On 25.3.2020 16.20, Joseph Eisenberg wrote:
In that case, please add it to wikidata with a reference, but it would
not be appropriate for Openstreetmap.
Unlike wikipedia and wikidata, which are based on references and
citations to "authoritative" sources, Openstreetmap has always been
designed as a primary source: we map real, current features which you
can visit in person.
I agree that names in fictional languages should not be added to
name:<iso> tags, nor any name which cannot possibly be confirmed to be
true or false by local mappers.
If an Indonesian mapper adds name:id=* tags to every village in
northern Finland, how is anyone to confirm that they are right or
I strongly disagree with this. IMO the OSM data should be useful, and
because of that we should use our common sense just as in any other
tagging issue. The data won't be useful, if we can't actually add all
the relevant information but instead limit to whatever information is
available strictly on-the-ground. If this were the case, I'd found a new
project for actually useful information and stop contributing to OSM.
In my email I provided two links for verifying the Finnish name Jokohama
for Yokohama. Do you believe it's easier to find someone local in
Yokohama to verify this instead of verifying it online? If an Indonesian
mapper adds a name:id=* to the villages in Northern Finland, I wouldn't
go on-site to verify the data, I'd rather look up the names online.
That's the same thing I do as a first step to any data in Finland, even
data in Finnish.
I know ground verifiability is the guideline in OSM, but I'm sure
there's plenty of other tags that also somewhat expand from it. The wiki
even acknowledges this: ”OSM data should, as far as is reasonably
possible, be verifiable.” – the point I'm making being _reasonably
possible_. IMO the requirement you're suggesting isn't reasonable.
Tagging mailing list