contributor license agreement (CLA)

[Brad Bishop]

On Wed, 2018-02-14 at 15:51 +0100, Paul Menzel wrote:
> Dear Brad,
> 
> 
> Thank you for your response.

No problem!

> 
> > I am not a lawyer but my understanding is that it protects the
> > users
> > of OpenBMC from a contribution to the project by a contributor that
> > did not have the necessary grants to make the submission in the
> > first place.  I found this page to provide a good layman's overview
> > of the benefits: http://oss-watch.ac.uk/resources/cla
> 
> Thank you for the URL, but, unfortunately, it doesn’t answer the 
> question what the advantage over the Linux kernel [1] procedure is.

>From the article:

"The purpose of a CLA is to ensure that the guardian of a project’s
outputs has the necessary ownership or grants of rights over all
contributions to allow them to distribute under the chosen license."

Again I am not a lawyer but based on my limited understanding, I don't
believe projects that only utilize a DCO (like Linux) can make these
assurances to their users.  This is the advantage - rather than
deferring this risk to the users of the project, we provide software
free of these concerns from the very start.

> 
> Brad, who decided, that CLA are a requirement? To my knowledge, CLAs
> are 
> only needed, if you think about changing the license in the future. 
> Besides that, there is *no* advantage, and the Linux Kernel
> procedure 
> should be used. 

I think the evidence suggests otherwise.  There are many, many OSS
projects out there - many highly successful, that require a CLA.
Including Linux Foundation projects.  If there were no advantage at all
then this would not be the case.

> It just scares off people wanting to contribute small 
> fixes and improvements.

I concede the point, but as with any decision it is about cost/benefit 
- the cost being the point you have raised here, the benefit being
providing our users with software guaranteed to be free of any legal
infringement.