Hi, first off this is my first post here so I just wanted to congratulate Nathaniel and the rest of the Natural Earth team for what is an amazing resource, if I ever come across any way I can contribute data I’ll do so.
–
Secondly, I’m trying to create a map where cities are shown with their local names, but I’m having difficulty with this. As far as I can see there are three fields for names.
–
The name field contains the most commonly accepted English rendition of the name, so Moscow, Munich and Cairo but recognises where newer more accurate renditions of certain city names have passed into common use, like Mumbai and Kolkata.
–
The namepar field contains local renditions such as Moskva and München where they differ from the common English rendition but nothing for Cairo and the old colonial names Bombay and Calcutta. It appears this field only contains the local alternative for languages in a Roman, Greek or related alphabet which can be unambiguously transliterated (although there are a few exceptions such as Dubayy for Dubai). It also contains accents and umlauts where applicable.
–
The namealt field contains versions of cities like München without the umlauts and accents (so Munchen) and also versions such as ‘Columbus, Ohio’ (presumably Columbus is quite a common place name in America) however it also contains renditions such as Al-Qahir for Cairo and Grung-thep for Bangkok as well as other far eastern cities.
–
So I’m a little confused as to what the definitions are of the namepar and namealt fields. Why for example, Dubai and Moscow have their local spellings in namepar, but Mumbai and Kolkata have their old colonial spellings in namepar and for Cairo and Bangkok, it is namealt which contains the more accurate rendition of the local name into English.
–
The closest I can get to creating an expression which will give me the names I want to see is to label with namepar, then namealt if namepar is blank and finally name if both of the others are blank. However this results in anomalies such as Bombay and Calcutta (and Columbus, Ohio, which looks odd on a map where you can clearly see it is in Ohio)
@McAllisterW:
Glad you are enjoying Natural Earth!
Map Feature Name Guidelines for Natural Earth (scroll down)
https://www.naturalearthdata.com/about/map-update-committee/
Local names are definitely a goal of this project, but it hasn’t been fully realized yet. Would love if you could help build this out!
name_par is for labeling places on a map like this:
name && “(” && name_par && “)”
Where && is the concatenation symbol in your scripting language.
name_alt is mostly targeted for use in disambiguation (eg: geocoding), but not for consistent display on a map. There can be multiple variants separated by comma , characters. There could be more logical sorting for this. But it’s more “you might remember this as / locals know this as”.
You also missed name_ascii which is where the plain ascii version of the name contents goes for better (simple) search support.
Really this should have a few extra columns:
name_local in latinized form, with accent marks.
local_script in native script (korean, japanese, chinese, thai, arabic, hebrew, etc).
One thing that’s prevented this from happening is UTF8 support lacking in ArcMap. But QGIS works very well for this.
_Nathaniel
Thanks for replying, I guess you could really go to town on placenames, I’d love to help.
You could have fields for the common English name as already exists, but separate fields for different types of alternative, such as different romanisations of non-roman languages (such as Peking and Běijīng), old names given by imperial powers (Bombay, Danzig), local names for places in the local languages (with another field for those languages not spoken by the majority in their own territory such as Welsh), old Soviet names (Leningrad), perhaps suffixes like the ’upon Tyne’ part of Newcastle upon Tyne.
Would require a lot of people’s imput though 
.
