[OSM-talk] [OSM-legal-talk] In what direction should OSM go?

[Anthony]

On Thu, Sep 30, 2010 at 1:04 AM, Toby Murray <toby.murray at gmail.com> wrote:
> On Wed, Sep 29, 2010 at 11:22 PM, Anthony <osm at inbox.org> wrote:
>> In the particular case we were talking about data that "cannot be
>> obtained by surveying" (*).
>> [...]
>> (*) I actually can't think of any boundary data for which this is the
>> case, though.
>
> In that case I guess I need a how-to. How should I survey my county
> border?

I never said *you* could survey it.  I suppose you could, with enough
time and money and resources though.

> So then since it can't be surveyed, we shouldn't have these borders in
> OSM?

If none of it can be surveyed, and it doesn't connect to anything
which can be surveyed, then we shouldn't have it in OSM.  That's what
I said, more than once actually.

However, I still don't know of any boundaries for which that is the case.

> That means you can no longer do queries on OSM data based on "is
> in ???" which would seem to be kind of a big feature for a mapping
> application.

If there is a boundary which isn't in OSM, obviously you'd need to use
a database other than OSM to load the boundary.  Fortunately this is
an easy thing to do.

> I guess any time you want to do anything useful with OSM
> data you would have to go out and find your own boundary data then?

No.  As I've said, I don't know of any boundary data which can't be surveyed.

> That seems rather unhelpful too. Don't we want to make it reasonably
> easy for people to use our data?

It is reasonably easy to load more than one database.  It would be a
miniscule portion of the design or even just installation of a system
to use that data.