[Board] [Fwd: Geotools legal matters for the OSGeo board meeting]

Frank Warmerdam warmerdam at pobox.com
Fri May 18 11:21:53 PDT 2007

Arnulf Christl wrote:
> Hey,
> this mail was a little late for me to read before the last board meeting
> (42 minutes). GeoTools explicitly asks us to decide on some things so they
> can carry on with Incubation. I suggest we go through them in the list so
> next time we meet we can motion and approve what has to be done.


I had a
>> ************************************************************************
>> ************************************************************************

Note that we did this as far as I know.  Chris took the action to fill
in the holes in the proposed agreement, and we agreed to then refer the
issue to our outside lawyer on open source issues (Heather?).

>> I.   Room for different types of contributions:
>>          Neither the OSGeo CA nor the FLA consider explicitly the three
>>          different types of contributions which are currently going into
>>          Geotools. We have contributions of code to be GPL'ed,
>>          contributions of documents to be FDL'ed, and contributions of
>>          demonstration code/website contents to be released under BSD (we
>>          would like to have this in the public domain but there is no
>>          legal way for us to place it there).
>>          Even my current draft doesn't deal quite well enough with this
>>          multi-contribution situation.
> This touches an issue brought up by the GRASS PSC. What is the scope of a
> project? Does it include data, website contents, community, brands, etc.
> There is no real definiton for any of the projects.

Item I. above does not seem to relate to scope of project.  To me this
concern suggests that any agreement should not be too tight.  For instance
statements about OSGeo having to offer contributions under an OSI license
should be specific that this applies to source code, giving us the
alternative of offering documents under non OSI licenses.

>> IV.  Intellectual property
>>          Paraphrasing Richard Stallman:
>>                  Anyone using the terms 'Intelectual property' is either
>>                  confused themselves or is trying to confuse you.
>>          Indeed, this is sloppy shorthand for 'all that legal stuff
>>          around rights like property rights but for things which are not
>>          in reality exclusive'. Since the idea groups together things
>>          which are totally different---copyright, trademark, patents---we
>>          would do well to steer very clear of the term, most especially
>>          in a legal document. The FLA doesn't use the term but the OSGeo
>>          CA does.
> +1 from Arnulf to not use this term in any context.

I understand this objection but I do not find it compelling.  However, I
would encourage referring specifically to copyright and licensing when that
is all we mean.

> I support to drop any references to patents altogether in copyright
> assignment. We may have to deal with it, but this can happen separately to
> GeoTool's graduation.

I am in favor of keeping the scope of a copyright assignment as narrow
as is practical, and excluding patent issues helps keep it much simplier
and makes it easy for folks to sign it.

>> VIII.Re-licensing
>>          Part of the reason to assign copyright to OSGeo is to provide
>>          some way to re-license the code under a new version of the GPL,
>>          e.g. GLPv3. However if OSGeo has the right to re-license, it
>>          necessarily has the right to re-license under any license
>>          including a proprietary license. While no one images OSGeo will
>>          do so, it doesn't hurt to make this explicit.
>>          This raises two issues: (1) How do we make this explicit and (2)
>>          how does OSGeo decide to re-license.
>>          The first issue can be solved in two ways: explicitly in the
>>          document like in the FLA or ignoring the issue in the document
>>          and trusting OSGeo to remain constrained by its charter.
>>          Discussions with Frank and others at various times, seemed to
>>          suggest that the second solution would be easier for you all to
>>          agree to while the first solution is preferred by the Geotools
>>          contributors.
> Version one will require to change copyright and licensing twice, once on
> the OSGeo level and once for the GeoTools project. If the Charter binds
> OSGeo it can react to changes in licensing or legislations once and
> resolve the problem for all its projects. If the project's license needs
> to be revisited individually it just ads overhead work. One reason for
> OSgeo was to be able to deal more efficiently with stuff that sucks.

I am a bit lost here.  As discussed above with regard to the documentation
license, I think we need to avoid applying OSI-only rules to all
contributions.  Explicit rules in the agreement may put us in a catch-22
where we can't do the sensible thing.

Ok, now I am getting Arnulf's point.  Yes, I think the OSGeo charter/bylaws
is the best place to embed rules about appropriate licensing.  Amoung other
things it is changable with a sufficient majority of the charter members

>> Board Decisions:
>> ---------------
>> Ideally, during your meeting you will reach the following decisions:
>>       1. If you are willing to work with us to craft a different form of
>>          the existing OSGeo Copyright Assignment document.
> (the following are votes from Arnulf assuming that we had moved and seconded)
> +1

The board was is general agreement with the proposed agreement assuming it
is finalized, and reviewed by appropriate legal counsel.

>>       2. If you want to ignore or explicitly deal with submarine patents.
> Ignore patents and remove corresponding wording from th F/CLA.
> +1


>>       3. If you want to require language for Ongoing Assitance by
>>          contributors
> (I don't really understand what this is supposed to be good for?)
> Drop it, if it is no-goer for GeoTools. Keep it if we can use it to appear
> more long lasting and can make us of it in marketing.

I don't know about this either.

>>       4. If you are willing to explicitly commit to keeping the code
>>          available under a 'free' or 'open source' license.
> I will happily commit to keep it under a Free and Open Source Software
> license.
> Currently OSGeo uses OSI-approved licenses. So as soon as it would
> relicense to anything else it would lose the project as it cannot have
> projects with non-OSI approved licenses because of its own rules. Cool.

I think offering OSGeo code under free/opensource approved licenses is
bedrock principle.  For now this is encoded as OSI approved licenses.
I think we need to be careful to not necessarily apply this to all "content".

>>       5. If you are willing to explicitly grant the user broader rights
>>          to their own contributions than they would get by being a
>>          licensee of the OSGeo code.
> This sounds a little like a dual copyright holder?! In European legal
> background (thanks for your help above) there is a differentiation between
> originator's (Urheber) rights and exploitation rights (Verwertungsrecht).
> Both are contained in the one English word "Copyright". The originator's
> rights can not be sparated from the code (work of art, etc.) ever. It is
> part of the work. Exploitation rights initially lie with the contributor
> but he can give them away. From my perspective this covers what you
> describe above.

I am generally not in favor of giving the original author special rights
over their code if they are assigning the copyright to OSGeo.  Of course,
where the law specifies otherwise that is fine.   But I don't care to craft
special language giving the original author special rights.

>>       6. How you want to address re-licensing.
>>From my point of view this is to keep the door open for the future. We saw
> many trouble and time going into changing from GPL 1 to 2 and want to
> avoid something similar to happen again.
> (((As an aside: OSGeo is just a bunch of people and source code flocking
> together. If anything really goes amiss make a fork and be happy. Where is
> the issue? We are building a paper tiger to show off how big we are, is
> that not it in the end?)))
>>       7. If you are willing to build on my version of the Fiduciary
>>          Agreement document.
> +1 from Arnulf
>>       8. What process you wish to adopt going forward which will lead to
>>          the rapid:
>>               1. completion of the text,
>>               2. review by your lawyers, and
>>               3. final approval of the completed document by the board

Yes, this is the process we agreed on in the meeting.

Best regards,
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org

More information about the Board mailing list