[Mesa-dev] [PATCH] docs: Document the optional usage of Signed-off-by

[Matt Turner]

On Tue, Nov 27, 2018 at 7:26 PM Jordan Justen <jordan.l.justen at intel.com> wrote:
>
> On 2018-11-27 18:04:17, Matt Turner wrote:
> > On Tue, Nov 27, 2018 at 6:00 PM Jordan Justen <jordan.l.justen at intel.com> wrote:
> > >
> > > On 2018-11-27 17:17:15, Matt Turner wrote:
> > > > On Tue, Nov 27, 2018 at 5:13 PM Jordan Justen <jordan.l.justen at intel.com> wrote:
> > > > >
> > > > > This adds the "Developer's Certificate of Origin 1.1" from the Linux
> > > > > kernel. It indicates that by using Signed-off-by you are certifying
> > > > > that your patch meets the DCO 1.1 guidelines.
> > > >
> > > > Do we gain anything if it's optional?
> > >
> > > As I recall, one thing that bothered you about Signed-off-by in Mesa
> > > is that it wasn't documented what it meant when it was used.
> > >
> > > Perhaps there are developers that don't want to use Signed-off-by with
> > > an undocumented meaning for Mesa. If that is the case, then this might
> > > help. I wasn't sure if you fell into that category.
> > >
> > > I use -s whenever I commit, so requiring it would not bother me. But,
> > > I notice that many people (such as yourself) do not, so I didn't see
> > > the need to push for that.
> > >
> > > If it's well documented, and becomes commonly used, then perhaps
> > > requiring it might be a reasonable thing to consider. I won't be
> > > holding my breath while waiting on that. :)
> >
> > I don't have a problem requiring it. I sign-off on commits I make to
> > Gentoo, to Linux, etc.
>
> If it has the same meaning as with the Linux kernel, but is not
> required, then you won't use it?
>
> I guess your concern might be that you are then giving something to
> the project that others can choose not to. ?

I'm not aware of any projects that define S-o-b but make it optional.
I suspect there's a reason that those that define it require it.

> > I'm just against cargo-culting it like we're doing now without a
> > defined meaning.
>
> The purpose of this patch is to give it a defined meaning. And the
> meaning I chose is the one that people are more likely to have in mind
> when they use Signed-off-by. Maybe that's too big of an assumption on
> my part, but I think several other open source projects have followed
> the kernel on this.

I understand, and agree. I'm just suggesting that it may not be good
to do something different than all the other projects that
meaningfully use S-o-b (i.e., by making it optional).

> > By all means, require it (with a git hook) if you like.
>
> I personally don't want to push for that right now.
>
> I guess I would like it to be required someday, primarily because it
> creates a standard process for open source projects to use. (So,
> people are more likely to be used to it in general when contributing
> to open source projects.)
>
> But, I'm not confident that the consensus for Mesa would be in favor
> of making that change right now. So, as an alternative I'd like to
> remove any barriers (such as ambiguity) to its usage in Mesa.

I don't think requiring S-o-b is onerous. Using S-o-b in a defined way
is just making explicit the current implicit promise to the community
that "I have the right to contribute this code" etc. I'm not sure why
you're hesitant.