Re: [gentoo-project] Re: [gentoo-core] Re: Poll: Would you sign a Contributer License Agreement?

[M. J. Everitt]

On 25/06/18 15:37, Ulrich Mueller wrote:
>>>>>> On Mon, 25 Jun 2018, Greg KH wrote:
>> And I'm dragging this back to -core, as I'm not on -project, so my
>> responses are not even going there, and you started this on -core.
> Nope, I started the thread on -project on 2018-05-30:
> https://archives.gentoo.org/gentoo-project/message/b1d92fc4275c15a052cf27bb2a5d75dd
> I cross-posted to -core once (on 2018-06-04) for wider audience,
> because I had received only a handful of replies by then.
>
> Otherwise, there is no reason why this discussion should take place in
> private, so it is off-topic in -core.
>
>>> With the license currently listed at https://developercertificate.org/
>>> ("changing is not allowed") nobody would even be allowed to commit
>>> the DCO to a repository under it's own terms. Catch-22.
>> And as the Debian developers said, "that's crazy-talk, don't worry
>> about it."  Seriously, don't.
> If anyone worries about non-free files in their repositories, then
> it's Debian. Certainly much more than we do.
>
> Also, encouraging people to falsely certify things (and "don't worry
> about it") is exactly what we want to avoid. If there is a S-o-b line
> included with a commit, then there must not be any doubt that this
> commit conforms to the wording of the certificate. If we allow people
> to commit non-free files and certify them under the Linux DCO 1.1 then
> the whole exercise is useless.
>
>> And if you do have a lawyer who is worried about such a thing,
>> please let me talk to them and I'll be glad to put them in contact
>> with loads of other lawyers who will be glad to discuss it.
>> What company or legal entity has concern with the DCO as-written?
> Everybody who wants to commit a license file to the Gentoo repository,
> and with the DCO 1.1 would have to lie about its status?
>
>> That's not the only thing that you have changed here, as you state.
>> You changed the wording of the types of licenses (hint, "free
>> software" is not the same as "open source" and has consequences by
>> changing that wording.)
> It is generally acknowledged that "open source" licenses and "free
> software licenses" are mostly congruent. (There are very few OSI
> approved licenses like Artistic 1.0 which the FSF classifies as
> non-free. The other way around, I am not aware of any.)
>
> Nevertheless, I don't have a strong opinion here. Our Social Contract
> says "free software", so we changed it to that for consistency, but
> replacement of the term alone wouldn't be a sufficient reason to
> create a modified version.
>
>>> Do you think that anybody would have difficulties understanding
>>> this? Then please propose a better wording.
>> I am saying, over and over and over, that it's not up to me to
>> change the wording.  I want _you_ to justify the change by getting a
>> solid legal opinion that what you are changing actually does what
>> you think it does, and is even needed in the first place.
>> Again, don't try to arm-chair legal issues.  That ends up causing
>> many more problems than you can ever imagine.  There's a good reason
>> that lawyers write licenses and legal texts as they understand
>> things that are not obvious to non-legally-trained people.
> (Sometimes I wonder how some people survive. Do they ask their lawyers
> before passing a green traffic light? Or before agreeing to a contract
> of sale in the grocery store? :-)
>
>> And again, you are ignoring the fact that we all are now going to
>> have to get the legal departments of our companies to evaluate this.
>> That will NOT take just 1 minute.  If you use the DCO as-is, that
>> would only take 1 minute.
> How about the following change then:
>
> --- a/glep-0076.rst
> +++ b/glep-0076.rst
> @@ -133,12 +133,17 @@ with the project's license.
>  For commits made using a VCS, the committer shall certify agreement
>  to the Gentoo DCO by adding ``Signed-off-by: Name <e-mail>`` to the
>  commit message as a separate line.  Committers must use their real
>  name, i.e., the name that would appear in an official document like
>  a passport.
>
> +As an alternative to the above, commits may be certified with the
> +Linux Kernel DCO 1.1.  Committers shall clearly indicate this by
> +adding ``(Linux DCO 1.1)`` at the end of the ``Signed-off-by`` line.
> +Using the Gentoo DCO is strongly preferred, though.
> +
>  The following is the current Gentoo DCO::
>
>      Gentoo Developer's Certificate of Origin, revision 1
>
>      By making a contribution to this project, I certify that:
>
>
> It would allow anyone who has issues with our modified version to
> commit under the original Linux DCO instead. Of course, certain files
> they couldn't commit then.
>
> Ulrich
I make a simple observation based on the thread here. I would personally
probably be more comfortable under a DCO that has "large organisation"
backing eg. Linux kernel, as the effort required to make changes is
likely to be significant, and it is likely to have been vetted by
"qualified persons". By contrast, Gentoo is likely to have been cobbled
together by a consensus of unqualified persons, and is quite unlikely to
be defended in court, -should- it come to that (see recent legal case of
McHardy et al).

Not that I have any issue with Gentoo having it's own, but it lacks
teeth and claws.

-----
another 2c from the peanut gallery. Apologies.