[GSoC 2015] Introduction

Dirk Hohndel dirk at hohndel.org
Sun Mar 15 16:00:33 PDT 2015


On Sun, Mar 15, 2015 at 11:36:30PM +0200, Claudiu Olteanu wrote:
> 
> I believe that it is time to introduce myself and to let you know that
> I intend to apply for the idea "Native Bluetooth support for Linux
> and/or Windows" - GSoC 2015.

Excellent! Certainly one of the projects that I was hoping we'd find a
talented student for...

(and before any of the other students start to worry... I'd say that for
pretty much every single one of the projects... that's why they are on the
list :-)  )

> My name is Claudiu Olteanu and I am a bachelor student at the University
> Politechnica of Bucharest. I have a strong background in C language and
> some bluetooth experience from GSoC 2013 edition where I implemented a
> bluetooth transport plugin for GNUnet organization.

That does sound like you would be a good fit for this project!

> Until today I managed to build the sources on a Linux platform, to play
> a little with the interface and to watch the tutorial. Unfortunately I'm
> not a diver and I don't understand all the terms but I have faith that
> I will learn all of them soon. I also fixed an issue (probably the
> easiest one :) ) and I hope that I will continue my contribution.

It's perfectly OK for the "two patches" to be really simple things.
What I am looking for is to make sure people are able to send clean
patches, can deal well with feedback (a couple other candidates who picked
harder problems had to go through a few "redo" cycles), and overall seem
to fit into the way we do things in Subsurface...

Being a diver is really not necessary. Lubomir, one of our top
contributors, is not a diver. Tomaz, our #1 contributor, wasn't a diver
when he started. He's now certified and has done a couple of dives... but
without meaning any disrespect to him or his skills as a diver... he still
isn't exactly a DIVER :-)

None of the four students from last year were divers, and two of them are
still quite actively contributing...

> Since the idea should be implemented both on Linux and Windows I decided
> to build the project for the last platform too. As I expected this
> process is not as easy as the one for Linux but I am confident that
> I will succeed.

Building natively on Windows is SO MUCH PAIN that I suggest that you don't
do this. It's reasonably easy to cross build from Linux for Windows (which
is how the official binaries are created).
I have switched tool chains about half a year ago, so I'm not sure the
scripts in packaging/windows are still 100% correct, but I can easily make
sure that's the case.

Which Linux distribution do you use?

> I also want to know if:
> - some ideas have a higher priority than others to be accepted, or the
> difference will be made by the applications;

Nope. It will be entirely about the application. I think last year we had
one that several of the principal developers here really really wanted to
have - and we ended up not picking any of the candidates for that idea
since we just didn't think their proposals were strong enough. On the flip
side we had an idea that was more of an afterthought and that we picked
because of a strong candidate (and he's still around, contributing).

> - I should talk here details about the implementation or I should contact
> the mentor assigned for the bluetooth idea;

We don't have the mentors assigned, yet.

> - I can publish my proposal here when it will be in a good shape to
> receive some feedback.

Certainly.


/D


More information about the subsurface mailing list