[OSM-talk] Frederik declares war on data imports...

[Matt Amos]

On Mon, Aug 9, 2010 at 10:41 PM, Julio Costa Zambelli
<julio.costa at openstreetmap.cl> wrote:
> On Mon, Aug 9, 2010 at 3:45 PM, Matt Amos <zerebubuth at gmail.com> wrote:
>> OSMF is not moving to "a PD license disguised as BY-SA", OSMF would
>> like to move to ODbL. however, it has to be pointed out that CC BY-SA
>> might be described as "a PD license disguised as BY-SA", since many
>> lawyers (including those at Creative Commons) think that CC BY-SA is
>> unsuitable for factual data (such as geodata) and may not be
>> enforceable in many jurisdictions (such as the USA).
>
> I know about the problems with (CC)BY-SA, and I also know that ODbL is
> supposed to solve those. And unless I am getting lost in translation I
> do not have any problem with ODbL, but with the point made by John
> Smith about the third condition on the CT ("OSMF agrees to use or
> sub-license Your Contents as part of a database and only under the
> terms of one of the following licenses: ODbL 1.0 for the database and
> DbCL 1.0 for the individual contents of the database; CC-BY-SA 2.0; or
> another free and open license").
>
> In the process of approving the change to ODbL the Foundation is
> asking us to let it change the license to something that may be PD in
> the future. That said the imports that we have made here in Chile are
> probably compatible with ODbL but not with letting the foundation
> change the license to something "more open" than BY-SA.
>
> Again, risking some misunderstanding with my far from perfect English,
> I understand from what you are saying that two problems are trying to
> be solved, the unfitness of the (CC)BY-SA license for our kind of
> data, and the risk of loosing data in future changes of license.
>
> The thing is that I am all about solving the first but not about
> lossing lots of data today speculating about that first solution
> failing in the future, risking lots of data then.

unfortunately, we will lose data this time around - it's unavoidable
because it's extremely unlikely that all contributors will be
reachable and, as many here have pointed out, be willing to agree to
the new CTs. if the data you're worried about is governmental
attribution datasets (such as OS opendata, LINZ, NRC, etc...) then,
pending legal advice, they could be fine.

> The only reason that I see to put that condition there is thinking
> about changing the license to PD in the future without asking all the
> contributors again.

or to change the license to something else which is also BY and SA, if
it turns out that's necessary. or to move to a BY-only license, if
that's necessary. is the possibility of needing to change the license
again in the future not worthwhile, given the problems it's causing
right now?

>> whether section 4 is enough to allow CC BY compatibility is something
>> that OSMF is currently seeking legal advice on.
>
> I guess this will help, but if the license can be changed in the
> future to PD without asking the Gov agency, I am almost sure that they
> will not comply with this.

i think that's a question for a real lawyer. ;-)

>> if (as i hope) the lawyers say that section 4 of the CT ensures
>> compatibility with CC BY, why would section 3 pose a problem? if
>> section 4 requires that OSMF provide a method of attribution then that
>> couldn't be taken away by changes under section 3 unless a new version
>> of the CT were released - which would require asking every single
>> contributor and re-raising the problem of data loss: the very problem
>> that section 3 is supposed to alleviate.
>
> I think this time I actually got lost in translation but as far as I
> understand it, the point 4 is useless if it can be discarded without
> asking the contributors. Am I getting something wrong?

point 4 cannot be discarded without asking all the contributors who've
agreed to the contributor terms. so it's far from useless in
guaranteeing attribution.

>> migration to PD is not part of the plan. the motivation for that
>> section is simply that needs and requirements change over time. when
>> the project was started CC BY-SA seemed like a perfectly valid
>> license. we're now 6 years on, and 2 years into trying to change the
>> license, because we were wrong about CC BY-SA. while we think ODbL is
>> far, far better - do we want to have the spectre of data loss again in
>> another 6 years if we prove to be wrong again?
>
> I think it is a perfectly reasonable risk in front of a sure damage.

and what might the damage be in the future if we need to change in the future?

>> no-one wants to see any data loss. that's one of the many reasons
>> we're moving from a BY-SA license to another BY-SA license. while
>> there is an option to declare your preference with regard to PD, this
>> is for information only.
>
> It is a BY-SA to BY-SA moving as long as you do not give the OSMF the
> right to move to PD (or anything different from BY-SA for this matter)
> without asking again.
> At this point I do not see any good reason to prefer PD and accept to
> consecuences of moving to it.

that's great - so you don't have to tick the "i consider my
contributions PD" box. that's the default anyway. ;-)

cheers,

matt