The Problem with GSOC.

[Joel Bryan Juliano]

GSOC undeniably had been very useful on providing many solutions &
innovations on various open source projects,
not only that, but it also make ways for many people to be involve in
open source software development, branching
many other open source projects.

Despite of the advantages of GSOC, there are also some aspect of it
that is worth looking mostly in the areas of its
mechanics, general outcome and integration, that I hope will provide
improvements on the next GSOC.

Many of the disapproved projects are crossed out, not because they are
not ingenious, in fact, most of them are really
ingenious, but the approver feels that they are easy to do. The
mentors are looking for really big and herculean projects
that requires many pages of coding, and reason out "it's the cause".

Now the consequences are some of the approved projects are finished,
but they grow molds in their CVS, SVN, GIT space,
for the same reason why they are accepted.. They are big, hard to
implement, requires much time and effort to be
integrated on newer versions, contrast to smaller projects that can
easily be implemented, much faster integration, and
number of potential developers that can easily collaborate and submit
patches on the project.

If productivity, participation & finished projects are more concerned,
there are more smaller rocks that can fit inside the
bottle than big rocks.

I hope Google to consider ingenious projects that the mentors say are
easy or not worth the whole summer time to make,
don't judge the solution because they are easy to do, look more
closely on the innovation they will bring.

[Raphael Ritz]

Joel Bryan Juliano wrote:

[..]

Hi Joel,

>
> Many of the disapproved projects are crossed out, not because they are
> not ingenious, in fact, most of them are really
> ingenious, but the approver feels that they are easy to do. The
> mentors are looking for really big and herculean projects
> that requires many pages of coding, and reason out "it's the cause".
>

[..]

Why are you saying this? I'm not sure I agree.
Of course I can only speak for the project I'm
involved in myself (Plone) and while it might be
true that we sort of did this when participating
for the first time we've learned in the meantime
that it is actually *better* to focus on (sometimes
only seemingly) easy projects - we all tend to
underestimate the efforts involved anyway, even
if we try to take this into account.

After all, it is not Google who decides out-of-the-blue
which student projects get accepted but it's the
individual project leaders/mentors who do the ranking and
thereby suggest to Google what to accept and what not.

Different projects may handle this quite differently.
And of course you are free to push them all to
rethink their evaluation criteria ;-)

Raphael

[Leslie Hawthorn]

Hi Joel,

We actually don't decide which project go forward, the mentors choose
those. As for some of the code getting bit rot, that's sad but we
can't compel anyone to continue working on their projects later. We
certainly hope that they will and, if not and is normal for open
source, that another community member will pick up the charge and
carry the work forward.

As for projects that are not accepted into Summer of Code, it's not as
though the ideas cannot be executed. Open source is about
contributing your time to improve things, which everyone is completely
free to do outside of the program. In fact, most people do.

I've heard from many mentors that they are willing to mentor a student
whose application is not accepted. Being a part of GSoC or no is
*not* a barrier to contributing unless someone is looking for a
financial incentive to do so.

Cheers,
LH

--
Leslie Hawthorn
Program Manager - Open Source
Google Inc.

http://code.google.com/opensource/

[Greg]

On Jan 27, 12:54 am, "Joel Bryan Juliano"

<joelbryan.juli...@gmail.com> wrote:
> Many of the disapproved projects are crossed out, not because they are
> not ingenious, in fact, most of them are really
> ingenious, but the approver feels that they are easy to do. The
> mentors are looking for really big and herculean projects
> that requires many pages of coding, and reason out "it's the cause".

Eh? Whatever gave you that idea? Can you produce evidence to support
your assertion?

As a mentor and org. admin, I certainly did not "cross out" a proposal
simply because it was "easy". We chose our applicants based on the
quality of their proposal. If they provided a well-written proposal
with a project plan, timeline, and expected milestones we scored them
highly. Yes, the nature of the project was considered, but we were
more interested in attracting talented and hard-working applicants
than the scale of the project. In fact, none of our students from
last summer worked on "herculean" projects.

-Greg
--
Greg Lund-Chaix
OSU Open Source Lab ~ http://osuosl.org

[Jbs...@gmail.com]

i have no place in this discussion but are there any links to the
projects you guys speak of i've recently stumbled on this website and
am interested in learning more about this.

thanks.

Jabril

[Kulbir Saini]

Hi!

   Checkout http://code.google.com/soc/2008 for details on Google Summer of Code. For projects and organizations involved last year checkout http://code.google.com/soc/2007.
 

thanks.

Jabril
International Institute of Information Technology,
Hyderbad, India - 500032.

My Home-Page: http://saini.co.in/
My Institute: http://www.iiit.ac.in/
My Linux-Blog: http://linux.saini.co.in/
-------------------------------------------------------

[Leslie Hawthorn]

Thanks Kulbir.  :)

--
Leslie Hawthorn
Program Manager - Open Source
Google Inc.

http://code.google.com/opensource/

I blog here:

http://googlesummerofcode.blogspot.com  - http://google-opensource.blogspot.com - http://www.hawthornlandings.org