Do you prefer a CLA or Developer Certificate Of Origin?
- Inicie sesión ou rexístrese para enviar comentarios
When contributing to a FLOSS project, some companies/organizations require a Contributor License Agreement (CLA) that is legally binding and gives you the rights to contribute code but also at the same time allowing the copyright holder to take your code and take complete credit for it while passing liability to you.
If they want, they can re-license your code to a non-free one without your control as they own your work. Canonical is known to be shady with their CLAs as they leverage the GPL to force you to use free software but then change it to non-free code since they have all the power to do so as the original copyright holder. Some say that Upstart may have been more popular and systemd would have never existed if Canonical was better with contributed code and them being less draconian with forcing their own copyrights.
On the other hand, projects like Docker and the Linux kernel use the Developer Certificate of Origin approach (https://blog.docker.com/2014/01/docker-code-contributions-require-developer-certificate-of-origin/) where you basically accept the certificate one time and don't have to go through the legal approach that a CLA may impose. It is simple and out of the way, but many FLOSS companies don't do this approach so it isn't that popular.
So from your experiences, what is your preferred approach?
Preferred approach: stop trolling already t3g. GTFO to the Mir's dev mailing list and raise philosophy thread there.
I don't give a damn about Docker shit and *buntu "Snappy Core".
It's a trap into exfiltrated data silo.
I simply asked if you guys had to work with a CLA or Developer Certificate of Origin in the past and had some thoughts.
Btw, who the fuck are you?
I am a translator!
t3g wrote:
> If they want, they can re-license your code to a non-free one without
> your control as they own your work. Canonical is known to be shady
> with their CLAs as they leverage the GPL to force you to use free
> software but then change it to non-free code since they have all the
> power to do so as the original copyright holder. Some say that
> Upstart may have been more popular and systemd would have never
> existed if Canonical was better with contributed code and them being
> less draconian with forcing their own copyrights.
If Canonical licenses the software under the GPL then it doesn't matter
too much if they also license it under a non-free software license.
Here is the FSF advice on assigning copyright (CLA):
http://www.gnu.org/philosophy/assigning-copyright.html
and the obligatory RMS philosophy article on "selling exceptions":
http://www.gnu.org/philosophy/selling-exceptions.html
I think there are advantages and disadvantages either way, but having
code assigned to an entity willing to defend free software seems like a
good idea, but in practice might be more trouble than it's worth.
Andrew
Thanks for the response Andrew. I read up on those links prior to this original post and got some background info on what a CLA was and how some companies are using it.
So back to the original topic... which one do you prefer if you had a choice? This applies to your own code and ones you contribute to.
- Inicie sesión ou rexístrese para enviar comentarios