Subject: RE: FOR APPROVAL: OZPLB Licence
From: Michael Sparks <zathras@thwackety.com>
Date: Mon, 27 Dec 2004 23:54:53 +0000 (GMT)

On Mon, 27 Dec 2004, Russell Nelson wrote:

> ian.oi@bdw.com writes:
>  > Many thanks for your comments - it was good to get some feedback, even if it
>  > seems rather hostile.
>
> Just an experiment to see if I could prod the license-discuss folks
> into action.  Didn't work; won't do it again.
>
> If your legal theory is correct, then this is a REALLY REALLY big and
> bad problem.

You seem to be inviting a general discussion here, so I think I'll put
in my tuppenceworth. (There were comment a little while back about
non-lawyers commenting putting off lawyers which may be a why to the lack
of reaction.)

I think there is a good chance it is a real problem and one that's liable
to get worse as time progresses rather than better. More countries
enforcing patents, so patent clauses become more issues, more countries
using open source in critical locations, so disclaimers become more
important, along with tort issues. (I have on a very limited understanding
of the specifics here)

I think sooner or later there does need to be a change in the way open
source licensing is dealt with, and it strikes me (in a highly unoriginal
fashion) that the approach that is most scalable in the end is the
approach taken by the Creative Commons crowd.

I'm not talking about the modular license idea - which I think would be a
real boost to open source personally - but specifically the simple fact
that they have taken the following approach:
   * Define a set of rights in non-lawyer, plain english
   * Define a machine readable representation of this
   * Define a simple way of represnting this (eg BY-SA-NC)
   * Then, using these as a specification, produce licenses as
     implementations.

This means that if I state on a webpage that the contents may be used
under a Creative Commons BY-SA-NC-2.0 license, I can be pretty certain
that the set of rights I want to grant *are* being granted in all the
countries/legal systems the licenses have been translated into.

As well as preventing proliferation of licenses due to "I don't like that
clause" or "I don't like this clause", it also avoids proliferations due
to "that doesn't work in australia", "that doesn't work in france", "that
doesn't work in germany" issues.

I doubt anything I'm saying here is news to anyone on this list, and
unfortunately I can't help provide translations into specific languages,
but one comment I might say is this:
   * The OSI has the OSL and AFL licenses as being cleaned up GPL/BSD
     equivalents (speaking as a developer view, not legal view).

   * Why not suggest (say) that there are OSL-au, OSL-uk, OSL-us, which
     may all be referred to as "OSL" in the same way that BY-NC-SA
     actually means BY-NC-SA-<your legal system> ?

Getting back to the original poster's reasons for posting their draft (the
disclaimer's validity in australia) - would you be happy with this sort of
approach ?

If the OSI don't do this, it would be great to see the FSF do this, but
it strikes me that the OSI would be the more natural group to go down this
route. It strikes me that collating a simple summary of the rights and
obligations of existing licenses in a CC style should be doable. (As in a
BY-SA-NC type approach.) Potentially something even a non-lawyer could do.
(If this is thought worthwhile, consider me volunteering :)

I suspect I'm saying something very naive, and hence making some kind of
gross mistake here, but it strikes me that the approach works very well
for the creative commons, so something similar for open source could work
well as well? (Let's face it "GPL", "LGPL" and "BSD" are treated by most
open source developers in the same sort of way that 'BY-SA-NC', 'BY-NC'
and 'BY' are used by people using CC)

Best Regards,


Michael.
--
I am not a lawyer, not affiliated with the OSI, this isn't legal advice.
etc, etc, etc...