Wikidata:Property proposal/Date of declaration as a town
Date of declaration as a town[edit]
Originally proposed at Wikidata:Property proposal/Place
| Description | The date when a settlement is declared a town |
|---|---|
| Data type | Point in time |
| Domain | item |
| Example 1 | Warsaw (Q270) -> 1323 |
| Example 2 | Mühlhausen (Q14925) -> 1135 |
| Example 3 | Birmingham (Q2256) -> 14 January 1889 |
Motivation[edit]
Do you think there is a right to exist for such a feature as: "declaration as a town" "loss of town status"? Currently, the town rank of a settlement is indicated by entering it as the qualifier for city (Q515) in instance of (P31): start time: DD-MMM-YYYY.
But not only Q515 can be included in P31, but also such: big city (Q1549591) or city in the United States (Q1093829) or town in Romania (Q16858213) etc., If we write the beginning time, then it is not clear what started at the given time: the fact that it became more than a hundred thousand inhabitants? Or that he became an American? (And many other elements can refer to city status: port city, border city, etc.)
The other ambiguous situation: a settlement that received city status in the 19th century and then lost it later, we cannot mark the period of city status there at all, since it is not even included in P31. (Or do we enter it as a city and then make it obsolete? If we don't make it obsolete, the result of a query can be spammed with settlements that used to be cities but lost this address).
If there was a property where you could specify the start and end time of the city rank, it would make the data about it unmistakable.
I know that there are administrative positions where the concept of "city" is incomprehensible (e.g. Denmark). For these countries, this proposed new feature should of course not be used, or if we want to give a city rank obtained in a previous administrative system, then it should be marked as completed. A good solution for this too is to create a separate property to mark this rank.
BTW: I checked how many times town privileges (Q28942) was specified for the significant event (P793) attribute. I found only 16, I don't think this possibility of data entry has spread in this form.
Data should only be valid if there is a source for it.
Notified participants of WikiProject Cities and Towns Pallor (talk) 19:00, 19 June 2023 (UTC)
Discussion[edit]
Comment @Pallor: "town" surely does not have uniform meaning across different languages and cultures. See en:Town#United_States for the widely differing meaning of the word "town" even within the United States. I can see why using qualifiers on instance of (P31) in this sort of case is one way to do it, even though generally I don't like to see qualifiers there. Maybe this can be handled with a different property than instance of (P31) with start and end dates? I don't like adding two separate properties to handle this use case. Maybe better modeled with a "settlement type" property or something like that? ArthurPSmith (talk) 19:14, 21 June 2023 (UTC)
- ArthurPSmith: I thought about what you wrote and I think it's a good idea. In order to avoid the repetition of the value in P31 (e.g.: American city, etc.), we must stipulate that only such value can be included here that has P31:classification of human settlements (Q98966309) (e.g.: city (Q515)). Pallor (talk) 23:30, 23 June 2023 (UTC)
Oppose instance of (P31) => town in Sweden (Q2261863) (or equivalent) with qualifiers start time (P580) and end time (P582) should cover this perfectly. /ℇsquilo 08:56, 27 June 2023 (UTC)
- Of course, in many cases the start time (P580) refers to the city rank, but there are many historical countries in which the transfer to the new country would be indicated with the same qualifier for settlements that became cities (think of the Soviet Union, Czechoslovakia, etc. or the end of the First and Second World Wars for territorial changes after peace agreements). Of course, it may be that there is no need for such a thing for Swedish cities (that is, we are sure that we will never add a qualifier to Lithuanian settlements under Swedish occupation), but my proposal also included a suggestion for this: in such cases, we simply should not use the attribute. I described these in the introduction, but you didn't respond to it, you repeated the same thing that was in the problem statement.
- If there are no other comments, I will withdraw this proposal and start the version suggested by ArthurPSmith. Pallor (talk) 09:13, 28 June 2023 (UTC)
Not done - not created per lack of support. Regards, ZI Jony (Talk) 05:30, 24 January 2024 (UTC)
