[OSM-talk] open data

[Tom Carden]

On 21/09/06, Tom Chance <tom at acrewoods.net> wrote:
>
> Anyway, I believe that a change in license would be impossible without
> contacting every single contributor, so I assume this discussion is rather
> pointless :o)
>

I think you'll find the 80-20 rule applies... in that you could change
the license on 80% of the data by obtaining permission from 20% of
contributors (likely it's even more weighted to the few than that...
Steve, Nick Hill, can we have a graph?).  I think Etienne's
observation also holds, that most areas only have one or two
contributors.

We don't have *that* many contributors.  We have email addresses that
have at least been confirmed once, for every user.  I suspect the
amount of data that depends on a contribution by someone who would
refuse to relicense is minimal, and could be thrown out.

Relicensing is definitely an option, but it must be to a license that
is appropriate for geodata.  As Richard has pointed out, there isn't
one yet... however I agree with Tom Chance that the fight shouldn't be
to loosen restrictions on OSM so that we can republish with
proprietary data, the fight should be to ask why the data is
proprietary in the first place...

We need a way to license a database of points (other features are more
complicated) such that a map with those points on can be distributed
freely, but such that reconstructing the database from the map is not
acceptable.  If someone was to draft such a license, I think it would
be a really useful tool for OSM consumers to use to approach people
with nice data sets (like recycling points).

Thoughts?

Tom.