[Tagging] wikidata <> other ref
marc_marc_irc at hotmail.com
Wed Nov 29 00:40:38 UTC 2017
Le 28. 11. 17 à 21:29, Kevin Kenny a écrit :
> Having nonvolatile ID's for ways really is technically infeasible
it's impossible but here it is "s/toT"
a id that match all osm id for highway "Avenue des Champs-Élysées" in
Paris taged as node, way, type=associatedStreet relation and type=route.
it resists id changes in osm, merged or cuted way and requires no
wikidata key in osm.
To find the way concerned, it's as easy to use as to make a request for
objects with the desired wikidata.
usecase : prefix the id with https://overpass-turbo.eu
like https://overpass-turbo.eu/s/toT (of course the query can be simplified)
In fact, we doesn't need this id at all, we just need the overpass model
filled with the name and the city (such fild are already available in
the wikidata files correctly described).
> Here is an Overpass query that will return the OSM data for the
> same Wikidata ID, irrespective of what happened to volatile
> objects: https://overpass-turbo.eu/s/toj
if you need overpass api to get all osm object using wikidata,
my query shows you that you can get the same without wikidata.
you're just showing that the solution is overpass, not wikidata.
If you feel that the info in osm is volatile (= modified sometimes by
mistake, forgetting to add it in the appropriate relationship for
example or have a name key with a bad value), then a query on the
wikidata key does not guarantee you to find these objects that are not
kept in good condition, which sometimes don't have the wikidata key.
I like ref for what they are : a tools to identify an object among many
others (for example a specific streetlamp in a city). these objects
there will never, in my opinion, have a wikidata. for all others, the
criteria used to find all objects on which you add the wikidata key can
also be used as a link to these objects. In reality those who use
wikidata need to have this request somewhere to do the regular
maintenance of objects without wikidata key. So why having to do regular
maintenance (and therefore have errors between 2 run) if you can reach
the same quality level without additional maintenance ?
I saw a usecase of wikidata, the user started by making a update to fix
missing wikidata tags and then make a request to retrieve the objects he
had just taged, that's absurd, the first query allready give all objects
More information about the Tagging