How many Developers on the Linux kernel?
-
We were wondering in the office (in comparison to vista): How many people hacking on the linux kernel, and on linux desktops? Anybody knows a link with numbers?
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify! || Fold With Us! || sighistHeh, I'm sure this is not the right forum for this. :rolleyes:
-- Mit viel Oktan und frei von Blei, eine Kraftstoff wie Benziiiiiiin!
-
We were wondering in the office (in comparison to vista): How many people hacking on the linux kernel, and on linux desktops? Anybody knows a link with numbers?
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify! || Fold With Us! || sighistI think the kernel has hundreds of developers (I'm thinking I've heard a number somewhere in the 200-300 range) partially due to volunteers and partially due to companies like Novell, IBM, et al contributing. For the other stuff it's next to impossible to know, since the software "base" is different for every distro, and there's no uniform architecture or plan to the set of libraries that come with a distro. You've got GTK, KDE, Gimp, OpenOffice, Beagle, GConf, DBUS, GStreamer, ARTS, JACK, ALSA, blah, blah, blah, ya da ya da ya da. Just the ones I mentioned entails 100-200 programmers.
¡El diablo está en mis pantalones! ¡Mire, mire! Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)! SELECT * FROM User WHERE Clue > 0 0 rows returned Save an Orange - Use the VCF!
-
I think the kernel has hundreds of developers (I'm thinking I've heard a number somewhere in the 200-300 range) partially due to volunteers and partially due to companies like Novell, IBM, et al contributing. For the other stuff it's next to impossible to know, since the software "base" is different for every distro, and there's no uniform architecture or plan to the set of libraries that come with a distro. You've got GTK, KDE, Gimp, OpenOffice, Beagle, GConf, DBUS, GStreamer, ARTS, JACK, ALSA, blah, blah, blah, ya da ya da ya da. Just the ones I mentioned entails 100-200 programmers.
¡El diablo está en mis pantalones! ¡Mire, mire! Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)! SELECT * FROM User WHERE Clue > 0 0 rows returned Save an Orange - Use the VCF!
Perhaps I misunderstood, but what do GTK, KDE, GIMP, OO, Beagle, GConf, etc have to do with the Linux kernel? The original question is a tough one to get accurate metrics on. Technically anyone who has had code accepted into the mainline kernel tree is a "developer of the Linux Kernel" myself included. However most of us "kernel developers" would be the first to admit that we shouldn't really be included in those metrics due to our very limited contributions. I have had 3 device drivers accepted into the mainline kernel. Every one of them was for a custom, 1 of 1, high speed DAQ developed for a customer. I am the only person that will ever get any use out of the kernel modules I submitted into the kernel tree, but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel. The benefits for doing so are great. When the next major update to the device subsystem breaks compatibility with the current kernel api, those nice guys (and gals) over at kernel.org will update my drivers for me to ensure that when someone downloads the new kernel source all of the modules compile. I know I went off on a bit of a tangent, but I wanted to illustrate my point that literally speaking there are probably thousands of kernel developers, but in reality probably fewer than 100 that are truly active in its day to day development.
-
Perhaps I misunderstood, but what do GTK, KDE, GIMP, OO, Beagle, GConf, etc have to do with the Linux kernel? The original question is a tough one to get accurate metrics on. Technically anyone who has had code accepted into the mainline kernel tree is a "developer of the Linux Kernel" myself included. However most of us "kernel developers" would be the first to admit that we shouldn't really be included in those metrics due to our very limited contributions. I have had 3 device drivers accepted into the mainline kernel. Every one of them was for a custom, 1 of 1, high speed DAQ developed for a customer. I am the only person that will ever get any use out of the kernel modules I submitted into the kernel tree, but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel. The benefits for doing so are great. When the next major update to the device subsystem breaks compatibility with the current kernel api, those nice guys (and gals) over at kernel.org will update my drivers for me to ensure that when someone downloads the new kernel source all of the modules compile. I know I went off on a bit of a tangent, but I wanted to illustrate my point that literally speaking there are probably thousands of kernel developers, but in reality probably fewer than 100 that are truly active in its day to day development.
"but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel." I didn't think it mattered whether or not you were paid, but that simply linking with GPL code required your code to then also be GPL. Granted, licensing your code as GPL I think just requires you to make the code available and perhaps distribute the code with the product, and wouldn't necessarily require you to submit it to any particular group. So maybe you're referring to an additional restriction. Anyway, sorry for the off-topic question, but that stood out to me.
-
We were wondering in the office (in comparison to vista): How many people hacking on the linux kernel, and on linux desktops? Anybody knows a link with numbers?
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify! || Fold With Us! || sighistpeterchen wrote:
How many people hacking on the linux kernel, and on linux desktops?
This is difficult to quantify for several reasons: 1) numbers will include "free time developers" (as in they only work during their "off hours" from a regular job) 2) You have higher turn over in open source (not everyone can afford to donate) 3) You have companies turning over old code (SGI, IBM, Sun all transferred volumes of code to Linux) 4) You have companies who's business is something else (like Sun servers, SGI graphics), who tell their employees if they find a problem with Linux to fix it and send it in. This means that no matter how you count it you will probably always find "quantifiable" numbers higher in open-source, especially something as large as Linux. That does not necessarily mean they are in any way comparable numbers.
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Perhaps I misunderstood, but what do GTK, KDE, GIMP, OO, Beagle, GConf, etc have to do with the Linux kernel? The original question is a tough one to get accurate metrics on. Technically anyone who has had code accepted into the mainline kernel tree is a "developer of the Linux Kernel" myself included. However most of us "kernel developers" would be the first to admit that we shouldn't really be included in those metrics due to our very limited contributions. I have had 3 device drivers accepted into the mainline kernel. Every one of them was for a custom, 1 of 1, high speed DAQ developed for a customer. I am the only person that will ever get any use out of the kernel modules I submitted into the kernel tree, but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel. The benefits for doing so are great. When the next major update to the device subsystem breaks compatibility with the current kernel api, those nice guys (and gals) over at kernel.org will update my drivers for me to ensure that when someone downloads the new kernel source all of the modules compile. I know I went off on a bit of a tangent, but I wanted to illustrate my point that literally speaking there are probably thousands of kernel developers, but in reality probably fewer than 100 that are truly active in its day to day development.
Peter asked: "How many people hacking on the linux kernel, and on linux desktops?" So I took that to mean the various desktop components, like X, toolkits, productivity suites, etc.
¡El diablo está en mis pantalones! ¡Mire, mire! Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)! SELECT * FROM User WHERE Clue > 0 0 rows returned Save an Orange - Use the VCF!
-
We were wondering in the office (in comparison to vista): How many people hacking on the linux kernel, and on linux desktops? Anybody knows a link with numbers?
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify! || Fold With Us! || sighistpeterchen wrote:
How many people hacking on the linux kernel
42 Well someone had to say it. :) best place to locate that info would be http://www.kernel.org/[^]
I'd love to help, but unfortunatley I have prior commitments monitoring the length of my grass. :Andrew Bleakley:
-
"but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel." I didn't think it mattered whether or not you were paid, but that simply linking with GPL code required your code to then also be GPL. Granted, licensing your code as GPL I think just requires you to make the code available and perhaps distribute the code with the product, and wouldn't necessarily require you to submit it to any particular group. So maybe you're referring to an additional restriction. Anyway, sorry for the off-topic question, but that stood out to me.
The way you described it is probably more accurate. I guess what I meant was that if I were to write a device driver to control a robot I was building then I would not be legally required to submit the source code for that to the kernel until I chose to distribute my derived work publicly. My only experience with kernel development has been as a paid contractor so my comment, incorrectly, was based only from that viewpoint. Thanks for the correction.
-
Peter asked: "How many people hacking on the linux kernel, and on linux desktops?" So I took that to mean the various desktop components, like X, toolkits, productivity suites, etc.
¡El diablo está en mis pantalones! ¡Mire, mire! Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)! SELECT * FROM User WHERE Clue > 0 0 rows returned Save an Orange - Use the VCF!
Gottcha. I read the topic and skipped over the desktops part. Sorry to have taken a critical tone in my response. I think that there is a common misconception as to what exactly Linux is, and I am generally quick to try to correct that misconception by pointing out that Linux is just a kernel. Everything else is just software built on top of it. Richard Stallman (GNU) doesn't get the credit he deserves :)
-
peterchen wrote:
How many people hacking on the linux kernel, and on linux desktops?
This is difficult to quantify for several reasons: 1) numbers will include "free time developers" (as in they only work during their "off hours" from a regular job) 2) You have higher turn over in open source (not everyone can afford to donate) 3) You have companies turning over old code (SGI, IBM, Sun all transferred volumes of code to Linux) 4) You have companies who's business is something else (like Sun servers, SGI graphics), who tell their employees if they find a problem with Linux to fix it and send it in. This means that no matter how you count it you will probably always find "quantifiable" numbers higher in open-source, especially something as large as Linux. That does not necessarily mean they are in any way comparable numbers.
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
As you are alluding to, a much more interesting statistic would be the number of man hours spent on Linux development per week as compared to Vista. It's too bad there is no way to get such a statistic with any degree of accuracy.
Scott Lee wrote:
a much more interesting statistic would be the number of man hours spent on Linux development per week as compared to Vista. It's too bad there is no way to get such a statistic with any degree of accuracy.
It would be, but for much the same reasons it is even harder to quantify. Linux has taken a large jump toward marketable/installable capability due to the donations of code. Those would actually be the only code pieces that would be quantifiable, but it would be tied up in legal issues trying to get it. The SMP capability is primarily code from SGI, but the time it took to adapt it was larger than expected. Do you use the time it took to adapt the code to Linux, or the time to develop the SGI SMP code, or both, or some portion therein?? The SGI code was all developed for Irix, their own OS, not Linux, so adaption of file system, SMP and other code sets were not always easy. In the end, you can't really compare the hours either. The question is in code maintenance hours. Which is easier to work with, which is easier to fix? That would be an interesting statistic. If Linux is more difficult to adapt new code and changes to, then it will never catch up to Windows. If Windows is more difficult to adapt changes to then development will eventually slow and Linux has a "chance" to catch up.
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Perhaps I misunderstood, but what do GTK, KDE, GIMP, OO, Beagle, GConf, etc have to do with the Linux kernel? The original question is a tough one to get accurate metrics on. Technically anyone who has had code accepted into the mainline kernel tree is a "developer of the Linux Kernel" myself included. However most of us "kernel developers" would be the first to admit that we shouldn't really be included in those metrics due to our very limited contributions. I have had 3 device drivers accepted into the mainline kernel. Every one of them was for a custom, 1 of 1, high speed DAQ developed for a customer. I am the only person that will ever get any use out of the kernel modules I submitted into the kernel tree, but legally if you get paid to write code that links against the Linux kernel you are obligated to submit it for inclusion in the kernel. The benefits for doing so are great. When the next major update to the device subsystem breaks compatibility with the current kernel api, those nice guys (and gals) over at kernel.org will update my drivers for me to ensure that when someone downloads the new kernel source all of the modules compile. I know I went off on a bit of a tangent, but I wanted to illustrate my point that literally speaking there are probably thousands of kernel developers, but in reality probably fewer than 100 that are truly active in its day to day development.
Scott Lee wrote:
etc have to do with the Linux kernel?
because my question included "..and on linux desktops"?
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify! || Fold With Us! || sighist -
I think the kernel has hundreds of developers (I'm thinking I've heard a number somewhere in the 200-300 range) partially due to volunteers and partially due to companies like Novell, IBM, et al contributing. For the other stuff it's next to impossible to know, since the software "base" is different for every distro, and there's no uniform architecture or plan to the set of libraries that come with a distro. You've got GTK, KDE, Gimp, OpenOffice, Beagle, GConf, DBUS, GStreamer, ARTS, JACK, ALSA, blah, blah, blah, ya da ya da ya da. Just the ones I mentioned entails 100-200 programmers.
¡El diablo está en mis pantalones! ¡Mire, mire! Real Mentats use only 100% pure, unfooled around with Sapho Juice(tm)! SELECT * FROM User WHERE Clue > 0 0 rows returned Save an Orange - Use the VCF!
That's what makes Linux so appealling in a commercial environment. Operating system design by circle jerk. X|
Software Zen:
delete this;