[OSM-talk] Could we just pause any wikidata edits for a month or two?

[Christoph Hormann]

On Wednesday 11 October 2017, Martin Koppenhoefer wrote:
> From what I have seen so far, this should probably be less of a
> concern,
>
> but it is an uncertainty (because it could be interpreted more
> rigidly in the future)

I am not judging here, i merely stated the observation that Wikidata in 
its current form would not be capable to function as an universal 
meta-database connecting OSM with other open data sets.

To be honest - even if Wikidata was willing to change their Notability 
rules in a way that fully covers OSM (like by adding a rule declaring 
everything in OSM notable) this would still not work because it would 
also need to cover everything in any other open database OSM might be 
connected to and that quickly scales to a point where Wikidata almost 
certainly does not want to go.

> For me, criterions pro wikidata are:
> - it has a very permissive license (cc0)
> - it is openly accessible
> - it is fully downloadable as a dump (i.e. I don't have to use APIs
> which might log what I look at or limit the speed or quantity of my
> access) - there is overlap with our field of interest

You should be aware that these same criteria are fulfilled by a large 
number of other data sets, like for example almost all geodata 
published by the USGS or other US federal institutions.

> > But my question was specifically to what extent data has been
> > transferred based on wikidata ID references.  The question if such
> > data transfer happened before based on other connections has
> > nothing to do with this.
>
> Is this about reliability of the information, or about licensing
> questions?

If and how wikidata tags in OSM are used by Mappers to transfer 
additional information from Wikidata to OSM is important to know i 
think.  I have no specific hypothesis here.  But obviously such 
transfer can result both in legal and quality concerns.

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