Google Summer of Code 2009
Burcin Erocal
Are we doing anything for the Google Summer of Code this year?
The applications for mentoring organizations open tomorrow (March 9),
and end on Friday (March 13). The timeline is here:
http://code.google.com/opensource/gsoc/2009/faqs.html#0_1_timeline_5354032302481437_
I started a wiki page for project ideas, basically by copying parts of
previous years page.
http://wiki.sagemath.org/gsoc09
More information about the program available here:
Here is a blog post by a gentoo developer with some relevant questions
we could answer in our application as well:
http://dberkholz.wordpress.com/2009/03/01/you-can-help-get-gentoo-into-the-summer-of-code/
Cheers,
Burcin
William Stein
>
> Hi,
>
> Are we doing anything for the Google Summer of Code this year?
organization application for Sage. If you or somebody else wants to,
volunteer now. After receiving a "no" answer three times in a row, I
don't personally want to take the rejection yet again. :-)
I'm definitely +1 on somebody (not me) applying Sage as an organization.
> The applications for mentoring organizations open tomorrow (March 9),
> and end on Friday (March 13). The timeline is here:
>
> http://code.google.com/opensource/gsoc/2009/faqs.html#0_1_timeline_5354032302481437_
>
>
> I started a wiki page for project ideas, basically by copying parts of
> previous years page.
>
> http://wiki.sagemath.org/gsoc09
>
>
>
> More information about the program available here:
>
> http://code.google.com/soc/
>
> Here is a blog post by a gentoo developer with some relevant questions
> we could answer in our application as well:
>
> http://dberkholz.wordpress.com/2009/03/01/you-can-help-get-gentoo-into-the-summer-of-code/
>
>
> Cheers,
>
> Burcin
>
> >
>
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org
Harald Schilly
> Are we doing anything for the Google Summer of Code this year?
maths related tasks, that are "easy" and comprehensible from a
software engineering standpoint. i.e. authentication with existing
systems (ldap, active directory,...) and embedding into things like
moodle come to my mind. otoh avoid anything that could confuse, don't
mention mathematical terms and so on. I think the chances are higher
if we could boil it down to this sort of technical tasks with short
and clear descriptions.
H
Burcin Erocal
+10
This is why I copied the notebook section as an example from the list
for 2008. Even though I would expect Google to be interested in funding
open source implementations of maths, I think it would be better if we
focus on the computer science/engineering side of things.
If we are going ahead with the application, we should try to make the
ideas list very clear, and add well defined milestones/deliverables for
the tasks on the list.
Cheers,
Burcin
Burcin Erocal
On Sun, 8 Mar 2009 06:46:17 -0700
William Stein <wst...@gmail.com> wrote:
>
> On Sun, Mar 8, 2009 at 4:50 AM, Burcin Erocal <bur...@erocal.org>
> wrote:
> >
> > Hi,
> >
> > Are we doing anything for the Google Summer of Code this year?
>
> I hope so. Note that I'm personally not going to fill out a mentor
> organization application for Sage. If you or somebody else wants to,
> volunteer now. After receiving a "no" answer three times in a row, I
> don't personally want to take the rejection yet again. :-)
We should definitely apply and not be discouraged from the rejections from the previous years.
As Harald wrote in a different post, as long as we state the tasks
clearly, concentrate on computer science, software engineering tasks,
and define clear milestones that any computer science student can
start working on, then we will have a competitive application.
I think you should still do the application, since you're officially
the BDFL of the Sage project, and you already have lots of experience
with the process. I can help with the paperwork, as many others on
this list would also do.
> I'm definitely +1 on somebody (not me) applying Sage as an
> organization.
I can do some work on this, but I don't think I can handle the task all
by myself. But then, if people are not volunteering to work as mentors,
suggest tasks, or help with the application, we shouldn't apply at all.
So please, everyone, suggest project ideas with well defined goals
achievable with less than 3 months of work by computer science majors.
It would be great if people familiar with the notebook volunteered to
be mentors, and extended the items already on the wiki page.
> > The applications for mentoring organizations open tomorrow (March
> > 9), and end on Friday (March 13). The timeline is here:
>
> It's first come first served, I think, so whoever does this should do
> it ASAP!
I looked at the FAQ shortly, but didn't see any mention of this. Does
anybody have more information on this?
We should still prepare the application ASAP, of course. :)
Please submit ideas to the wiki page:
http://wiki.sagemath.org/gsoc09
I'll also start a wiki page for the questions we need to address as a
part of the applications.
Cheers,
Burcin
Burcin Erocal
On Sun, 8 Mar 2009 16:08:22 +0100
Burcin Erocal <bur...@erocal.org> wrote:
> We should still prepare the application ASAP, of course. :)
>
> Please submit ideas to the wiki page:
>
> http://wiki.sagemath.org/gsoc09
>
>
> I'll also start a wiki page for the questions we need to address as a
> part of the applications.
I copied some of the questions from Google's FAQ, and put them on this
wiki page:
http://wiki.sagemath.org/gsoc09_application
Now all we need is some answers. :)
Burcin
Ronan Paixão
>
> +10
>
> This is why I copied the notebook section as an example from the list
> for 2008. Even though I would expect Google to be interested in funding
> open source implementations of maths, I think it would be better if we
> focus on the computer science/engineering side of things.
>
> If we are going ahead with the application, we should try to make the
> ideas list very clear, and add well defined milestones/deliverables for
> the tasks on the list.
>
>
> Cheers,
>
> Burcin
I'd recommend improvements to plotting functions. It's the "face" of
most math software (from a non-mathematician standpoint) and there's
room for improvement. Implementing different axes (like log), fixing
jmol's issues that happen for some people, the mayavi output that's
being worked on by cwitty and Prabhu, generalization to allow plotting
by Pyx, or new 3D output like Blender or Yafray, plotting of new types
that currently don't work like Piecewise functions, etc come to mind.
Also it looks more computer-related than math-related.
Another idea would be improvements towards the Windows port. The only
problem with that is that it would be less noticeable from a user's
standpoint, since most probably a complete port wouldn't be viable in 3
months time by a single person.
I guess the switch to new symbolics and pynac would be more math-related
than desirable by the GSoC program, but maybe transitioning Maxima to
that less-buggy Lisp (which I don't recall which one right now) is also
an idea.
That's my 3 cents,
Ronan
davidp
http://people.reed.edu/~davidp/412/
and I have a student that would love to work with me this summer to
really polish it up. (A preliminary version of a sandpile applet he
wrote is at www.reed.edu/~headb/sandpiles.) It might not be what
Google is looking for, though.
David
Harald Schilly
On Mar 9, 3:04 am, davidp <dav...@reed.edu> wrote:
> and I have a student that would love to work with me this summer to
> really polish it up. (A preliminary version of a sandpile applet he
which mentoring organizations are selected. This is independet of your
student! If (and only if) Sage is selected as a mentoring
organization, your student could write a proposal to Sage for his/her
own project. Then, the Sage group decides (ranks all submissions)...
H
Tim Lahey
in plotting is a good idea. I recommend a canvas matplotlib
back end. This would greatly benefit the notebook and would
benefit the matplotlib project as well.
Of course, I'd prefer improvements to pynac based integration,
limits, and ODEs, but I think the canvas back end has a
decent chance of being funded.
Cheers,
Tim.
Harald Schilly
> ... but I think the canvas back end has a
proposal for Sage as a mentor group. Therefore it should focus on such
tasks that have no maths background and average IT students are able
to solve them. Later on, iff Sage is selected as a mentor group and
students submit proposals for different tasks (integration, etc.) it
could be discussed. If Sage is not selected, nothing will happen :\
So, I hope some mentors pop up and we manage to write a decent
proposal in the next few days! There is enough non-maths stuff that
needs to be done.
H
Jason Grout
Is Cython going to be entering again this year? What about scipy/numpy,
sympy, ginac, or other projects that Sage relies on? I can see
enhancing pynac as falling under the ginac project, for example.
As another idea for something that would be fantastic: how about
finishing Davide Cervone's javascript equation editor? jsmath is used
all over the place, and having an online equation editor would be great
too. I'm referring to:
http://www.math.union.edu/~dpvc/talks/2006-12-08.IMA/editor.html
Thanks,
Jason
David Joyner
>
> Based upon a lot of the comments in recent weeks, improvements
> in plotting is a good idea. I recommend a canvas matplotlib
> back end. This would greatly benefit the notebook and would
> benefit the matplotlib project as well.
Robert Bradshaw
> Harald Schilly wrote:
>> On Mar 9, 11:27 am, Tim Lahey <tim.la...@gmail.com> wrote:
>>> ... but I think the canvas back end has a
>>> decent chance of being funded.
>>
>> Yes, but first of all, this week until the 13th we need to have a
>> proposal for Sage as a mentor group. Therefore it should focus on
>> such
>> tasks that have no maths background and average IT students are able
>> to solve them. Later on, iff Sage is selected as a mentor group and
>> students submit proposals for different tasks (integration, etc.) it
>> could be discussed. If Sage is not selected, nothing will happen :\
>> So, I hope some mentors pop up and we manage to write a decent
>> proposal in the next few days! There is enough non-maths stuff that
>> needs to be done.
+1 to all the suggestions to improving plotting. I think that's an
area that could be tackled by someone with less math background, and
also produce tangible results.
> Is Cython going to be entering again this year? What about scipy/
> numpy,
> sympy, ginac, or other projects that Sage relies on? I can see
> enhancing pynac as falling under the ginac project, for example.
It doesn't look like we'll have enough potential mentors to be a full
organization, but we'll probably try and do a project or two under
the Python umbrella (or perhaps even scipy/numpy).
- Robert
Burcin Erocal
+1
We need mentors to have a decent proposal. So please volunteer if
you can supervise a student during the summer. Note that nobody agreed
to be a mentor so far.
The GSoC FAQ says that on average mentors need to spend about 5 hours
per week for this. For each mentor, the organization gets $500, and
the mentor gets a t-shirt. I guess the organization to receive the said
$500 would be the Sage foundation, though maybe there are options to
pay the mentors directly.
Especially people who are familiar with the tasks already mentioned in
this thread, and on the ideas page
http://wiki.sagemath.org/gsoc09
should volunteer.
Talking about a decent proposal... I've added some answers to the
application at
http://wiki.sagemath.org/gsoc09_application
The first question still has the text Harald copied from the web site.
I plan to edit this to contain a better (and slightly longer)
introduction to Sage and the community, mentioning the # of downloads,
contributors, etc.
The question about the application template is still unanswered. If you
have any ideas of questions to ask students who apply, please add them
here, or even better, add them to the wiki.
I wrote most of the answers on and off during the day, and some in the
evening now. It would be great if some people could go through it and
add corrections, write comments, or rewrite it all.
We should also work on the ideas list. We can list the ideas in the
wiki page initially, but it would be good to turn these into projects
for the application. I suggest putting a few of the ideas together,
such as refactoring the notebook code and adding an authentication
framework, then the list of possible mentors with the project. We can
also add some motivating text for the tasks, and possible milestones.
BTW, there is no "first come first serve" policy for applications. I
think this was an idea William came up with to make me do all this work
today and not put it off till the end of the week. :)
Special thanks to Harald for helping out with the questions and ideas.
With the suggested additions of plotting improvements (maybe
including the matplotlib canvas, if they are not applying as an
organization), javascript equation editor, and maybe pynac (ginac is not
applying, afaik), and of course commitments from our first rate sage
developer mentors, we'll have a great application.
That's all for now... I'm going to sleep. :)
Cheers,
Burcin