GSoC 2014

newer
Forthcoming Boost.Fiber review

older
[build] referring to a tool from...

Bjorn Reese

29 Dec 2013 29 Dec '13

11:22 a.m.

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas?

Show replies by date

Christopher Kormanyos

29 Dec 29 Dec

1:56 p.m.

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? Yes, I think Boost should go for it again. I've got a few ideas for Boost.Math and/or Boost.Multiprecision. I need to run

these by my co-authors, but I think we'll be able to put together a well-suited project. I bit off a bit more than I could chew last year with multiple master's students and a good GSoC student. So I should stick with just GSoC in 2014. I would definitely like to give it another run. Wow, that sure came up fast! We're still wrapping up our 2013 efforts. Thanks for the heads up. Sincerely, Chris. On Sunday, December 29, 2013 12:22 PM, Bjorn Reese wrote: Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Vicente J. Botet Escriba

3:57 p.m.

Le 29/12/13 12:22, Bjorn Reese a écrit :

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

Should we start collecting ideas?

Hi, Do we have some results of GSoC 2013? What has been achieved? What is still on going? ... Best, Vicente

Christopher Kormanyos

4:59 p.m.

...

...
Should we start collecting ideas?

...

Do we have some results of GSoC 2013? What has been achieved? What is still on going? ... Best, Vicente

In GSoC 2013, we were awarded 7 slots. As far as I know, all mentors of all 7 projects reported a successful GSoC 2013. I don't think we have a dedicated list of all the results. We finished the core of the Boost.Math project (Bernoulli numbers and their use in the gamma function), but did not get to all of the optional research parts. The results are being merged to a branch of the development branch now and are expected to flow into the next or next-next Boost release. https://github.com/boostorg/math/tree/bernoulli-numbers Sincerely, Chris. On Sunday, December 29, 2013 4:57 PM, Vicente J. Botet Escriba wrote: Le 29/12/13 12:22, Bjorn Reese a écrit :

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

Should we start collecting ideas?

Hi, Do we have some results of GSoC 2013? What has been achieved? What is still on going? ... Best, Vicente _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Niall Douglas

6:05 p.m.

On 29 Dec 2013 at 12:22, Bjorn Reese wrote:

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

Should we start collecting ideas?

I'm thinking I could mentor adding any of the following features to AFIO for GSoC 2014: 1. ASIO style callback API layer, if it is technically possible. If it is not possible, instead we could add a new API layer for ASIO which reworks ASIO into an AFIO style API :) 2. Fiber support. 3. Directory change monitoring, if Paul doesn't get a chance to finish it next few months. As Paul has found, this one is far, far harder than it looks to get right. 4. More wrapping of other Boost libraries into an AFIO style API layer. More of Boost.Filesystem for example. 5. Potential: I'm somewhat into an async batch hash library for AFIO, but while my present implementation works and works well (amortised 1.854 cycles/byte SHA-256 on a four core Intel CPU), I'm not happy with its present design. IF I get enough free cycles to refactor that design into what I think is possible with lots of extra template metaprogramming to help along the compiler's optimiser, I MAY be able to mentor some student work on this area as well. I know this list expressed some interest in having more batch hash algorithms than SHA-256, CityHash and SpookyHash. Niall -- Currently unemployed and looking for work. Work Portfolio: http://careers.stackoverflow.com/nialldouglas/

Bjorn Reese

2 Jan 2 Jan

6:41 p.m.

Here is another idea. Audio library: Provide primitives for playing audio using an Asio-style API. Initially it should be limited to the functionality of OpenSL ES [1], but with the possibility of extending it to the OpenMAX suite[2]. [1] http://en.wikipedia.org/wiki/OpenSL_ES [2] http://en.wikipedia.org/wiki/OpenMAX_AL

rasmus ekman

4 Jan 4 Jan

7:39 p.m.

On 2014-01-02 19:41, Bjorn Reese wrote:

...

Here is another idea.

Audio library: Provide primitives for playing audio using an Asio-style API. Initially it should be limited to the functionality of OpenSL ES [1], but with the possibility of extending it to the OpenMAX suite[2].

[1] http://en.wikipedia.org/wiki/OpenSL_ES [2] http://en.wikipedia.org/wiki/OpenMAX_AL

Audio library, yeah, sure! But absolutely NOT modelled on these libraries. Audio interfaces that limit programmers to playing prerecorded files are useless for most music applications. (Also, did you *look* at the API:s? Piles of overdesigned dross) Anyway, such a project would be doable for a single person, but not sure how exciting conceptually. The interface can be very simple. You basically need to * match client settings with existing capability (ie sample-rate, output data-type), * register a callback function with the underlying system, and * wrap the start/pause/stop functions. The trickery lies in * getting the implementations right across several slightly different underlying systems with different quirks. * making start/stop threadsafely robust wrt client code. Does this fit the scope of GSoC projects? Where to go to get a head start: RtAudio is an audio playback/recording with a simple interface that could serve as starting point: http://www.music.mcgill.ca/~gary/rtaudio/ The interface could be improved in some ways (I'm using it in a wrapper that I'd be happy to share). The PortAudio library covers more systems, and is more widely used. http://www.portaudio.com/ (but to me it has not provided the right combination of functionality and interface simplicity, so no experience) The necessary knowledge about underlying audio systems is present in the two libraries mentioned. They both cover ASIO, ALSA, MacOS, and several Windows systems (PortAudio supports a couple more than RtAudio, but they may be actually obsolete by now). I'd be interested in contributing in any useful way (discussion, review etc - with my decent coding but limited design skills). - rasmus

Klaim - Joël Lamotte

8:19 p.m.

Another kind of audio library that would be interesting would be Plaid Audio: http://interactopia.com/code/plaidaudio/ It's cpu processed audio, mainly used for interractive work like SoundSelf: soundselfgame.com I believe it could be used as a core for a higher abstraction audio library. Another library idea I've been thinking about recently: A Markdown translation library: text in a string as input, html as output, following markdown and kramdown. Other than doing it in a portable modern C++ way, I think focus on efficiency would be important as another attempt was made CPP-Markdown, which had poor efficiency (from several benchmark I gathered only)

David Bellot

6 Jan 6 Jan

3:46 p.m.

I'm very interested about Boost.Compute. We started to explore GPU computations too for Boost.uBLAS. So far, nothing in the main tree but that's a highly desired feature too. What do you guys wanna do with that ? Regarding Audio, I was thinking about another idea: a GFX library in Boost. OK, Herb Sutter proposed that to Cairo a few days ago and maybe we could do a joint GSoC project with Boost, Cairo and the ISO C++ Committee. Regarding uBlas, we've still have plenty of projects again: - solvers. We need solvers - matrix factorization: here I'm thinking more about non-negative matrix factorization approximations, Bayesian factorizations, etc... - more specific support of ARM and/or Power architecture On Sat, Jan 4, 2014 at 8:19 PM, Klaim - Joël Lamotte wrote:

...

Another kind of audio library that would be interesting would be Plaid Audio: http://interactopia.com/code/plaidaudio/ It's cpu processed audio, mainly used for interractive work like SoundSelf: soundselfgame.com I believe it could be used as a core for a higher abstraction audio library.

Another library idea I've been thinking about recently:

A Markdown translation library: text in a string as input, html as output, following markdown and kramdown. Other than doing it in a portable modern C++ way, I think focus on efficiency would be important as another attempt was made CPP-Markdown, which had poor efficiency (from several benchmark I gathered only)

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Marshall Clow

7 Jan 7 Jan

8:10 p.m.

On Jan 4, 2014, at 12:19 PM, Klaim - Joël Lamotte wrote:

...

Another library idea I've been thinking about recently:

A Markdown translation library: text in a string as input, html as output, following markdown and kramdown. Other than doing it in a portable modern C++ way, I think focus on efficiency would be important as another attempt was made CPP-Markdown, which had poor efficiency (from several benchmark I gathered only)

I would love to see this, just because I would use it ;-) — Marshall

Klaim - Joël Lamotte

9:37 p.m.

On Tue, Jan 7, 2014 at 9:10 PM, Marshall Clow wrote:

...

I would love to see this, just because I would use it ;-)

I have a tool project idea (for my own use) that would use it too, but - I don't think I can start working on it soon, certainly not before several weeks or months - I'm not a student so I think I don't fit the GSoC requirements (but I didn't check) However if I manage to work on my project meanwhile, I would definitely propose it to boost (that's what I noted anyway).

Kyle Lutz

3 Jan 3 Jan

9:16 p.m.

I could mentor a student interested in developing for the Boost.Compute GPGPU library [1]. The current focus is on improving performance and implementing missing algorithms. A list of potential projects can be found on the issues list [2]. Also would be interested in proposals to implement a higher-level domain-specific framework to ease the development of GPU-accelerated operations (some ideas are map-reduce, LINQ-style, task-graph / data-flow, etc). -kyle [1] https://github.com/kylelutz/compute [2] https://github.com/kylelutz/compute/issues On Sun, Dec 29, 2013 at 3:22 AM, Bjorn Reese wrote:

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

Should we start collecting ideas?

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Aditya Avinash

11:54 p.m.

Hi Kyle, On Friday, January 3, 2014, Kyle Lutz wrote:

...

I could mentor a student interested in developing for the Boost.Compute GPGPU library [1].

I have spent 3 years programming in cuda and opencl. The current focus is on improving

...

performance and implementing missing algorithms. A list of potential projects can be found on the issues list [2]. Also would be interested in proposals to implement a higher-level domain-specific framework to ease the development of GPU-accelerated operations (some ideas are map-reduce, LINQ-style, task-graph / data-flow, etc).

-kyle

[1] https://github.com/kylelutz/compute [2] https://github.com/kylelutz/compute/issues

What are you really expecting? -- Aditya Avinash Atluri

Christopher Kormanyos

12 Jan 12 Jan

5:20 p.m.

...

Deadline for registering as a mentoring organization for GSoC 2014 is 14 Feb 2014.

...

Should we start collecting ideas?

I believe we have several good ideas. We need to get moving. Does anyone know how to put up the skeleton of this kind of page? (Or can anyone point me in the right direction on how to do it?) https://svn.boost.org/trac/boost/wiki/SoC2013 The hard part, I believe, will be finding one or more org admins. The guys did a great job last year, and set the bar pretty high. * I can probably mentor a student. * I can help collecting and writing the ideas on SVN trac. * I could also help writing the GSoC 2014 proposal and with recruiting. But org admin would potentially be too much of a commitment for me. Is anyone interested in being org admin for GSoC 2014? Best regards, Chris.

Vicente J. Botet Escriba

5:45 p.m.

Le 12/01/14 18:20, Christopher Kormanyos a écrit :

...

...
Deadline for registering as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? I believe we have several good ideas. We need to get moving.

Does anyone know how to put up the skeleton of this kind of page? (Or can anyone point me in the right direction on how to do it?)

https://svn.boost.org/trac/boost/wiki/SoC2013

The hard part, I believe, will be finding one or more org admins. The guys did a great job last year, and set the bar pretty high.

* I can probably mentor a student. * I can help collecting and writing the ideas on SVN trac. * I could also help writing the GSoC 2014 proposal and with recruiting.

But org admin would potentially be too much of a commitment for me.

Is anyone interested in being org admin for GSoC 2014?

Hi, I have just added a page for GSoC 2014 here https://svn.boost.org/trac/boost/wiki/SoC2014. Be free to add your projects and ideas. Vicente

Christopher Kormanyos

8:38 p.m.

...

I have just added a page for GSoC 2014 here https://svn.boost.org/trac/boost/wiki/SoC2014

Thank you Vicente. I have added "Boost.Math Generalized Hypergeometric Functions". Sincerely, Chris. On Sunday, January 12, 2014 6:45 PM, Vicente J. Botet Escriba wrote: Le 12/01/14 18:20, Christopher Kormanyos a écrit :

...

...
Deadline for registering as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? I believe we have several good ideas. We need to get moving.

Does anyone know how to put up the skeleton of this kind of page? (Or can anyone point me in the right direction on how to do it?)

https://svn.boost.org/trac/boost/wiki/SoC2013

The hard part, I believe, will be finding one or more org admins. The guys did a great job last year, and set the bar pretty high.

* I can probably mentor a student. * I can help collecting and writing the ideas on SVN trac. * I could also help writing the GSoC 2014 proposal and with recruiting.

But org admin would potentially be too much of a commitment for me.

Is anyone interested in being org admin for GSoC 2014?

Hi, I have just added a page for GSoC 2014 here https://svn.boost.org/trac/boost/wiki/SoC2014. Be free to add your projects and ideas. Vicente _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Ankit Vadehra

13 Jan 13 Jan

5:14 a.m.

Hi, Could someone please. Add the list of Advanced Data-Structures that Boost is trying to fund/get this summer of SoC ?? Either as an individual project/library, or as a part of an existing Library ?? Thanks. Ankit Vadehra On 1/13/14, Christopher Kormanyos wrote:

...

...
I have just added a page for GSoC 2014 here https://svn.boost.org/trac/boost/wiki/SoC2014

Thank you Vicente.

I have added "Boost.Math Generalized Hypergeometric Functions".

Sincerely, Chris.

On Sunday, January 12, 2014 6:45 PM, Vicente J. Botet Escriba wrote:

Le 12/01/14 18:20, Christopher Kormanyos a écrit :

...
...
Deadline for registering as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? I believe we have several good ideas. We need to get moving.

Does anyone know how to put up the skeleton of this kind of page? (Or can anyone point me in the right direction on how to do it?)

https://svn.boost.org/trac/boost/wiki/SoC2013

The hard part, I believe, will be finding one or more org admins. The guys did a great job last year, and set the bar pretty high.

* I can probably mentor a student. * I can help collecting and writing the ideas on SVN trac. * I could also help writing the GSoC 2014 proposal and with recruiting.

But org admin would potentially be too much of a commitment for me.

Is anyone interested in being org admin for GSoC 2014?

Hi,

I have just added a page for GSoC 2014 here https://svn.boost.org/trac/boost/wiki/SoC2014.

Be free to add your projects and ideas. Vicente

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

-- Ankit.Vadehra "The Code Is free, But its Priceless" "Exactly, which way is Forward ?"

Christopher Kormanyos

14 Jan 14 Jan

10:13 p.m.

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

...

Should we start collecting ideas?

We now have two suitable projects and two veteran mentors. This is probably enough to apply. https://svn.boost.org/trac/boost/wiki/SoC2014 * We need to write the application. * I can assist with writing the application. * Which does not imply that I desire to be the org admin 2014. Can anyone help me find the application materials for GSoC 2013? Can anyone assist in the application process? Is there anyone better to charge ahead here? Thank you. Sincerely, Chris On Sunday, December 29, 2013 12:22 PM, Bjorn Reese wrote: Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014. Should we start collecting ideas? _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

Boris Schäling

18 Jan 18 Jan

12:58 p.m.

On Sun, 29 Dec 2013 12:22:10 +0100, Bjorn Reese wrote:

...

Deadline for registring as a mentoring organization for GSoC 2014 is 14 Feb 2014.

Should we start collecting ideas?

Please collect as many ideas as possible and put them on https://svn.boost.org/trac/boost/wiki/SoC2014 (thanks for setting up this page, Vicente). It's really important to have lots of good project proposals on that page as Google is going to look at it to decide whether Boost should become a mentoring organization (and how many GSoC projects we might get). Regarding GSoC admin: As some of you know I was the admin last year. I would be happy to pass the torch to someone else though. Anyone interested? (The most important job of the admin is to make sure that GSoC deadlines are met.) Boris

Bjorn Reese

1:06 p.m.

On 12/29/2013 12:22 PM, Bjorn Reese wrote:

...

Should we start collecting ideas?

Another idea: a rolling hash [1]. [1] http://en.wikipedia.org/wiki/Rolling_hash

Andrea Grimaldi

20 Jan 20 Jan

11:04 a.m.

Hi, I don't know if it's place and time to do this question. I'm a student of Phisics from Padua and I'm interested in compute science like simulation, Monte Carlo Method, PDE, EDO, etc. I read that there's a projects about it and I would like to know if there is any chance work as a part of Gsoc. Andrea 2014/1/18 Bjorn Reese

...

On 12/29/2013 12:22 PM, Bjorn Reese wrote:

Should we start collecting ideas?

...
Another idea: a rolling hash [1].

[1] http://en.wikipedia.org/wiki/Rolling_hash

_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/ mailman/listinfo.cgi/boost

-- Andrea Grimaldi Matricola: 1027593

4112

Age (days ago)

4134

Last active (days ago)

List overview

Download

20 comments

13 participants

participants (13)

Aditya Avinash
Andrea Grimaldi
Ankit Vadehra
Bjorn Reese
Boris Schäling
Christopher Kormanyos
David Bellot
Klaim - Joël Lamotte
Kyle Lutz
Marshall Clow
Niall Douglas
rasmus ekman
Vicente J. Botet Escriba

GSoC 2014

Bjorn Reese

Christopher Kormanyos

Vicente J. Botet Escriba

Christopher Kormanyos

Niall Douglas

Bjorn Reese

rasmus ekman

Klaim - Joël Lamotte

David Bellot

Marshall Clow

Klaim - Joël Lamotte

Kyle Lutz

Aditya Avinash

Christopher Kormanyos

Vicente J. Botet Escriba

Christopher Kormanyos

Ankit Vadehra

Christopher Kormanyos

Boris Schäling

Bjorn Reese

Andrea Grimaldi

tags

participants (13)