Fwd: [OSM-talk] BSD/CC-by/LGPL vs. SharedAlike - decide now and forever

[Lars Aronsson]

Immanuel Scholz wrote:

> > I don't know what it would take in practice to enforce
> > this transfer of freedom.
> 
> I think there are two parts required: Specify a license that 
> allow as many people as possible to access the data *and* to 
> restrict other people so they cannot suppress the community 
> making advantage of the data.
> 
> I think a viral license like CC-By-Sa or GPL fit for both. BSD 
> likes only provide the first part well.


Perhaps you don't understand the meaning of the word "enforce"?

Case 1: I release my data in the public domain and somebody uses 
it for a derived work, and they don't publish the source to the 
derived work.  This is OK, because I released my data in the PD.

Case 2: I release my data under CC-SA-BY and somebody uses it for 
a derived work, and they don't publish the source to the derived 
work.  This is not OK, it is a violation of the license.  But they 
don't care.

There is a clear risk that the practical outcomes of cases 1 and 2 
are identical.  If people can just ignore the licensing terms and 
get away with it, the choice of license is meaningless.  
Enforcing the license means achieving a real, practical difference 
in the outcome, such as making these violators comply with the 
license by releasing their data, or pay damages, or go to jail.  
Just going to the police station and filling out a form, reporting 
the crime, is not going to help me.  The prosecutor will say "case 
closed" before I leave the building, and the violators will laugh 
out loud. I will have to pay lawyers, some lawyers that understand 
CC licensing issues.  And I will have to prove that the derived 
work was based on my data.  How much does this cost in time and 
money?  Is it feasible?

Imi, I know you have practical experience from writing software.  
I know the Opentom people have practical experience from enforcing 
the GPL.  But who has any practical experience from enforcing 
CC-SA-BY for map data?  Anybody on this planet?  I doubt that.

> > Imi and other GPL fundamentalists need to explain the *practical*
> > differences between licensing options *for map data*.
> 
> 1) may redistribute our data for free.
> 2) must put published modifications of the data under same license.
> 3) must put published deriverates (maps, things created using the data)
> under same license.
> 4) must provide the "source" (vector data or even raw gps tracks) that
> lead to published deriverates
> 
> Public domain: 1)
> BSD-like: 1) and 2)
> CC-By-Sa: 1), 2) and 3)
> GPL: everything.

I'm sorry, but this is a completely theoretical analysis, and has 
nothing at all to do with the practical details of enforcement.



-- 
  Lars Aronsson (lars at aronsson.se)
  Aronsson Datateknik - http://aronsson.se