Google Summer of Code Project Breakdown

behdad writes "Google's Summer of Code final per-organization project breakdown is out. The Apache Software Foundation is on the top of the list with 38 projects allocated out of total 410 slots, followed by KDE, FreeBSD, and 38 other mentoring organizations. The accepted applications will be posted early next week. More than 8700 applications have been submitted. Thanks Greg Stein and Chris DiBona for the hard work."

Comments

"Breakdown"

[Guanix]

Did anyone else think that the project had broken down?

Re:"Breakdown"

[Prophetic_Truth]

i certainly did

Re:"Breakdown"

[cryptoz]

Yep. Sure did, but don't worry, that /. story has a couple more weeks to brew before it can really be posted. You know, the Summer of George broke down pretty quickly, so how long can the Summer of Code last?

Re:"Breakdown"

[michalf]

of course. about 5500 people that were rejected ;-)
it was a sort of lottery. imho if you had an independent project not following the ideas posted earlier by the mentoring organizations - your chances were low - judgin on some discussions after acceptance/rejection of proposals.

mine got rejected ;-)

michal

Re:"Breakdown"

[Mr. Underbridge]

of course. about 5500 people that were rejected ;-) it was a sort of lottery. imho if you had an independent project not following the ideas posted earlier by the mentoring organizations - your chances were low - judgin on some discussions after acceptance/rejection of proposals. mine got rejected ;-)

That was obvious from the first paragraph. In general, following the recommendations for any application process is a good idea. You'll find this theme repeated pretty much everywhere. Reviewers are looking for something. Your best chance is to give it to them.

Re:"Breakdown"

[Praetorian42]

Yeah, for a second I thought maybe the rejection letter I got today was a mistake.

Oh well.

Re:"Breakdown"

[Davin811a]

Does anyone know if Google plans to open source Google File System? That is the one thing I would love to see. I have looked at PVFS and Lustre, but want the redundancy of GFS.

Re:"Breakdown"

[AKAImBatman]

Does anyone know if Google plans to open source Google File System?

Considering the competitive advantage it gives them? Not bloody likely. Besides, it's very much a system that's keyed to Google's usage patterns. Most regular businesses and users would see little to no benefit in Google's system. If you really think it's useful for your line of business, then read the paper (html) and implement a version more specific to your needs. Good luck!

Re:"Breakdown"

[mattyrobinson69]

Depends if you work for yahoo or altavista

Re:"Breakdown"

[mattyrobinson69]

i believe your thinking of the phrase "break it down!", after which people get up and dance like a mo'fo'

Not surprising.

[nonymous Covvard]

Those seem like fairly obvious choices.

psf?

[3770]

Excuse an ignorant Swede, but what the heck is psf, number 4 on the list?

Re:psf?

[Bob Loblaw]

http://www.google.ca/search?q=psf

Python Software Foundation would be my guess since it comes up first in Google.

Re:psf?

[Anonymous Coward]

Python Software Foundation
http://www.python.org//

Re:psf?

[Anonymous Coward]

python software foundation

Re:psf?

[Anonymous Coward]

Project Star Force!

Re:psf?

[Uber Banker]

Panty Source Fetish

Re:psf?

[pegasustonans]

Plaid Slacks Federation

Re:psf?

[dapyx]

Phree Software Foundation

Re:psf?

[Anonymous Coward]

link plz k thnx

$_$

[Anonymous Coward]

Anyone else think it's a little odd to be pushing open source development with cash?

Then again, I wont be complaining if I get an opportunity like this when I study CS next year.

Re:$_$

[michalf]

oh, yes. this is odd. I forgot that open-source developers drink water, eat earth and live in primitive cottages in the center of jungle/desert/iceland. have no other needs than internet access that is free and wear simple self-made clothes.

and yes, this is strange. I always thought open-source software does need any money - it was always here since I remember.

michal

Re:$_$

[inode_buddha]

No, I don't think it's odd at all. Just ask (for example) IBM or Novell. Or RedHat for that matter. Developers need to eat. They also need to pay for their home, car, kids, etc. Companies like Google need relatively cheap access to bleeding-edge tech, and the PR value of a project like this doesn't hurt. Anyway, last I checked "open source" is not mutually exclusive with "cash".

Re:$_$

[MikeFM]

Why? Free software is about freedom and not about money. The fact that free and opensource software is usually available for free is mostly just a happy side effect. Software developers need an income just like everyone else so there is no reason they shouldn't expect to get paid for their work even if it is licensed as opensource.

I think we need a formal non-profit organization to act as a union of sorts for opensource developers. Something similar to what SoC is (but much bigger) and what LinuxFund tried to be. The OSDU could coordinate things like SoC which would take much of the burden off of companies such as Google that might be involved. Also they could request sponsorships from multiple companies which would allow greater amounts of projects to be sponsored. The group could collect extra funds through various programs such as LinuxFund's branded credit card program and through similar programs such as branded license plates (who doesn't want Tux on their license plate?).

OSDU could maintain relationships with mentoring organizations such as those participating in SoC to keep a pulse of what those organizations need and to help shepard programmers to those projects as appropiate. OSDU could pay those programmers out of their funds as well as covering special needs of the programmers or mentors such as providing special hardware or funding needed artwork and documentation. Along with sponsoring programmers that mentors need OSDU could keep tabs on interesting and underfunded projects by monitoring Sourceforge, Freshmeat, and similar community websites. Such projects, when identified , could approach the project leaders to find out how they could best be sponsored.

OSDU would be a super-organization that could tie together many companies and mentoring organizations under a single umbrella to enable the community to better address the needs of these as well as the needs of the programmers actually doing the work.

Now.. how do I bootstrap this. :)

Re:$_$

[commodoresloat]

Anyone else think it's a little odd to be pushing open source development with cash?

Yes, they should really be using bash.

Re:$_$

[danharan]

Odd that we could do something we love and get paid for it?

Yeah, but an odd I could get used to.

13 google

[cryptoz]

Google? Google will have open source code? Is this a first, or am I just nuts? Has anyone else heard that THEY were to be developing open source code, too? I mean, yeah, it's obvious they want to tap into its power with the whole Summer of Code thing, but I didn't realize they were actually going to have code developed for themselves as well...

Re:13 google

[behdad]

Indeed they are opening some of their small projects, moving development to SourceForge.net. See http://code.google.com/ for a list. Small still, but nonzero.

Re:13 google

[gstein]

Actually, the projects aren't being developed "for us". Students who had ideas that didn't match up well with the organizations requested Google for a sponsor. So "our" projects are actually a pretty eclectic sort.

And remember that it will all be Open Source. The SoC students are developing code for *everybody*.

Where are the proposal details?

[Anonymous Coward]

Are the proposal details listed anywhere? I would love to find out if there is any proposal to support yahoo messenger webcam/voice in kopete or gaim.
Dont troll about gaim-vv, it doesnt support voice.

Re:Where are the proposal details?

[Nimrangul]

The proposal information for each project is on their own website, the ones accepted you will not know until Google releases more information, as noted in the link. "We'll post the applications and have further data early next week."

Re:Where are the proposal details?

[fboliv]

My proposal details that got accepted by the Gnome Foundation are available here. Its basically a documentation repository for open source projects, with an API to access them from external applications and websites.

It also fosters the architecture of participation paradigm by making commenting and rating documents as easy as possible.

Re:Where are the proposal details?

[tannhaus]

Well, I can tell you that it's not there for gaim:

http://gaim.sourceforge.net/summerofcode/

KDE

[daviq]

Why isn't KDE higher up the bloody food chain?

So...

[JamesTRexx]

FreeBSD in third place. So who is dead now? :-P

Re:So...

[pegasustonans]

So who is dead now?

Well, everybody over 150 for starters...

Re:So...

[Anonymous Coward]

Do they count as over 150 if they're dead?

Re:So...

[behdad]

That's because there's a lot to do on that front, you know, to catch up... :)

Re:So...

[Anonymous Coward]

Elegy For FreeBSD

I am a *BSD user
and I try hard to be brave
That is a tall order
*BSD's foot is in the grave.

I tap at my toy keyboard
and whistle a happy tune
but keeping happy's so hard,
*BSD died so soon.

Each day I wake and softly sob
Nightfall finds me crying
Not only am I a zit faced slob
but *BSD is dying.

go KDE go...!

[bogaboga]

I am glad KDE is way up there. I am sure no slashdotter can say KDE is doing badly at all. GNOME too, did not do that bad either. We surely have resources in the OSS world.

Re:go KDE go...!

[RPoet]

GNOME got more than KDE anyway - 27 to 24 - if you count gaim too, being a fullblood GNOME application. KDE's similar application, Kopete, had projects in the KDE pool.

Re:go KDE go...!

[mpupu]

Well, gaim works in Windows, XFce and many other enviroments while Kopete doesn't. That could've made a difference.

Re:go KDE go...!

[pherthyl]

No. Gaim is an application that happens to run on Gnome as well as several other platforms/environments. Notice how there is no mention of Gnome on the gaim website.

Kopete on the other hand is an IM client specifically for KDE (although it can be used on Gnome too). It is tightly integrated with the KDE framework and other KDE applications.

Re:go KDE go...!

[bhalo05]

Is really Gaim a Gnome application? I can't see it mentioned here:

http://gaim.sourceforge.net/about.php

Why is that every GTK+ app is always counted as a Gnome app? It's an honest question :-)

Re:go KDE go...!

[Anonymous Coward]

Why is that every GTK+ app is always counted as a Gnome app? It's an honest question :-)

Because otherwise Gnome is missing several important applications, and most GTK+ applications seem to use Gnome HIG standards (more or less), so people just assume they're Gnome applications, rather than general applications that happen to fit best with Gnome.

Re:go KDE go...!

[Anonymous Coward]

Why is that every GTK+ app is always counted as a Gnome app? It's an honest question :-)

Because that is essentially what they really are, especially apps like Inkscape - also on the list - which attempt to follow the Gnome Human Interface Guidelines. The real problem here isn't one of technology, it's terminology. The line between "just GTK" and "fully Gnome" is purposely soft, with most new advances being pushed directly into GTK to get the most usage and buy-in. So it is indeed fair to say that Gnome and Gnomish applications are very well represented in the results.

Re:go KDE go...!

[imr]

KDE: 24
Gnome = 12 gnome + 15 gaim + 14 ubuntu + 13 fedoracore + 16 mono = 70 Gnome.

Re:go KDE go...!

[thinkfat]

Well, I think your equation is only remotely correct.

• GAIM: not exactly a GNOME application. It just happens to use GTK+ and glib
• mono: from what I learned from guadec, the GNOME community seems to be quite divided regarding mono
• ubuntu, fedoracore: lets first see if those bounties are connected to GNOME exclusively. Of course one can argue that FC and ubuntu are primarily GNOME, but you can have KDE with FC and there's kubuntu, too, so every improvement there can also be a win for KDE

But it doesn't really matter anyway. From the outside (news media that are not slashdot or osnews, or newsforge etc), nobody knows about those connections, they will list the topmost three at max, and GNOME is not among them.

Re:go KDE go...!

[imr]

From my trolling point of view, my equation is pure genius, but from any other point of view and especially good sense and good taste, it's pure crap.
But hey, i got moderated as interresting!! So i was probably right!

Dear Michal

[TimeTraveler1884]

mine got rejected ;-)

Dear Michal,

Yeah, sorry about that. Unfortunately your particular proposal did not align with my goals for total world domination. Your submission "The Free Simulator for Coconut-Swallow Aerodynamics" while interesting, would not have been useful as I have already researched this topic throughly and concluded ducks are the optimal fowl for coconut transport.

Sincerly,
Google Inc.

Re:Dear Michal

[Jonner]

Ah, but would that be mallards, or wigeons?

gnome?

[geo.georgi]

I think Apache is on first place because they are the only one with java projects.
But why is gnome so behind KDE?
Does someone have an explanation?

Re:gnome?

[ziggamon2.0]

Actually, GNOME didn't do so bad at all:

GNOME: 12

GNOME/GTK-based applications:
Gaim: 15
Inkscape: 4

Total: 31

To this figure: add Ubuntu and Fedora, two GNOME-oriented distros, and I bet some of their combined 27 Coders will do something that's GNOME-related in some way.

And Apache is on the first place because it makes improvements to Apache are improvements to the web, and the web is what gives Google money. That alone is reason enough to sponsor Apache.

Re:gnome?

[newsblaze]

My guess is that there were more - and more interesting project suggestions for KDE than Gnome. I doubt that Apache's high number has anything to do with Google's use of apache. It will be related to innovation and usefulness. If you look at the lists of suggested projects, ASF probably has the biggest number. Apache projects also have the capability to be useful to more people.

Re:gnome?

[Illissius]

Actually, iirc GNOME actually got significantly more proposals than KDE did. However, note that Gaim was seperate, and got 15 itself, which combined with GNOME's 12 is more than KDE got.

Re:gnome?

[gstein]

The counts were not based on the organizations' utility to Google. Most of it is based on popularity, on the capacity of the org, and some feedback from the orgs.

The ASF ended up with the most because it was the second-most popular, and because they have a LOT of people available for mentoring.

GAIM had the most applications, but were unable to mentor all that many.

Re:gnome?

[uss_valiant]

But why is gnome so behind KDE? Does someone have an explanation?

The number of students per project is more or less proportional to the number of applications a specific project received. Some projects received a lot more applications than others. The only exception is google itself. They have received a lot more applications than the number of their students indicates. I guess they wanted to give (other) F/OSS projects priority.
So, to answer your question. Either KDE is a lot more popular than GNOME amongst students or KDE did a better job attracting students for its Summer of Code tasks.
And, lets not forget that each mentoring project needs human resources to actually mentor the students.

Re:gnome?

[grilo]

Because KDE suffers from featuritis?

Because KDE has a better infrastructure for further development?

Because KDE developers are more fanatic about the project?

I can surely say that at least one GNOME developer submited a proposal for a desktop agnostic enhacement (if you're really interested, you might want to check this).

Seriously, the possibilities are too broad to make a correct statement. But I can point some of the possible reasons:

• People think GNOME is pretty much feature complete, regarding high level enhacement possibilities.
• People are afraid of developing stuff for GNOME due to some issues (real or not) with the platform itself.
• People that use and prefer GNOME have a more broad view of their system, and think GNOME is just another component, instead of their whole computer experience, thus thinking there are other areas that need more attention than their Desktop Environment.

I think the most accurate, though, is that KDE simply has a larger user base that have programming skills. I guess we can't be too far from reality if we establish a relationship between that (attracted people with programming skills) and the proposals present in this "contest".

I'm even tempted to speculate that GNOME (as a Desktop Environment, but certainly not as a development platform) is much more successful than KDE. Even if KDE's userbase is larger, it just means that people that use GNOME are much more oblivious to all this programming stuff, meaning it's prefered by Joe Sixpack.

Of course, all this becomes non-sense when you realize that Linux itself isn't even listed, while FreeBSD is so high (considering that Linux has a much wider user base and many more people contributing to it).

So, two possibilities remain:

1. The ammount of people with spirit of initiative, high coding skills, creative and rethorical (afterall, you got to convince them you are "worthy").
2. People that have their kids starving and really need this to feed them.

So, either KDE attracts super human beings or GNOME attracts only rich bastards to whom 1000 dollars mean nothing.

Re:gnome?

[Anonymous Coward]

Gaim is used by Windows users too. Inkscape is the currently best Open Source vector graphics tool. I use KDE and can't stand GNOME but I use both Gaim and Inkscape. KDE is way more popular than GNOME so it's not surprising it's far in the lead.

Re:gnome?

[Rick and Roll]

I found that the Gnome experience varies by distro. I used Gnome under Gentoo and found it to be a mess, so I switched to KDE. Now I switched to Ubuntu and use Gnome. I had to tell it to use Nautilus instead of the spatial file system, but once that was done I found it purty and reliable and I can use it to get work done.

And Gnome is definitely not a mess on Ubuntu. Far from it.

Re:gnome?

[Muramasa]

If you combine ubuntu and gnome then gnome actually has more projects. This makes sense considering all the ubuntu projects are probably preference apps for the gnome desktop. Most of the mono projects probably have a lot to do with gnome as well.

Re:gnome?

[pantherace]

The reason for this is not Gnome being up to snuff, but that distros will customize Gnome. Gentoo relies on what the project itself provides, and Gnome doesn't provide things very well out of the box. Thus Gnome on Gentoo requires a lot of user-tweaking. KDE provides a much more higher baseline, thus, distros don't need to do as much customization. (Additionally, while Gnome seems to require tailoring, Red Hat's (relatively minor, once actually looked at) changes to KDE created a large outcry.)

So, basically it boils down to KDE being a more centralized, and consistant base, with usually a few custom (config) apps added in. However, Gnome isn't, so there's a lot more parallel effort to get it to the state KDE is in. (You've got to pull in a IM client, media player and lots of other apps which are part of KDE's base (meaning by that the common packages (kdegames, kdepim, kdemultimedia, etc), not just 'kdebase').

I happen to feel that KDE's way is better, but that's my personal opinion.

Re:gnome?

[Robert Douglass]

Apache is an umbrella organization for like a zillion projects. The http server for example. It's not surprising, therefore, that they are largely represented. We'll see the breakdown of apache projects eventually, but I too had noticed that asf and codehaus are the only two Java projects. I think this is more a matter of Java groups not applying on time than any bias for/against them.

Re:gnome?

[thinkfat]

Well, Gaim is not so much of a GNOME application, it's just using GTK+ as a widget toolkit and glib for convenience. Otherwise, no GNOME technologies are used. On the same grounds you could argue that "scribus" is a KDE application because it happens to use Qt.

The documentation tells that it "integrates well with GNOME 2 and KDE 3.1 system tray". So one could argue that it's a KDE application as well :-)

Re:gnome?

[arose]

Gaim also integrates with Nautilus and Evolution. IMHO at least some of those Gaim projects are about Gnome integration.

Re:gnome?

[thinkfat]

So let's just say that GNOME has adopted GAIM as an IM client.

Re:gnome?

[thinkfat]

I think the most accurate, though, is that KDE simply has a larger user base that have programming skills. I guess we can't be too far from reality if we establish a relationship between that (attracted people with programming skills) and the proposals present in this "contest".

I'm even tempted to speculate that GNOME (as a Desktop Environment, but certainly not as a development platform) is much more successful than KDE. Even if KDE's userbase is larger, it just means that people that use GNOME are much more oblivious to all this programming stuff, meaning it's prefered by Joe Sixpack.

Well, if this was true it's a bad thing for GNOME. A project that is unable to attract developers is a dying project. Being at mercy of corporate sponsorship alone to drive development is not freedom anymore. For the sake of GNOME lets hope this is not the case.

Re:gnome?

[cortana]

Good god, rational and reasoned debate about the merits of Gnome and KDE, on Slashdot! I neve thought I'd see the day!

Re:gnome?

[Anonymous Coward]

In this thread, I think that panterace has come closest to what I believe is the actual answer. Without reading the project breakdown, my initial impression is that it is a case of integration.

KDE seems to take a more integrated approach to getting applications under the KDE umbrella with Kthis, Kthat, etc. all of which fall under the "KDE Project" banner.

Gnome on the otherhand, while it may have Gthis and Gthat applications, for better or worse, it does not have the same push to put every little application under the "Gnome Project" banner. Gnome (not surprisingly) seems to take the Gnu philosophy of "here are some tools, do something with them."

While I think that the KDE methodology may produce more cohesive results in the near term; Ultimately, developers prefer being their own bosses; at least when it comes to writting software in their spare time. It also seems to be conducive to the goals of professional software development, as evidenced by the level of gnome development that goes into a goodly number of commercial Linux distributions and other Unicies.

Re:gnome?

[Anonymous Coward]

Gnome on the otherhand (...) does not have the same push to put every little application under the "Gnome Project" banner.

Really not? GNOME nevertheless seems to be happy to officially include Firefox and OO.o as well as many GTK+ apps under their banner...

Re:gnome?

[imr]

Gnome (not surprisingly) seems to take the Gnu philosophy of "here are some tools, do something with them."
This is untrue.
If it had been true, there would have been a functionnal "file selector" in Gnome for years as it is a fundamental tool of any desktop and UI based app.
Instead, it was usable only with the keyboard, defeating its GUI nature.
Then, as a more mature "file selector" was done, it was then impossible to use it with the keyboard without advanced configurations (which the gnome projects claims are bad and therefore remove from being reachable by the front-end user).
From worst to just bad.

The Unix philosophy of having simple efficient tools that are the bricks of more complex ones doesnt relate in a GUI envirronment with "here are some libs, do what you want". It should have been "here are the basic tools any UI will need, do what you need".
In that regard, KDE did better and that's why I and lot of other desktop font-end users have switched wawy from gnome.
Look at konqueror in that aspect, it is the perfect example of a good major Unix tool, the terminal, translated into a GUI envirronment:
it is just a frame where anything can be accomplished.
It is not a browser, nor a file manager.

The wide acceptance of gnome has nothing to see with respecting the Unix philosophy, it is tied to economics, politics and ideology.
Some distros have invested heavily in gnome for ideological reasons when it was sound to do it, then refused to change course when it wasnt necessary anymore for the unreasonable love of anything that doesnt start with a G.

Summer of code in the News

[newsblaze]

Top story at NewsBlaze The Summer of Code for students begins.

Re:Python ! haha

[Anonymous Coward]

I don't get it.

Juice in the Blender

[MindPrison]

Yay! 10 project accepted for Blender! (*claps my little sweaty hands*)

...and NO... This is NOT an opensource kitchen appliance!

Re:Juice in the Blender

[pablodiazgutierrez]

And one of them is mine, yay! :). One summer without ramen noodle overdose is going to be strange...

Re:Juice in the Blender

[jimmypw]

I Love blender :D Viva Open Source

Bah

[Umbral Blot]

My proposal got rejected (I'm not too surprised). The only thing that makes me a bid upset is that it seems all the projects are going to bsd and linux. At least some people should have been allowed to work on mac and windows projects. If you aren't a linux or bsd developer than you would have been looking for java projects, which there wern't many of either.

Re:Bah

[Anonymous Coward]

Stop being a crybaby and port centum to a unix-like OS.

I hear REALbasic works on Linux now!

Re:Bah

[Quantum Jim]

I am not from Google, but my project was accepted for the Summer of Code. My project has nothing to do with BSD or GNU/Linux. It involves a homogenizing the various interfaces for all of the semantic web frameworks around like Jena, Sesame, Redland, and others. Sort of like SAX or DOM for RDF.

Don't be sad that your project wasn't accepted. (What was it tho?) There are lots of opportunities for open source developers. Some alternatives were posted in summer-discuss.

Thank you very much Google and Semedia, by the way.

Re:Bah

[Anonymous Coward]

Hey that's great. There are interfaces for the semantic web. Now if you guys can just get around Godel's theorm, you're good to go.

Re:Bah

[Umbral Blot]

I'm woking on it. Or actually I'm having someone else work on it since I don't use linux.

Cygwin

[ImaLamer]

I've been using a lot of (ported) GPL Windows software lately (Apache2, MySQL, Gaim, so forth) and depend on it greatly. I've been starting to use Cygwin more because I've noticed it has matured a great deal - but more work is always needed.

Can we get Cygwin in on this?

It sure would be nice to have more GNU tools for Windows.

Re: This isn't just "developers", either...

[symbolic]

This is an educational push, if I understand it correctly, because one of the requirements is that you have to be a student in order to participate. Feel free to correct me if I am mistaken, but I seem to remember seeing some complaints by more seasoned developers (non-students) that this limitation seemed somewhat arbitrary.

Too geeky for my acceptance :P

[AigariusDebian]

I am accepted and I am too geeky to celebrate that as it should have been celebrated. Doh!

Summer of Lisp

[heowbert]

Being an mentoring organization, LispNYC has received a lot of feedback regarding the summer of code.

Because of the overwhelming popularity and financial support of Google's Summer of Code, LispNYC is proud to announce it's continued dedication to same great idea: Summer of Lisp.

The reality is that although many of our projects are worthy endeavors, not all will get funding from Google, which is why our mentors are willing to work unfunded with students who are willing to learn Lisp. So if you're serious about computer science and are interested in continuing your education in order to beat the averages ( http://www.paulgraham.com/avg.html ) while working on projects that matter, please stop by.

http://www.lispnyc.org/summeroflisp.html

Re:Summer of Lisp

[Rick and Roll]

That sounds great! There is a lot of need among C. S. people for good mentors.

Definitely something to think about.

Re:Google bubble is about to burst

[Darth Cow]

"They are loading up with PHds and scientists. If these guys had any useful brains, they would be super rich already."

Excuse me? Maybe PhDs and scientists are smart enough to realize that there is more to life than just money, such as perhaps having an interesting and fun job?

There are many different types of intelligence, and the most capable scientists and engineers may likely not have the best business sense. That doesn't meant that Google doesn't have people with good business sense, too, though. Perhaps a little division of labor?!? (Microsoft does that, too.)

Re:gnome morons

[fboliv]

Well you're quite naturally a dumbass, but allow me to challenge you into doing one of the documentation or memory optimization projects yourself then. Put your fingers where your mouth is (pun intended).

After all, desktop management elitism is so 1999.

total application count

[Anonymous Coward]

The breakdown is a combination of two factors: what each project could handle, and how many people applied for it. Here's the breakdown of total applications:
997 google
859 gaim
698 asf
460 gnome
421 kde
373 other
338 mono
319 freebsd
305 psf
260 gallery
236 ubuntu
233 nmap
216 oo
203 fedoracore
180 wine
178 mambo
169 mozdev
162 codehaus
143 lj
139 blender
134 drupal
124 internet2
123 plg
121 jabber
117 lispnyc
113 winlibre
108 netbsd
104 svn
93 jxta
92 tpl
88 handhelds
76 semedia
74 samba
73 xwiki
62 psu
57 horde
55 inkscape
49 oscar
48 bricolage
48 asterisk
47 ohiolink
39 monotone

MOD UP PLEASE

[Rick and Roll]

That is funny. I hope somebody is archiving the genius /. comments for posterity.

Re:Google bubble is about to burst

[be-fan]

They are loading up with PHds and scientists. If these guys had any useful brains, they would be super rich already. At least they'll be wasting private funds and not public funds. Doesn't anyone think Google needs more Microsoft-type people...people with tech and biz smarts ?

Without Google-type people, who would Microsoft-type people copy? If you doubt this, I dare you to name a single Microsoft project that isn't ultimately copied from something a PHD or scientist dreamed up. Even their "research" projects tend to be of the applied type, rather than the "new ideas" type.

Re:Google bubble is about to burst

[ClosedSource]

"I dare you to name a single Microsoft project that isn't ultimately copied from something a PHD or scientist dreamed up."

Given that the vast majority of tech ideas and implementations weren't dreamed up by either PHD's or scientists, it shouldn't be too hard.

Re:Google bubble is about to burst

[ma_luen]

Seriously, check out microsoft research. They do a lot of "real research". They have some very smart people and they publish a fair amount of quality work in prestigious conferences.

I know in my area of program analysis for reliability/safety that there have been some very subtantial papers (with "new ideas") on predicate abstraction that involve people from msr. It is some cool work and it looks like it will have a real impact on driver/low level code quality.

Mark

Re:Python ! haha

[Anonymous Coward]

You had to be there, really.

Google uses a lot of Python

[cpghost]

Yes, that should be it. According to Peter Norvig (of "Paradigms of Artificial Intelligence Programming: Case Studies in Common Lisp" fame):

"Python has been an important part of Google since the beginning, and remains so as the system grows and evolves. Today dozens of Google engineers use Python, and we're looking for more people with skills in this language." said Peter Norvig, director of search quality at Google, Inc. (Source).

Google-type?

[pedantic bore]

Gosh, I guess there weren't any research organizations before Google came along and put a bunch of smart people in cubicles together?

Google-type? I'd rather be PARC-type, or DEC-SRC-type. I'd settle for MSR-type, BBN-type, SRI-type, SunLabs-type, or IntelLabs-type, or any-decent-university-type.

Re:Google-type?

[be-fan]

You missed my point completely. The original-poster was complaining that Google was filled with PHDs and scientists, and suggested they should hire Microsoft-type people instead. The statement suggests that Microsoft type people are not PHDs and scientists, but rather, applied folks with "tech and business smarts". My point was that without PHDs and scientists (Google-type people, using the poster's terminology), where would the applications people (Microsoft-type people, using the poster's terminology), get their ideas?

Re:Google-type?

[pedantic bore]

I'm sorry I didn't express myself more clearly; I don't mean to start a flame-war.

Let's look at this another way. The "average" person who works for Google isn't technical at all. They spend their time managing Google's ad-based revenue. They are primarily salespeople. The last time I spoke to someone at Google, their estimate was that less than 15% of the employees at Google are involved in the technical side of Google -- and they thought perhaps it was a lot less.

The average person at Microsoft is a code monkey. It's fairly astonishing how small of a sales force they have. (one of the advantages of being a monopoly, of course).

Now, on to the second point: it pushes my button to hear "Google-type" used as a synonym for PHDs and scientists, the best and the brightest, etc. They hire good people who are great at turning ideas into programs, but they're not particularly inventive when it comes to those ideas. They grab ideas that other people have invented and build them into businesses. That's about as Microsoft-like as you can get.

Re:Google-type?

[be-fan]

I'm not the one who characterized Google employees as PHDs and scientists, the original poster was. I was simply using his terminology, to defend phds and scientists in general.

For all you rejects

[Anonymous Coward]

You can check out the list of rejected ideas, or add your own here:

http://summerofrejects.pbwiki.com/index.php?wiki=R ejectedIdeas

and besides that, we have a irc support / collabration channel on :

irc://irc.freenode.net/google-rejects

check out the wiki as well:
http://summerofrejects.pbwiki.com/

the point here is to help those who got rejected work on their projects together.

Students only get $3150 from original $4500

[astrab]

"Students will receive only $3150 from the original $4500. According to Pedro Macedo, US government takes 30% of the grant via taxes."

http://google-blog.dirson.com/post.new/0265/

IN CAPATILIST AMERICA, GOVERNMENT TAXES YOU!

[Anonymous Coward]

"According to Pedro Macedo, US government takes 30% of the grant via taxes."

No, really? Film at 11!

summer what'd mean summer it's winter

[Grizzlysmit]

well here in the southern hemisphere it is. :-P

seriously way to go google this is a good(TM) thing to do.

Re:Google bubble is about to burst

[be-fan]

In retrospect I think my characterization wasn't entirely fair. I based my judgement on the papers from MSR that I'd seen in my field of interest (compiler optimization), and they had seemed to me of the applied type rather than the theoretical type. Even looking through MSR's website, a lot of the high-profile stuff (eg: Singularity) seems a bit derivative. On the other hand, there does seem to be some genuinely new stuff in some of the research areas.

How Valuable Is It?

[rchoetzlein]

Ok, so I'm going to develop free software this summer and get $4500 for it. That seems like a good idea (and getting paid to create open source is unusual, but should be the norm).

But.. When Google sees all these cool ideas, grabs a few, rewrites them from scratch, and their market value goes up by several million -- i'm going to feel a bit differently about that $4500 i'm holding. Hey, if I knew it was _that_ valuable... As far as I remember it is the expressed work, the code, that is copyright (yes, in open source licenses too), not ideas. So nothing prevents Google from doing this. I'm all for giving away ideas to the community, but Google seems to gain a lot more than I, the originator of the idea, does. Ah, the American Dream: "Where any little guy can rise up and give away a good idea to a big company."

Ok, perhaps I'm paranoid, but when was the last time you checked how valuable _your_ idea was? Any comments? Does anyone think this is an issue?

c.c.

Re:How Valuable Is It?

[Nimrangul]

Uh, why would Google rewrite an HFS+ implentation on NetBSD from scratch? And why would that increase it's market value?

You're paranoid, no doubt about it. A bit delusional as well. The BSD projects they are sponsoring don't need to be rewritten, but why would any of them be of use to Google for making money? I cannot see how most of them would.

Let's run down NetBSD's Summer of Code projects, shall we?

tmpffs' idea is a more efficient way to do memfs, or at least it is in theory; fewer copies, disable buffer cache and try to use as little memory as possible.

wcurses' wide-character support is useful for multiple languages that use characters that are outside the normal ascii character set. Pretty much anything that doesn't use Roman letters; Russian, Japanese, Korean, etc... So, if you're not a native English speaker this could be handy.

bpg's just like gnupgp, only BSD licensed. - This one may actually be useful, adding pgp to Gmail.

zeroconf would be a daemon that automatically assigns IPv4 link-level addresses and probably a Multicast DNS responder and a BSD-licensed Multicast DNS library.

regress is an automated testing the system.

userfs would probably be for allowing unprivileged users to manipulate parts of the filesystem.

ndis is an attempt at a standard nic api, developed by 3com and Microsoft, though Windows have their own "extended" versions.

hfs+ would be that filesystem Apple likes so much - the Hierarchical File System.

So, only one thing has really any use to Google and they don't even need to rewrite it. This is only out of the NetBSD projects, but I am sure the FreeBSD ones will be much more profitable.

Re:How Valuable Is It?

[srowen]

Well, the thing is that an "idea" isn't worth anything by itself; its application in the market is. The internal combustion engine didn't make Ford wealthy; Ford Motor Company did. Ideas are cheap; there are tons of them. Execution on ideas is the important thing.

You might have a great idea, which you you don't have the time or resources to develop. Unless something changes, it's worth $0 to you and the world doesn't benefit from it because it is never implemented.

Google pays you $4500 to compensate you and give you time to develop your idea. Now, your idea is worth $4500 to you at least. The world benefits from it.

I'm not sure what the terms are, but, on the off chance that one of these ideas greatly benefits Google, I'm not sure you've lost anything (unless you were planning to develop the idea yourself). Remember that again Google has spent and will have to spend a lot to develop any idea -- productionizing and maintaining any service costs money. They've invested in developing a good idea and that's paid off.

It's really not as if you had a $1MM check in your head in the form of an idea and they've tricked you into selling it for $4500.

Now on the other hand, if you think it's such a great idea, so much that you think the expected value of trying to develop it on your own is more than $4500, by all means, don't enter it into the Summer of Code!

Re:How Valuable Is It?

[jimmypw]

Welcome to the world of IP rights. Nice idea until somone steals it. Its never the person thats has the idea who reaps the rewards but the person to write it down and had £30,000 to spare.

But please dont give google a hard time as they haven't done anything wrong yet. But they have done somthing Fantastically right. Open source needs a very loud voice... Google is louder than that. And frankly i dont mind them making money off a competition like this of it means at the end of it we will have a catalogue of the best software that you can get that has been donated by the community. As long as they dont backstab anybody like you said.

Drupal - Wow!

[sien]

It's really surprising and impressive that drupal has so many projects in the list. Kudos to the folks who are working on it, they obviously have their stuff together.