[OSM-legal-talk] Licensing of works containing geocodes pinpointed on OSM data (Richard Fairhurst)

[Jani Patokallio]

On Wed, Oct 17, 2012 at 6:49 AM,  <legal-talk-request at openstreetmap.org> wrote:
> From: Richard Fairhurst <richard at systemed.net>
> On this specific issue: I'd suggest you consider whether your combination of
> OSM-derived data and other data is a Derivative Database (has to be shared)
> or a Collective Database (doesn't have to be shared). As a rough guideilne,
> we say that it's Derivative if you've adapted the two datasets to work with
> each other, Collective if you haven't.

Thank you for the response.  For the feed, basically we'd be looking
at a list of points of interest in this format:

<poi>
  PROPRIETARY DATA
  PROPRIETARY DATA
  PROPRIETARY DATA
  <location>
    <lat>OSM</lat>
    <long>OSM</long>
  </location>
</poi>

Would this be "adapted" or not?  If yes, what if we store the geocodes
in a separate file and cross-reference the two, like this?

Closed:
<poi>
  PROPRIETARY DATA
  PROPRIETARY DATA
  PROPRIETARY DATA
  <location>1</location>
</poi>

ODBL:
<geocodes>
  <location id=1>
    <lat>OSM</lat>
    <long>OSM</long>
  </location>
</geocodes>

> On the broader issue: I'd be interested to see a discussion as to how we
> should define 'Substantial', and 'Collective' vs 'Derivative', for geocoding
> (in terms of principles). I think it's reasonably uncontroversial to say
> that geocoding an unsystematic set of self-collected points is a less
> substantial use of OSM data than distributing the roads as part of a
> connected dataset. But I've not got much further in my thinking than that. I
> may go and hunt for some relevant case law (*shudders at thought of William
> Hill vs BHB*)...

Yes, this would be very interesting for us as well.

Cheers,
-jani