[OSM-legal-talk] Merano Italy will be donating their data! CC-by-SA and donated data?

[Tom Hughes]

In message <8fcd02310710300428j6644d831wf83679a37f62c849 at mail.gmail.com>
        80n80n at gmail.com wrote:

> On 10/30/07, Tom Hughes <tom at compton.nu> wrote:
>>
>> In message <9319.42062.qm at web30810.mail.mud.yahoo.com>
>>         Mikel Maron <mikel_maron at yahoo.com> wrote:
>>
>> > It occurs to me that one simple, clean way to address attribution
>> > more generally is a new action in rails which simply lists all the
>> > users who have contributed within some bounding box. The map, and
>> > any other use of the data, could then link to that page.
>>
>> There are two problems with that...
>>
>>   - Firstly we can't include any users that have not made their
>>     identity public yet we are still (in theory) bound to attribute
>>     them as that is what they signed up for.
>
>
> If a user has asked for their identity to be concealed then are they not
> implicitly saying that they do not wish to be attributed?  Otherwise we
> can't use their data, period.

That's a nice line in theory, and perfectly reasonable when 
considered from a common sense perspective, but does it hold any
water legally?

When a user signs up they agree that "all work uploaded to
openstreetmap.org and all data created by use of any tools which
connect to openstreetmap.org is to be licensed under this Creative
Commons license (by-sa)" which is clearly saying we're going to
publish their work under CC-BY-SA which in turn requires attribution.

If we have a checkbox that a user could actively tick to say that
they waived their right to attribution that might standup, but to
not attribute them by default when the signup screen effectively
said we would?

> If we take that line then we might as well throw away all anonymous edits as
> we can't attribute them, and can't use them without attribution, so we can't
> use them.

Yes, we have a problem because the intentions of what we've been
doing don't align very well with the strict legal details.

>   - Secondly we have dropped all history as of the 0.5 API upgrade
>>     so such a scheme could not provide attribution for any edits
>>     made before that date.
>
> We should maintain on each element a cumulative list of contributors
> as well as the most recent editor.  This data could easily be mined
> from the 0.4archive and applied to the current database.

Well "easy" might be overstating things, but we could certainly do
something.

Tom

-- 
Tom Hughes (tom at compton.nu)
http://www.compton.nu/