City Name Manager development

[Leoreth]

I've already mentioned that I was redoing the CNM and that's finished and working now.

Besides speed improvements (city names are a dictionary now instead of huge if/else blocks), the CNM now also supports multiple and dynamic languages for every player. What does that mean? Previously, every civ had their own city name map for cities they found and their own rename list for cities the acquire from other players. Now, both of these are associated with languages, and every player can potentially have more than one language.

For example, the Turks have Turkish and Arabic as languages. Which means that if there is no Turkish name available, they can use an Arabic name if possible. I will use this in the future to further cut down the number of maps and rename entries.

The languages available to a civ can theoretically also change depending on circumstances, although I haven't done that yet. For example, Egypt could switch to Arabic when converting to Islam (currently they have both languages in the same list which usually produces anachronisms). If you have further ideas, you can suggest them here.

Even more importantly, the new implementation makes it a lot more convenient to add missing rename entries. So if you notice a civ acquiring a city where it should have its own name for, but the name doesn't change, please report that here so I can fix it.

(Please don't submit suggestions for the city names maps themselves because they're still a pain to edit.)

 

[InsanelyApple]

How about America's language be based off whoever has the most cities in their flip zone, defaulting to English if there is no majority or no cities?

 

[Lennon789]

Which languages are associated with the Mughals?

 

[Leoreth]

How about America's language be based off whoever has the most cities in their flip zone, defaulting to English if there is no majority or no cities?

Yeah, that could work.

Which languages are associated with the Mughals?

Persian -> Arabic -> Indian in that order. They're the only ones with three languages at the moment.

The Persian language is a bit of a mess still because there's a lot of overlap between Ancient (Achaemenid era) Persian and medieval Persian right now.

 

[ales_]

I think that every civ should have every language in its list so that every plot has its own name for every civ.

 

[Leoreth]

I think it makes more sense to create a generic language that covers the whole map, and let everyone use that as a fallback.

 

[Tomorrow's Dawn]

I think it makes more sense to create a generic language that covers the whole map, and let everyone use that as a fallback.

Esperanto?

 

[Leoreth]

Haha, I meant it more as a combination of all settler maps based on the language spoken in the area.

 

[awesome]

I think it makes more sense to create a generic language that covers the whole map, and let everyone use that as a fallback.

Especially since that's kind of what happens in real life. A city can have different names based in a certain language, but some languages will just call it the same thing that another language does. And that's not even getting into how some civilizations share languages.

 

[Leoreth]

Excatly. Also, just giving every civ all languages necessarily creates a global order of preference that is not always appropriate. E.g. it might make sense for Germany to give Portuguese preference over Chinese but that would have the effect that they rename Guangzhou to Cantao etc.

 

[civ_king]

Yeah, that could work.


Persian -> Arabic -> Indian in that order. They're the only ones with three languages at the moment.

The Persian language is a bit of a mess still because there's a lot of overlap between Ancient (Achaemenid era) Persian and medieval Persian right now.

There is at least one thing seriously wrong in that post
1) There should be two sets of Indian city maps with one being Indo-Aryan (and the other being Dravidian (you could call them Hindustani and Tamil respectively), they are two separate families (for reference Hindi, Spanish and Russian are in the same family).

Map of language families:

2) Mughals should actually have Hindi as primary and Persian as secondary perhaps with Turkish as tertiary until 1500 with Arabic being tertiary after.

 

[Leoreth]

2) Mughals should actually have Hindi as primary and Persian as secondary perhaps with Turkish as tertiary until 1500 with Arabic being tertiary after.

That doesn't make sense because we would get Dilli, Pataliputra and Lavapuri instead of Delhi, Azimabad and Lahore.

There's similarly little point in subdividing Tamil and Hindi languages because there is no overlap where that would result in different names.

 

[civ_king]

That doesn't make sense because we would get Dilli, Pataliputra and Lavapuri instead of Delhi, Azimabad and Lahore.

There's similarly little point in subdividing Tamil and Hindi languages because there is no overlap where that would result in different names.

For the first there are a number of hypotheses about the original name, the second one was renamed by a Mughal ruler after himself, for the third one there is only folk history that it was called that. Urdu is close enough to be mutually intelligible with Hindi.


There is overlap, but it would take a bit of research to find the romanizations.

 

[BenZL43]

Hindi: Anuradhapura
Tamil: Anuratapuram

needs research, though. I'll help.. we have to make this CNM more detailed than the very-detailed Rhye's! hahahahaha

 

[Leoreth]

Yes, that is exactly the point of this thread. Currently, India and Tamils use exactly the same city names. If you want me to split them, give me city names to make it worthwhile.

The inverse is true for Mughals. I'm aware that many names in the Indian map are modern reconstructions (Bhagyanagar is another example I think), but unless you want the Mughals to use exactly the same city names as India, they should not use the same language. And I don't.

How similar the actual languages of their time were isn't very relevant. "Language" in the context of the CNM is just an abstract concept that allows me to link multiple civs to the same city map or rename list. The Persian map includes city names used by the Mughals so it makes sense to make it their primary language, even though the actual languages present in the map are fundamentally different (Urdu and Ancient Iranian for example). I don't know how similar Hindi and Urdu are/were, but the point is that their city maps are wildly different.

 

[srpt]

Urdu and Hindi are extremely similar at a conversational level, but they take their "higher" vocabulary from different sources, Hindi from Sanskrit and Urdu from Persian.

 

[BenZL43]

From what I read and imagine... (haven't take a look at the code yet), the ideal is each Civ should have it's own unique "language" ?

 

[Leoreth]

No, then the whole ordeal would have been pointless, because that was effectively the previous situation. Every civ had their own city map and rename list, even though there were many civs that shared (almost) exact copies.

In such a situation languages would have been 1-to-1 relations between a civ and their city names and therefore an unnecessary level of abstraction.

This situation produced a lot of redundancy and therefore trouble for me because e.g. Arabia, Moors and Iran were using the same city names, but each had there own map so I had to make every change three times.

 

[BenZL43]

For example China converts to Islam and become East Turkestan, then if there's Islamic name for Chinese mainland cities in the Arabic "language" pool, the China will use that name in the CNM?
Much more efficient then!

 

[ales_]

So, now it's easier to make changes to the maps?