[Ns-developers] Google Summer of Code 2009

[Joseph Kopena]

On Fri, Feb 27, 2009 at 10:51 AM, Jan Henning <herr.jth at googlemail.com> wrote:
> I`am very interested in participating in Googles SoC 2009, so I took a
> look at the ns-3 wiki-page and it features a number of very
> interesting project-ideas.

Hi Jan,

I'm keeping this message on-list in hopes that other potential GSOC
applicants will also see it and resolve some of their questions.

To answer some of your questions:

- Experience and knowledge requirements depend greatly on the specific
  project.  For example, working on Agent-J and providing
  infrastructure to run Java programs over ns-3 would require some
  Java familiarity.  I tend to believe that working on the NSC would
  require solid familiarity with the Linux kernel to be successful.
  Some of the other projects may just require some familiarity with
  network programming.

- As far as C++ abilities, ns-3 GSOC candidates do not necessarily
  need to have deep, expert knowledge on C++.  Again, it's somewhat
  project dependent, but I think most projects would be fine for
  someone who's comfortable with C++ but not necessarily prepared to
  dive into really deep subjects.  As an example, there's a huge
  difference between using ns-3's callback and Object mechanisms
  versus developing them; the former doesn't require nearly the same
  level of depth and trickery...

Have you used ns-3 at all?  If not, the best place to start thinking
about putting together an application is to download it, start playing
with the examples and writing some simple code, and determining:

- Are you comfortable working with the codebase?

- What aspects interest you most?

- How would you approach that, given the current architecture and code?

In this case, if you're interested in fragmentation and MTU discovery,
then the next thing to do, if you have not already, after getting some
familiarity with the code is to think about what parts of the code you
would have to work with to resolve that issue.  That conception and
plan would be the key parts of a really good GSOC application.

We'll be cleaning up the projects list a little bit over the next week
or so as we put together our organization application, though I don't
expect it to change too much.  Feel free to keep the questions coming!

Thx

-- 
- joe kopena
right here and now