[OSM-talk] Attribution guideline update

Christoph Hormann osm at imagico.de
Thu Feb 20 11:25:48 UTC 2020


On Thursday 20 February 2020, Simon Poole wrote:
>
> Artificial "yes", but the main thing is that it is small enough to
> ensure that it will essentially never be a substantial extract, on
> the other hand large enough that you can cover the location of your
> entrance, parking lot or whatever in it, with other words, large
> enough to be useful.

First: This has absolutely no place in an attribution guideline, in 
particular since we already have a guideline specifically dealing with 
the subject of what is a substantial extract of OSM data.

Second: You are here essentially declaring almost all indoor mapping 
performed within OSM (with the exception of really large structures 
like large airports) to be insubstantial and therefore not protected by 
the ODbL and free to take and use without attribution or share-alike.

Given the highly variable mapping density in OSM and the fact that there 
is no limit in how detailed people may map things the whole idea of 
having a physical area limit for defining what is substantial seems 
inappropriate for OSM.

And yes, that even more applies to the 1000 inhabitants limit which even 
back in 2014 when that was adopted was not appropriate.  You can find 
areas with less than 1000 inhabitants in OSM with tens of thousands of 
features and many megabytes of data.  Considering that insubstantial is 
fairly outrageous and as others have pointed out it would also not be 
compliant with the the obligations OSM has towards data providers who 
provide us data under the condition we distribute it under the ODbL, 
not to put it effectively in the public domain.

-- 
Christoph Hormann
http://www.imagico.de/



More information about the talk mailing list