Towards the end of incorporating some of the good work from clojure-
contribs and others into Clojure, I have been looking at adopting
Sun's contributor agreement for Clojure:
The main purpose of the SCA for Clojure would be to prevent it from
becoming license-bound, i.e. everyone contributed under license X,
preventing the project from moving to license Y, or X + Y.
I like it because contributors give up none of their own rights, vs
projects which require a copyright assignment. It also includes a
promise that it will remain open source.
On Wed, Jul 2, 2008 at 9:37 AM, Rich Hickey <richhic...@gmail.com> wrote:
> Towards the end of incorporating some of the good work from clojure- > contribs and others into Clojure, I have been looking at adopting > Sun's contributor agreement for Clojure:
On Jul 2, 9:07 am, Chouser <chou...@gmail.com> wrote:
> On Wed, Jul 2, 2008 at 9:37 AM, Rich Hickey <richhic...@gmail.com> wrote:
> > Towards the end of incorporating some of the good work from clojure-
> > contribs and others into Clojure, I have been looking at adopting
> > Sun's contributor agreement for Clojure:
My main concern would be that this kind of requirement would reduce
the contributions made to the language. IIRC, there was some some
gnashing of teeth by people who wanted to contribute to I think it was
OpenOffice, but the Sun policy discouraged some contributors.
There are several other large projects that don't bother with this
kind of stuff. Are they being less diligent, or do they not have to
worry about it because of the license they use?
> My main concern would be that this kind of requirement would reduce
> the contributions made to the language. IIRC, there was some some
> gnashing of teeth by people who wanted to contribute to I think it was
> OpenOffice, but the Sun policy discouraged some contributors.
If someone feels up to the task, could they give us a brief overview
of the arguments from the dev community for or against this
contributor's agreement?
On Wed, Jul 2, 2008 at 10:39 AM, Allen Rohner <aroh...@gmail.com> wrote:
>> On Wed, Jul 2, 2008 at 9:37 AM, Rich Hickey <richhic...@gmail.com> wrote:
>> > Towards the end of incorporating some of the good work from clojure- >> > contribs and others into Clojure, I have been looking at adopting >> > Sun's contributor agreement for Clojure:
> My main concern would be that this kind of requirement would reduce > the contributions made to the language. IIRC, there was some some > gnashing of teeth by people who wanted to contribute to I think it was > OpenOffice, but the Sun policy discouraged some contributors.
Apparently there have been earlier version of the agreement which caused more unhappiness that the current version has.
> There are several other large projects that don't bother with this > kind of stuff. Are they being less diligent, or do they not have to > worry about it because of the license they use?
From what I've seen, it's easy to overlook this kind of detail, and its possible doing so would never actually cause a problem. But things can happen later that without this can cause a huge headache with just about any license. For example, if a legal challenge to the main license requires it to be even slightly adjusted and there's no centralized control (either the SCA, signing over of copyright, or something) such a minor adjustment would require every author of every patch to confirm the license change. It may be very difficult or even impossible to contact some authors, which would then require the removal of their patches until someone can re-write them (without violating the author's copyright!) ... just painful. As a user of Clojure, I'd be happy to know that the legality of the project is robust and can adapt to real-world challenges without this kind of mess.
I believe some projects have required contributors to sign over all rights to an organization such as the FSF. This solves the above problem, but leaves the author without the right to re-license their own work. By using the SCA, as a contributor to Clojure I could still use my own contributions in my own closed-source product, re-license them under some license I prefer, or whatever I want. I still own my own work.
I'm not sure what Rich has in mind, but it may be good to allow clojure-contrib to continue with less strict rules, which might help to continue encouraging contributions and the kind of trial-by-fire we've seen with lib.clj. Only when a contribution has proved itself there would the author be required to sign the CA to allow the code to be moved into Clojure proper.
On Jul 2, 11:02 am, Duane <dsearsm...@gmail.com> wrote:
> > My main concern would be that this kind of requirement would reduce
> > the contributions made to the language. IIRC, there was some some
> > gnashing of teeth by people who wanted to contribute to I think it was
> > OpenOffice, but the Sun policy discouraged some contributors.
> If someone feels up to the task, could they give us a brief overview
> of the arguments from the dev community for or against this
> contributor's agreement?
IIRC, it originally drew criticism for not having the "Any
contribution we make available under any license will also be made
available under a suitable FSF (Free Software Foundation) or OSI (Open
Source Initiative) approved license." language.
> On Wed, Jul 2, 2008 at 10:39 AM, Allen Rohner <aroh...@gmail.com> wrote:
> >> On Wed, Jul 2, 2008 at 9:37 AM, Rich Hickey <richhic...@gmail.com> wrote:
> >> > Towards the end of incorporating some of the good work from clojure-
> >> > contribs and others into Clojure, I have been looking at adopting
> >> > Sun's contributor agreement for Clojure:
> > My main concern would be that this kind of requirement would reduce
> > the contributions made to the language. IIRC, there was some some
> > gnashing of teeth by people who wanted to contribute to I think it was
> > OpenOffice, but the Sun policy discouraged some contributors.
> Apparently there have been earlier version of the agreement which
> caused more unhappiness that the current version has.
> > There are several other large projects that don't bother with this
> > kind of stuff. Are they being less diligent, or do they not have to
> > worry about it because of the license they use?
> From what I've seen, it's easy to overlook this kind of detail, and
> its possible doing so would never actually cause a problem. But
> things can happen later that without this can cause a huge headache
> with just about any license. For example, if a legal challenge to the
> main license requires it to be even slightly adjusted and there's no
> centralized control (either the SCA, signing over of copyright, or
> something) such a minor adjustment would require every author of every
> patch to confirm the license change. It may be very difficult or even
> impossible to contact some authors, which would then require the
> removal of their patches until someone can re-write them (without
> violating the author's copyright!) ... just painful. As a user of
> Clojure, I'd be happy to know that the legality of the project is
> robust and can adapt to real-world challenges without this kind of
> mess.
> I believe some projects have required contributors to sign over all
> rights to an organization such as the FSF. This solves the above
> problem, but leaves the author without the right to re-license their
> own work. By using the SCA, as a contributor to Clojure I could still
> use my own contributions in my own closed-source product, re-license
> them under some license I prefer, or whatever I want. I still own my
> own work.
> I'm not sure what Rich has in mind, but it may be good to allow
> clojure-contrib to continue with less strict rules, which might help
> to continue encouraging contributions and the kind of trial-by-fire
> we've seen with lib.clj. Only when a contribution has proved itself
> there would the author be required to sign the CA to allow the code to
> be moved into Clojure proper.
The way this CA works is the rights get granted when the contribution
is made, i.e. by the act of posting or submitting. So there need to be
places where posts constitute contributions, and they are clojure and
clojure-contrib. Since things in clojure-contrib will have been
contributed to Clojure, no other step will be required for me to
integrate them, either by moving code into boot.clj et al or shipping
Clojure with a contrib directory. That's not to say there won't be
utility to some kind of vetting or indication of readiness of contrib
work, but that's independent of it being code intended for inclusion
in Clojure.
I don't want there to be any question about my ability to use
everything in contrib, so I don't think it can have relaxed rules. The
plan is that a CA will be required before access to clojure-contrib or
clojure is granted. Of course, if anyone already in clojure-contrib
has an issue with a CA, their stuff will be moved out first. This is
why I haven't pulled from contrib yet - I want to do this with
consensus on a CA in place.[*]
The ultimate objective is for Clojure to have a sound way of
incorporating the enhancements and contributions of others, and
fostering community, which requires my moving from sole author to
steward. The CA is a tool enabling stewardship, and as you say,
assuring users of same.
I understand that a CA will put a speed bump before some
contributions, but as this CA is specifically about sharing, and not
giving up anything, I would hope all potential contributors would
consider it acceptable, considering the spirit in which Clojure was
originally shared.
Rich
* I've also looked at the Apache and FSF agreements, but prefer the
SCA. Other suggestions welcome.
On Jul 2, 10:07 am, Chouser <chou...@gmail.com> wrote:
> On Wed, Jul 2, 2008 at 9:37 AM, Rich Hickey <richhic...@gmail.com> wrote:
> > Towards the end of incorporating some of the good work from clojure-
> > contribs and others into Clojure, I have been looking at adopting
> > Sun's contributor agreement for Clojure:
The terms look good to me. I'll be mailing off a copy soon. I appreciate your thoroughness in safeguarding Clojure's long term future.
--Steve
(As a minor point, I see that this phrase comes from Sun:
"In order to track contributors, you understand that your full name and username may be posted on a public URL of authorized contributors"
The phrase "on a public (uniform resource locator)" doesn't make perfect sense to me as English. I take it to mean "at a publicly accessible location with a published URL".)
