feeds BoostBook/DocBook:

http://www.boost-consulting.com/boost/tools/quickbook/doc/html/index.html

I'll warn you that quickbook can be a bit tricky to get setup -- actually I 
believe a couple of last year's students helped improved the getting started 
docs.  Anyway, students that went this extra mile during the proposal showed 
that they were capable of independently handling basic environmental setups 
needed for boost.

I"ll also say that at least one of the projects from last year was 
significantly revised during the evaluation process.  The consensus of the 
mentors was that the student was very capable, but that the project was just 
too ambitious.  John Maddock made some suggestions of a similar direction and 
the student rewrote the entire proposal in a day -- it was a rather impressive 
demonstration of skill and he wound up with a Boost SoC project.

>> There project has not been pursued yet, so it would likely be a good choice 
>> for this year.  Our biggest problem might be finding a mentor, but we'll work 
>> something out.
> I will keep my options open in case this topic does not work out; 
> however, I'm sure I'd be able to also draw from the knowledge of the 
> group at school   as well as from a mentor on your end, which certainly
> can't hurt.

Yes, that's helpful.

Jeff