incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Rob Weir <robw...@apache.org>
Subject Re: CWS swbookmarkfixes01 rebasing and licensing
Date Tue, 03 Jul 2012 17:18:29 GMT
On Tue, Jul 3, 2012 at 11:44 AM, Bjoern Michaelsen
<bjoern.michaelsen@canonical.com> wrote:
> Hi Ross,
>
> Thanks for the constructive reply!
>
> On Tue, Jul 03, 2012 at 03:58:54PM +0100, Ross Gardler wrote:
>> If the CWS was included in the original SGA then it is available under
>> the AL2.
>
> How can I check that? It was not integrated in m106 at OOo, if that is the
> criteria, but I assume there is more to that ;)
>
>> If it was not included in that original SGA, you want to bring it here and
>> the OpenOffice project want to see the work committed then Oracle will, in
>> all likelihood, make it available to us.
>
> Great!
>
>> Therefore, I suggest the following order of execution:
>> - determine whether the OpenOffice committers want the work
>
> Where would that be done?
>
>> - confirm that Oracle have already or will make the code available
>> under the AL2 at our request
>
> Are there established workflows for that, or should I just go about that ad-hoc?
>
>> - submit patches against AOO trunk
>
> Are the patches already released AL2 by Apache when they are published on the
> list then? Given that AOO might switch to the Symphony codebase and fail to
> even release a version with the patches against the original OOo, I would need
> that assureance that the effort to rebase the work is not in vain.
>

The particular circumstances you are in, and the task you are trying
to accomplish, is not necessarily covered by a specific single
process.  So rather than focusing on process, I'd recommend
considering the overall forces at play and the known patterns for
resolving these forces.

If someone wants to contribute code to the project, then they need to
deal with three basic things:

a) Does the project want the code?

b) Technical integration

c) Getting the code under ALv2 or a compatible license.

This is true for committers as well.  We all need to be concerned with
these questions.  The main difference is that a committer operates
under Commit Then Review (CTR) rules.

#1 and #2 above are relatively easy.  It is social and code.  #3 is
not so much about a single process, but a logical argument.  Someone
contributing code should have a logical argument for why the code is
properly licensed.   There are several well-known conventions for
making this kind of argument, using procedures like an SGA.  Even
committers implicitly make this argument whenever they check in code,
based on their iCLA obligations.

However, these procedures are neither necessary nor sufficient.  But
they are well-known and following these conventions helps us all avoid
a lot of extra analysis.   Of course, if necessary we'd be open to
considering other forms of argument aimed at proving the suitability
of license for a code contribution.

Back to your original question:  An author posting original work to
the list, with an intent of contributing it to the project would
typically be accepted as an ALv2 contribution.  An iCLA for the author
would be preferred for larger contributions, but an unambiguous
statement on the list (or in Bugzilla) is a good start.

But posting 3rd party code to the list, i.e., not original work, that
would probably raise a red flag.

Hope this helps.

-Rob

> Looking forward to hopefully some constructive cooperation.
>
> Best,
>
> Bjoern

Mime
View raw message