Turkey's biggest cities are not in the dataset

[decodemyproject]

For example, Antalya and Istanbul are missing. Two major cities that should definitely be there.

[dosubot[bot]]

Hey @decodemyproject! I'm here to help you with any bugs, questions, or contributions you have. Just waiting for a human maintainer, but I'll do my best to assist you in the meantime.

I found similar issues that have been discussed and resolved:

1. 2 major cities from Turkey missing (Ankara, Istanbul) - This issue is closed, and the cities were confirmed to exist in the dataset.
2. Istanbul is Missing from List of Cities in Turkey - This issue is also closed, and Istanbul was confirmed to be in the dataset.

The major cities like Antalya and Istanbul are not missing from the dataset. They are listed as provinces:

• Antalya:

  {
    "id": 2169,
    "name": "Antalya",
    "state_code": "07",
    "latitude": "37.09516720",
    "longitude": "31.07937050",
    "type": "province"
  }

• Istanbul:

  {
    "id": 2170,
    "name": "İstanbul",
    "state_code": "34",
    "latitude": "41.16343020",
    "longitude": "28.76644080",
    "type": "province"
  }

You can find these entries in the dataset ^[1].

To continue talking to Dosu, mention @dosu.

[ChristianGreinke]

According to Wikipedia a.s.o. Antalya is a province of Turkey and at the same time also a major city. Please add Antalya to the cities subset of Turkey.

[dr5hn]

Fixed: https://github.com/dr5hn/countries-states-cities-database/commit/531c01e3a7819e13a8b26fbbba6ff4e4276cae3a

[oytunyuksel]

Hi @dr5hn ,

I noticed a significant issue with the Turkey cities/districts data following recent changes. There are incorrect entries particularly related to Metropolitan Municipalities (Büyükşehir).

In Turkey's administrative structure, Metropolitan Municipalities (Büyükşehir Belediyesi) have a special status where:

1. The province and central district share the same name
2. The central district is not listed separately as it's administered under the metropolitan municipality

For example:

• "Antalya" should not appear as a district because it's a Metropolitan Municipality
• Same applies for "Istanbul" and other metropolitan cities

This was correctly structured in the previous version, but appears to have been broken after recent PR changes. The current structure doesn't accurately reflect Turkey's unique administrative system where metropolitan municipalities have different organizational structure than regular provinces.

I suggest reverting the recent PR that caused these changes to restore the correct administrative structure.

[decodemyproject]

For my take on the matter, while administratively it may be accurate not to list Antalya separately as a city, geographically, it makes sense to do so. Antalya is a well-known, distinct urban area, and omitting it could lead to confusion, especially for purposes like mapping, travel, or general geographical analysis. Personally, I use the cities dataset for geographical purposes so omitting Antalya and Istanbul as cities makes no sense in practical, real-world use. Geography matters just as much as governance structures in this regard.

[ChristianGreinke]

I agree with you.

decodemyproject @.***> schrieb am Fr., 18. Okt. 2024, 00:08:

For my take on the matter, while administratively it may be accurate not to list Antalya separately as a city, geographically, it makes sense to do so. Antalya is a well-known, distinct urban area, and omitting it could lead to confusion, especially for purposes like mapping, travel, or general geographical analysis. Personally, I use the cities dataset for geographical purposes so omitting Antalya and Istanbul as cities makes no sense in practical, real-world use. Geography matters just as much as governance structures in this regard.

— Reply to this email directly, view it on GitHub https://github.com/dr5hn/countries-states-cities-database/issues/849#issuecomment-2420685571, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNE4ARYZZA7EWTXFH34EBLZ4AYNRAVCNFSM6AAAAABN6PQ2WWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRQGY4DKNJXGE . You are receiving this because you commented.Message ID: @.***>

[oytunyuksel]

Hello,

While this approach might make sense for your use case, it's not correct to generalize it. For example, I live in Istanbul, and no one living within Istanbul's provincial borders would ever say "I'm going to Istanbul" because there isn't a district called "Istanbul" within Istanbul. The same applies to all metropolitan municipalities in Turkey.

Therefore, adding a non-existent district as if it exists is incorrect. If we're going to use the province and district data from this repo as a reference in any project, the data should reflect the official administrative structure. Even for geographical purposes, data that doesn't reflect the administrative structure can be misleading. For instance, it could lead to errors in address selection, official procedures, or local services.

The root issue stems from Turkey not having states, yet Istanbul appears as if it were a state-level entity. However, this isn't unique to Turkey - it's the same in all countries without states in this repo. The approach is based on the "administrative level" - in countries without states, the state area shows cities, and the city area shows districts. When you accept this structure, there's no problem.

If geographical usage is important, a separate "geographical_regions" or "urban_areas" dataset could be created. However, the current state/city structure must reflect the administrative organization accurately.

[ChristianGreinke]

Thank you for your explanation. Then I agree with you and we need to add that marker.

Oytun Yüksel @.***> schrieb am Fr., 18. Okt. 2024, 09:25:

Hello,

While this approach might make sense for your use case, it's not correct to generalize it. For example, I live in Istanbul, and no one living within Istanbul's provincial borders would ever say "I'm going to Istanbul" because there isn't a district called "Istanbul" within Istanbul. The same applies to all metropolitan municipalities in Turkey.

Therefore, adding a non-existent district as if it exists is incorrect. If we're going to use the province and district data from this repo as a reference in any project, the data should reflect the official administrative structure. Even for geographical purposes, data that doesn't reflect the administrative structure can be misleading. For instance, it could lead to errors in address selection, official procedures, or local services.

The root issue stems from Turkey not having states, yet Istanbul appears as if it were a state-level entity. However, this isn't unique to Turkey - it's the same in all countries without states in this repo. The approach is based on the "administrative level" - in countries without states, the state area shows cities, and the city area shows districts. When you accept this structure, there's no problem.

If geographical usage is important, a separate "geographical_regions" or "urban_areas" dataset could be created. However, the current state/city structure must reflect the administrative organization accurately.

— Reply to this email directly, view it on GitHub https://github.com/dr5hn/countries-states-cities-database/issues/849#issuecomment-2421639419, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNE4ATMZDM4IWRJFTAI5MTZ4CZW7AVCNFSM6AAAAABN6PQ2WWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRRGYZTSNBRHE . You are receiving this because you commented.Message ID: @.***>