Google Announces Summer of Code 2008

morrison writes "The 2008 Google Summer of Code is on. We have discussed this four-year-old tradition before (2005, 2006, 2007). Google will once again be hosting a program that gives computer science students a $4,500 stipend to work on open source software projects. Last year, Google funded over 900 students' projects in more than 90 countries. As noted in the program FAQ, this year they hope to do even more. The #gsoc IRC channel on Freenode is already buzzing with activity."

Comments

What should get precedence?

[calebt3]

Insert worthy projects below here.
I personally hope Blender gets work.

Re:What should get precedence?

[Breakfast Pants]

Obama needs some tweaks.

Re:What should get precedence?

[Anonymous Coward]

Fix the Firefox memory leak! No wait, something more realistic... how's about world peace?

Re:What should get precedence?

[Nimey]

The anti-ballistic-chair defense system. Google's going to need it some day.

Re:What should get precedence?

[smittyoneeach]

Emacs, though thought by some to be older than the Renaissance, could enter one...

Regarding the Obama ad served on this page: "It's a cookbook!!!"

Re:What should get precedence?

[mithro]

Thousand Parsec (a game framework for turn based strategy games) was one of the mentor organisations last year.

The effect on our project was really huge, not only did the students do some very cool work. We now have the creditability to approach Universities and help get their students involved with our project.

We already have one student working on Thousand Parsec as part of a high school internship and two students from the University of South Australia working on a Java MIDP client.

Thanks a huge amount to Google and the Summer of Code team, hopefully we can get in again this year and have even more fun!

Re:What should get precedence?

[Matt Perry]

The anti-ballistic-chair defense system. Google's going to need it some day.

I'm surprised that Microsoft doesn't star their own "Summer of Code" considering how they keep saying that developers are so important.

Re:What should get precedence?

[Enderandrew]

This is a bit of an urban legend at this point.

1 - Any complex app will likely have some memory leaks. The code has been very thoroughly examined and cleaned up for Firefox 3.
2 - Most "leaks" come from poorly written extensions/add-ons. Run without them and check out the difference.
3 - There is a feature in Firefox that you can easily turn off, that people mistake for a memory leak. Firefox keeps fully rendered versions of pages in memory, in addition to the standard cache on the hard disk. If you hit back, Firefox doesn't need to re-render the page. Browse a while, and Firefox will use up plenty of memory. If you don't like this behavior, then turn the feature off.

Re:What should get precedence?

[Matt Perry]

they already have winter, spring, and fall. MS allows free (as in no cost, not five finger discount) downloads of all their programming/OS/Server software at my uni. (This is something MS "donated", they aren't getting paid for it).

It isn't about donating software. Software is cheap. Those same students can get free operating systems and development software that's non-Microsoft too. What Google is doing is donating the organizational skills to help students. They get to work on something that's larger than just a small personal project. They learn how to work within a larger team structure that may have established rules for code style, structure, documentation, etc. Most importantly, they are assigned a mentor who can help them navigate this new environment and help them to become better programmers. The financial reward isn't bad either. Microsoft isn't doing anything like that.

Re:What should get precedence?

[jesser]

Why? For most users, having fast (and accurate!) back/forward is more useful than having Firefox allocate less RAM. The feature even automatically turns itself off if you don't have a lot of physical RAM.

Turning off bfcache might be useful for rudimentary leak detection, but a proper leak-detection tool is less likely to be confused by fragmentation, other caches, or the OS simply not reclaiming memory that the application has relinquished.

Re:What should get precedence?

[Mandrake]

actually, that is not true. in fact, X allocates pixmaps for drawing to the screen most of what you're displaying in firefox - many of which are in subwindows of the main firefox application. X is allocating memory for the application, in other words. if not freed properly (due to, say, potential bugs in this person's X server) this can, in fact, cause memory to leak that lasts beyond the termination of the actual application (firefox).

4th year in a row?

[Breakfast Pants]

My open source Visual Basic extension for Word 97 has been rejected 3 times already; I'm gonna try one last time.

Re:Kids have it lucky these days

[redalien]

Actually, Google Highly Open Participation contest produced some excellent pieces of code that were all submitted by "high school" students. If I didn't know better I'd say they were professional developers.

Re:Kids have it lucky these days

[LetterRip]

Granted pretty much nobody in high school will write quality code (even if they honestly think they do, like I once did), the chance to get paid experience and a mentor to help you improve is fantastic.

Actually Nick Bishop who did SoC with Blender 2 years ago, and with Inkscape (I think?) last year, had pretty good code quality already as a high school student.

LetterRip

OpenMoko, coreboot, and ATI video drivers

[Anonymous Coward]

GO!

Too bad I get paid too much and actually have a real job. I'm being serious, it would be awesome to do a summer of code.

Re:OpenMoko, coreboot, and ATI video drivers

[CSMatt]

Agreed. I really need new ATI drivers. Neither the free drivers nor fglrx will allow me to suspend my laptop.

Metascore sure needs developers

[Anonymous Coward]

The development team is meeting for the first time in March. It is a rather ambitious project, but the code itself seems like it would be simple.

http://sourceforge.net/projects/metascore/

My hope...

[bogaboga]

...is that enough resources get geared to having KDE 4.1 as complete as can possibly be. Guys, KDE 4 rocks and can be made better. Go guys.

Awesome

[katterjohn]

I successfully participated last summer working with Nmap. Leslie (from Google) and Fyodor were wonderful to work with, and I hope I can get in again this year!

Great job, Google!

Check out Gladex

[charlie763]

I don't think we'd make it into GSoC, but if you are into Python and Glade you should checkout Gladex. We're even a Featured Project on Launchpad.net! Gladex isn't in the Ubuntu or Debian repositories yet, but we do have a PPA going of an alpha release. Alternatively, you can download the stable packages directly.

Gladex is a Python application which takes a .glade file created in the Glade User Interface Builder and generates code in Perl, Python, or Ruby. The generated code uses libglade to draw a GUI and is not raw pygtk code (support via a plugin is in development). Support for additional languages can be added through the plugin API.

Re:Check out Gladex

[AlXtreme]

Looks interesting, but one question that I couldn't find on the site: Why generate code at all?

I might be missing something, but libglade has a python wrapper (and probably Ruby/Perl too, I'm too lazy to check). You can connect signals and handle events, everything you can do with generated or self-written GTK+ code. Calling libglade in Python is about 4 LOC. Why would anyone need a generator for this?

Again, I'm not trolling here, just curious and both an avid GTK+ and Python user.

Re:Check out Gladex

[ricegf]

You can connect signals and handle events, everything you can do with generated or self-written GTK+ code.

The one very important thing you can do with generated code that you can't do with a 4-line call to libglade is customize the code. A library can't do everything custom code can do, though good libraries seem to cover 80-90% of the most common use cases. For many applications, the libraries do everything you need. For everything else, you'd like to generate the commonplace code from your glade design and then customize the small portions that are unique.

Have ANY projects been completed and integrated?

[batkiwi]

I know from looking the last 2 years that the projects for both PSI and MythTV were accepted and started but never completed to a point where the maintainers put code into the full product.

Are there ANY success stories?

Re:Have ANY projects been completed and integrated

[katterjohn]

Are there ANY success stories?

Absolutely. My fellow SoC students and I participating with Nmap last year have lots of code in Nmap proper. And the years before that (Nmap has participated every year of the SoC) there were a whole lot of cool things added to Nmap proper from SoC work.

Have ANY projects actually been useful?

[SuperBanana]

MythTV? You're joking, right?

More importantly, are they going to work on anything actually *useful*, instead of sexy stupid stuff that is the 2008 equivalent of "skinning" mp3 players? Every time I heard about SoC participants, I noticed that a)it wasn't something really useful or important and b)the main development team was really lazy about integrating in the work the student had done.

A great example of where some SoC lovin' would be great: Netatalk *blows*. It doesn't handle sleeping clients that try to reconnect, and they've sat on their fucking hands for YEARS with the whole openssl/GNU licensing debacle. It's still impossible for any distribution to distribute netatalk with SSL support compiled in (Debian and Ubuntu being two big examples.) Leopard now *requires* encrypted password support- you get an immediate error if the server doesn't support it (rightfully so.)

And no, Samba isn't an acceptable alternative. It vastly underperforms versus AFP on the same hardware/network, and doesn't support a lot of functionality Macintosh programs require- Quickbooks, for example, won't open a quickbooks file on a SMB/CIFS server.

If one or two Summer of Code students sat down and worked on improving netatalk, they'd be instantly loved by many the world over. I dare say that netatalk would do well from (another) code split; they haven't done a release in over TWO YEARS.

Re:Have ANY projects been completed and integrated

[jesser]

I believe Firefox 3's implementations of resumable downloads and the APNG image format came from GSoC participants. The continued support of MathML in Firefox 3 may also be due in part to the work of a GSoC participant. We've also had a few not-so-successful GSoC projects.

Re:write a decent reporting tool

[DNS-and-BIND]

No way, man! Do you know how boring that would be to develop? What we really need is transparency and alpha channel blending effects. And skins! There is a serious dearth of quality skins for mplayer!

Re:computer science students?

[Lincolnshire Poacher]

> This isn't limited to computer science students.

Quite true, but why do Google restrict participation to students?

The first goal listed on their SoC FAQ is:

``Get more open source code created and released for the benefit of all''

So why exclude professional developers who could crank out code?

I would dearly like to take a two-month sabbatical from work and
concentrate solely on writing code. There are huge voids in the
provision of free astronomical tools that could be addressed. But
finances dictate otherwise.

Instead, vast swathes of time and money will be wasted as students
learn about version control, rediscover elementary mistakes and
become entrapped in the politics of open source.

Thanks for nothing, Google.

Re:Small Winner and Big Winner

[Nazlfrag]

Even when you add the small winners together, Google still wins. Lot's of outstanding brainpower for dirt cheap. You're right, but it's not only brainpower Googles getting. The publicity itself would be worth the money, with the headhunting opportunities a distant second.

Re:Have ANY projects been completed and integrated

[Fweeky]

Quite a few FreeBSD SoC projects make it into the system or ports, or at least had some of their work help with it; a quick glance at the SoC wiki pages I see enhancements to libalias and ipfw (I think some of this eventually made it; we now have kernel NAT with ipfw), bsnmpd bridge monitoring, FUSE port, gvinum enhancements, GEOM storage virtualisation, Apple hardware support enhancements, and what became the name service caching daemon.

Other things may not have made it in, but were good research projects both for the project and for the students; FreeBSD now has a very functional port of OpenBSD's hardware sensors suite, though it wasn't accepted into base because of architectural concerns. gjournal started life as a SoC project, and while rejected it did help spur development of a new more functional one, and the student went on to produce gvirstor, the aforementioned GEOM storage virtualisation layer which *did* make it. The Linux KVM port got far enough to boot FreeBSD 7 as a guest and will hopefully continue development. I'm sure I've left lots out.

Just because a SoC project doesn't make it into a "product", doesn't mean that project wasn't a success. Even if it never produces something deployable, it's given a student some experience in development, it's given the project some interesting if not necessarily immediately useful code and it's helped lay groundwork for future development, even if it only does so by providing those concerned some experience.

Quick introduction for those thinking of applying

[morrison]

BZFlag participated in the Google Summer of Code for the first time in 2007. Our participation was documented in this detailed article (Warning: 15 MB PDF).

Another higher-level summary was put together for a presentation and is available here (Warning: 5 MB PDF)

See the presentation for the quick introduction. I highly recommend the article to any students and projects/mentors that are seriously thinking about participating for the first time.

On the whole, it's a great opportunity for projects but you do have to put in a lot of time and effort. You have to have your act together. If you do, the students and the projects will both have a great time.

Re:Kids have it lucky these days

[Corsix]

While you are correct about many people in high school, some can write high quality code, as shown in the Google Highly Open Participation contest. With the Drupal project, there is a 12 y/o who is too young even for GHOP who writes very good code. Corsix -- GHOP Drupal grand prize winner

Re:computer science students?

[nuzak]

Psst. code.google.com works for everyone.

Some people will bitch about anything and everything I guess.