75% of Linux Code Written by Paid Developers
Linked by Thom Holwerda on Sun 24th Jan 2010 16:22 UTC, submitted by Dale Smoker
Linux LWN.net founder and kernel contributor Jonathan Corbet offered an analysis of the code contributed to the Linux kernel between December 24 2008 and January 10 2010. 18% of contributions were made without a specific corporate affiliation, 7% weren't classified, and 75% were from people working for specific companies in roles where developing that code was a major requirement. "75% of the code comes from people paid to do it," Corbet said.
E-mail Print r 4   52 Comment(s) http://osne.ws/hky
Thread beginning with comment 405699
To read all comments associated with this story, please click here.
Well
by Xaero_Vincent on Sun 24th Jan 2010 16:52 UTC
Xaero_Vincent
Member since:
2006-08-18

The Linux kernel has become so big and complicated that only those paid have much incentive to work full time on the project. In addition, there are strict requirement before code is accepted into the mainline kernel. Not just any casual hacker will get their contributions accepted.

Distribution specific, non-vanilla kernels likely have many more kernel patches and driver contributions from non-paid community developers; its just these aren't accepted in the mainline kernel and thus aren't counted.

Edited 2010-01-24 16:53 UTC

Reply Score: 3

RE: Well
by Zifre on Mon 25th Jan 2010 00:09 in reply to "Well"
Zifre Member since:
2009-10-04

The Linux kernel has become so big and complicated that only those paid have much incentive to work full time on the project.

This is true. Almost nobody can work full time on something without getting paid for it.

In addition, there are strict requirement before code is accepted into the mainline kernel. Not just any casual hacker will get their contributions accepted.

Any random person can get a patch in the Linux kernel if it does something important (in fact, I am working on a patch to the kernel right now; I don't know if it will be accepted though). I don't think that there is any bias toward accepting code from corporations. However, corporations often have more experience and thus write better code than some random hacker on their first try.

Distribution specific, non-vanilla kernels likely have many more kernel patches and driver contributions from non-paid community developers; its just these aren't accepted in the mainline kernel and thus aren't counted.

I don't think this is true. The amount of code in the patches that most distributions apply is probably not even 1% of the total kernel code. And, for example, I'm guessing that many of Ubuntu's kernel patches are written by Canonical, and many of Red Hat's kernel patches are written by Red Hat.

Reply Parent Score: 2