Google Summer of Code 2008?

[Chiyuan Zhang]

Hi,

Ruby community is participating Google Summer of Code
through Ruby Central. However, although project ideas
of other organizations have been announced, I found
that there's still no under the Ruby Central page.
I'm willing to take part in Google Summer of Code this
year, especially working for a Ruby related project.

So, what's the current status of Ruby of GSoC? Is there
any project ideas collected? Maybe something about Ruby 1.9,
Rubinius etc. I myself haven't come up any good idea
yet. So I'm looking forward to the suggested ideas.

Thanks!

 

[pat eyler]

Hi,

Ruby community is participating Google Summer of Code
through Ruby Central. However, although project ideas
of other organizations have been announced, I found
that there's still no under the Ruby Central page.
I'm willing to take part in Google Summer of Code this
year, especially working for a Ruby related project.

So, what's the current status of Ruby of GSoC? Is there
any project ideas collected? Maybe something about Ruby 1.9,
Rubinius etc. I myself haven't come up any good idea
yet. So I'm looking forward to the suggested ideas.

I hope to have this up today. I'd hoped to have it posted
yesterday, but the best laid plans ...

If anyone has ideas they'd like to share, I'm all ears.

 

[James Gray]

Is there any project ideas collected? Maybe something about Ruby 1.9,
Rubinius etc.

I'm sure we can brainstorm some for you.

Regarding your Ruby 1.9 idea: many libraries don't work correctly yet
in Ruby 1.9. You could come up with a list of favorites that don't
work and offer to patch them for Ruby 1.9. I think that could be
quite a resource to the community. I got this idea from Greg Brown
who may be doing a similar project soon, but I think there are more
than enough libraries to go around.

Beyond that, here are a few other random ideas:

* There's been a lot of talk about better documentation tools lately.
I believe Eric Hodel is working on revamping RDoc. Perhaps you could
see what you could do to help with that. James Britt has also looked
at better ways to assemble the documentation for ruby-doc.org, so that
may be another place you could help out.
* I've recently had need of a multiprocess event safe logger. What I
mean by that is that I have two processes writing to the same log
file. They may log five things for each event. In the log file, I
would prefer the resulting ten entries not be interleaved. You could
provide a start event method and an end event method and only add
events to the log file after the event ended. This is more
complicated than it sounds since you will want to deal with large
logging data for a single event (probably using Tempfile), badly
behaved programs that don't manage to call the end event method, long
running events, etc.
* Daniel Berger has requested a pure Ruby implementation of zlib a few
times now. This might give new options for Windows installers.

I do encourage you to think up your own ideas though. I've been a
mentor for the last two years and I can easily say that the
applications for unique ideas were generally more appealing. There
are two reasons for that. The first is that lots of applications come
in for the ideas we post publicly, so you're competing with a lot of
people who want to do the exact same thing. The other is that you
will just be more creative when running with your own idea. Don't
hesitate to latch onto one of your own needs and run with it.

James Edward Gray II

 

[Michal Suchanek]

I'm sure we can brainstorm some for you.

There's been some talk about the memory management - the GC. The GC
used in 1.8 is clearly suboptimal. Running the same code under jruby
results in great cut on memory requirements and thus great performance
gain, at least for my application.

I am not sure about the status and plans for the 1.9 GC.

If you are into this kind of stuff you will likely not face too much
competition, and could make a lasting contribution to the 1.9
implementation ;-)

Thanks

Michal

 

[Chiyuan Zhang]

Hi, Thank you for so many ideas. And maybe you're right. I might
also think about my own ideas besides those published.

 

[RaL]

I want to create a project devoted to Neural Networks. Would Ruby
Central be interested to support me and this project for Google Summer
of Code?

Right now, there are two gems with a Neural Network support -
http://ai4r.rubyforge.org/ =C9 http://ruby-fann.rubyforge.org/. But both
of them implement only one learning algorithm - back-propagation (and
accordingly to that fact only one type of Neural Network - multilayer
perceptron) - also they don't have ability to choose different
activation functions for different layers and so you can't even build
Radial Basis Function Network with those gems. They don't have a built-
in interface to use even widespread learning strategies (e.g., begin
learning with a high learning rate and then switch to a low learning
rate after approaching some threshold).

So as a GSoC project I decide to implement several Neural Network
types, including multilayer perceptron, self-organizing feature map
(Kohonen net), auto-associative memory (Hopfield net) and recurrent
neural network. I want to build an easy but powerful interface to use
with. So it will support a range of learning strategies for fast start
but it will be very customizable in case of need. It will have a built-
in visualization module so you could see a net topology with all
values of weights, errors, input and output values in any moment of
time. You could also see a graphic of values (good to see a final
approximation) and errors (to understand a dependence on epoch count).
And as the last step i want to build a Neural Network web-editor via
silverlight and IronRuby.

So what does RubyCentral think about my proposal?

I'm sure we can brainstorm some for you.

Regarding your Ruby 1.9 idea: =9Amany libraries don't work correctly yet = =9A
in Ruby 1.9. =9AYou could come up with a list of favorites that don't =9A
work and offer to patch them for Ruby 1.9. =9AI think that could be =9A
quite a resource to the community. =9AI got this idea from Greg Brown =9A
who may be doing a similar project soon, but I think there are more =9A
than enough libraries to go around.

Beyond that, here are a few other random ideas:

* There's been a lot of talk about better documentation tools lately. =9A
I believe Eric Hodel is working on revamping RDoc. =9APerhaps you could = =9A
see what you could do to help with that. =9AJames Britt has also looked = =9A
at better ways to assemble the documentation for ruby-doc.org, so that =9A=

 

[James Gray]

I want to create a project devoted to Neural Networks. Would Ruby
Central be interested to support me and this project for Google Summer
of Code?

Obviously I don't speak for Ruby Central, but I think your idea is
great. Definitely send it in. Be sure and mention your plan, as you
did in this email, and the kinds of problems it would help users
solve. Good luck!

James Edward Gray II

 

[pat eyler]

Obviously I don't speak for Ruby Central, but I think your idea is
great. Definitely send it in. Be sure and mention your plan, as you
did in this email, and the kinds of problems it would help users
solve. Good luck!

I agree, this sounds like a very interesting project. No decisions will be
made about which projects to sponsor until we see all the proposals,
and the prospective mentors can weigh in on them.

 

[Michael Neumann]

I want to create a project devoted to Neural Networks. Would Ruby
Central be interested to support me and this project for Google Summer
of Code?

Right now, there are two gems with a Neural Network support -
http://ai4r.rubyforge.org/ É http://ruby-fann.rubyforge.org/. But both
of them implement only one learning algorithm - back-propagation (and
accordingly to that fact only one type of Neural Network - multilayer
perceptron) - also they don't have ability to choose different
activation functions for different layers and so you can't even build
Radial Basis Function Network with those gems. They don't have a built-
in interface to use even widespread learning strategies (e.g., begin
learning with a high learning rate and then switch to a low learning
rate after approaching some threshold).

If you are also interested in pulsed (or spiking) neural nets, take a
look at Yinspire [1], which is a fast simulator I am developing at my
university using Ruby and C++ (using Cplus2Ruby).

Regards,

Michael

[1]: http://www.ntecs.de/projects/yinspire/