[OSM-talk] Verifiability of names (was: Persian/Arabian Gulf Tagging)

[Niels Elgaard Larsen]

På Fri, 10 Dec 2021 13:49:43 +0100
Christoph Hormann <osm at imagico.de> skrev:
>On Friday 10 December 2021, Martin Koppenhoefer wrote:

>As i have explained many times:  OpenStreetMap is deliberately limited 
>in the scope of information it records to knowledge that it locally
>and independently verifiable.

OSM emphasizes local knowledge.

The wiki say "OSM data should, as far as is reasonably possible, be
verifiable"

When it comes to maritime names, they are generally very few signs.
Because ship would not pass close enough to read them, especially at
night. 

On the other hand there are official names usually recorded be maritime
authorities.

For example this I claim is "Nordre Røse"
https://www.openstreetmap.org/way/504812763


I have seen it many times and made photographs from every angle
https://commons.wikimedia.org/wiki/File:NdrRose.jpg

So unlike Atlantis I have verified on the ground that it exists.
However there is no sign with a name on it.

But anyone sailing in Øresund would know that it is indeed Nordre Røse.
And we can check the Danish lighthouse directory and nautical charts.

I think we have done what is reasonable possible. 

>  That is how we manage to cooperate 
>successfully on equal levels across language and culture barriers.  If 
>everyone wants and tries to record what they perceive to be part of 
>their individual or collective perception of the geographic reality 
>independent of local verifiability

No one on this list has proposed to do that.

> that would not work.  
>
...

>
>> That’s how names work, 
>> people use the same name so they know what they are talking about.
>
>So you think as long as a large enough number of people agree on a 
>common perception of reality that reality becomes true?  World history 
>is full of epic fails as a result of that belief.
>
>What is true is that names are part of the cultural sphere, they are
>how people refer to geographic features.  But that does not mean names
>in OSM stand above the paradigm of verifiability, we can record and 
>maintain names only when they are verifiably used to refer to the 
>actual feature in question locally.


There are places that have well known names, but no people to use them
locally. E.g., the North Pole and the Challenger Deep

And there are objects that are too big to have one name that is used
locally.
E.g. the Pacific Ocean.
Or the Continent Asia.


But I believe it still makes sense to name them in OSM.

>  Names that are used exclusively 
>by "people in far away countries" to refer to the virtual image of the 
>geographic reality they have collectively assembled in their culture 
>(like the mentioned example of Atlantis - but also plenty of much more 
>mundane examples of distorted and biased images of geography that
>swirl around,

"biased" is a very subjective term. Locally verified names can be
biased too. E.g, Berlin, ON was renamed to Kitchener because of
anti-german bias in WWI.

> you could say the tourism industry has essentially built

We are not talking tourism here.
I think it makes sense to consider what names the captain on a ship in
the area would use when talking to another ship or a marine traffic
controller.



>a business model on nuturing such distorted perceptions) do not belong
>in OpenStreetMap.  Leave those to projects like wikidata/wikipedia
>with their 'reliable sources' paradigm.
>