I'd suggest using the 6 main United Nations languages for the "name=*"
tag of Oceans and Continents: Arabic, Chinese, English, French,
Russian and Spanish.

That would be very nice, actually. Although a bit redundant, as this information is already present in the six “name:UN:<iso>” tags.

But there is no perfect solution, and as mentioned, most database
users will want to pick a localized name of the form "name:<iso>=" so
these tags should be added.

I’m sorry, but I still have issues understanding why it would be so harmless… just to remove the “name” tag (in the case where there is no main local language). No information would be lost as all the “name:<iso>” (and its variants) would be still there. It would be up to the renderrer to have to make a choice. It looks much less ad hoc to me: OSM is before all the database, not its renderrers. (Again, amongst OSM’s principles, I believe that there is a “semantic first, not renderring” one.) I would understand if there would have been some well-used renderrers that assume a “name” tag for large objects, but it doesn’t seem to be the case from this discussion.

Adding a “name” tag to a place with no local name seems artificial, and as you have seen, raises quite a lot of tensions because it implicitly imposes the assumption that there should be one main language… and this assumption seems so far away from the principles of OSM. As Oleksiy Muzalyev said it very nicely: “Translation is becoming the true international language”.


By the way, I’ve seen quite unusual changesets related to this issue. I’m linking some here, as I think that it illustrates the issues of the discussion in a more concrete matter:

There is an edit war here: https://www.openstreetmap.org/node/424311641/history Basically, there are some people insisting that the “name” tag of the Maldives be in English instead of the local name “ދިވެހިރާއްޖެ”. This is very strange to me: in this case there is a clear local language, but some people still insist in having it in English. English is locally recognised, but it is not the official language. I’m sorry, but it’s difficult not to see that as English imperialism: people wanting to impose English locally without any reason. I furthermore notice that changeset relative to Esperanto are prompt to trigger ban policies, but English-related not that much: there seems to be an asymmetry here which doesn’t feel like the values of OSM.

Speaking of which, some reverts are done in the name of “Esperanto vandalism” while the situation is more complex than this. For instance, this revert: https://www.openstreetmap.org/changeset/77883111 The initial changeset didn’t updated “name” tags from English to Esperanto, it just removed them, and added localised notes “<tag>:eo”. These additional tags has been removed because of the revert. I fully understand that one shouldn’t remove the “name” tag until we have set up this discussion here, but with such as revert description, it seems as if the main issue of the original commit was to add localised tags Oo Please don’t use such changeset description unless the original changeset really did just update a bunch of “name” tags to Esperanto for no apparent reason.


Anyway, as Pierre Béland yesterday evening said it very well: let’s be positive, the new year is coming ☺

Cheers,
Martin.


_______________________________________________
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Reply via email to