[OSM-talk] Draft Geocoding Guideline

Martin Koppenhoefer dieterdreist at gmail.com
Thu Jul 20 19:35:00 UTC 2017


2017-07-20 20:35 GMT+02:00 Michael Steffen <michael at mapbox.com>:

> These changes are now reflected in the wiki: https://wiki.
> openstreetmap.org/wiki/Draft_Geocoding_Guideline
>


Is my interpretation correct that if this guideline goes into effect I
could get millions of addresses and their coordinates from OSM, without any
obligation to attribute, as long as I only store names, addresses and
coordinates, and use a different systematic approach then "everything in a
given area at least as big as a city"?
(Btw., for practical reasons, wouldn't it make more sense to specify rather
than "city" an actual size like "contiguous area smaller than 1000 sqkm"?)

E.g. my algorithm could take a list of all streets, query all housenumbers
from 1 to x (until it doesn't find any more hits for a sequence of
numbers), but not the numbers 3 and 4 (so this is explicitly not
"everything", I am taking particular care not to take "everything").

Then I could make a specific address geodatabase which is different to a
"general purpose geodatabase" (or is it not?) and distribute it under
whatever license?

Or I could open a company, geocode for other people, and let them use the
results without attributing to osm, even if I'm doing repeated and
systematic queries for much bigger numbers as what the substantial
guideline defines?

Why do we treat addresses, which are among our most valuable information
(as they are tedious to collect and it is hard and often impossible to find
free existing sources), differently than other data that is in the db, e.g.
hotels, shops or hospitals? Is this a political decision to "spread the
word"? There is already a guideline about what is "substantial", why are
there different criteria for substantial in the geocoding db, and not e.g.
simply a reference to the substantial guideline?

In particular: "Individual Geocoding Results are insubstantial database
extracts if they are based on a Direct Hit." , yes, individual geocoding
results are not substantial, but geocoding is typically executed many times
(i.e. systematically), and it is the sum of the geocoding results that
makes the extract substantial.

Even (to a lesser extent) the following statement could be contested:
"Individual Geocoding Results that are based on an Indirect Hit contain no
OSM data and so are free of any obligations under the ODbL."  Rather than
"contain no osm data" this could be seen as "contain only transformations
of osm data, no raw osm data". A collection of Indirect Hits can be seen as
a derivative database according to the ODbL.

Cheers,
Martin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20170720/d46f807a/attachment-0001.html>


More information about the talk mailing list