Note: this post is also written in bahasa Indonesia, please scroll to the first reply
Many regencies (kabupaten) in Indonesia do not have an autonomous city (kotamadya) inside their boundaries. Their regency capital is just a regular district-level settlement administratively. It relatively has no special status distinct from other districts under the law. Yet, it practically functions as the most important center in the whole regency. It usually hosts the regency government offices, the main hospital, the main market, and the densest economic activity in the area.
In daily conversation, the name of a regency or its capital is often used almost interchangeably with how people refer to an autonomous city, but just slightly lower in perceived weight. Someone might casually say they are going to Jombang or Jember the same way they would say about cities like Malang, even when one is a full city and the others are regency capitals with no kotamadya status at all. The map currently does not reflect this everyday sense of importance.
Current situation regarding regency centers
Right now there is no consistent way these regency centers are tagged. Some mappers use place=city since it is the tag that renders most prominently. Others leave them as place=town, which does not reflect their real importance and sometimes collides with other tags. The result is that a regency capital’s visibility on the map today depends mostly on which mapper happened to edit it, not on any agreed criteria.
These centers do not stand out consistently and there is no uniform editing convention across regencies because the guideline for this specific case is vague. Two regency capitals with a very similar real-world profile can end up with completely different place=* values depending on who mapped them and when. Regency capitals could also sometimes rendered larger than nearby city because of this.
Another situation with place=town
An alternative that has come up is to use place=town for regency capitals instead, but this creates a different problem. Tag place=town is already the tag commonly used for district (kecamatan) center of activity in some places across Indonesia and this usage is actually supported by the OSM wiki’s own criteria. The English wiki page for place=town describes it as an important urban centre, larger than a village and smaller than a city, with a good range of shops and facilities used by people from nearby villages. Most district capitals genuinely meet this description: they typically have their own market, government office, and act as the service center for the surrounding villages.
The complication is that this criteria based tagging is not applied uniformly. Some mappers deliberately omit place=town on district centers, either because they see it as too minor to be worth a dedicated place node, or because they are reserving place=town conceptually for something bigger like a regency capital. So even before regency capitals enter the picture, district centers are already tagged inconsistently.
It is also worth noting that place=district does exist and is used in some regions of the world. The problem is that it does not reflect settlement notability at all. place=district is meant purely for data classification, referring to a subdivision, without conveying anything about the size, prominence, or urban character of the settlement at its center. It does not compete with place=town for the same purpose, since a place=town node signals a notable place in local perspective, while place=district only signals administrative status. So even where place=district is used elsewhere, it would not solve the visibility problem this discussion is about, and place=town remains the tag that actually does the job of making a district capital stand out on the map as a real place, which is why it has become the de facto choice for district capitals in Indonesia.
If place=town is now also used more consistently for regency capitals, we risk losing the ability to distinguish a district capital from a regency capital just by looking at place=*, especially in places where both categories genuinely qualify as towns by the wiki’s own size criteria.
Situation with Indonesian guidelines in the OSM Wiki
The current Indonesian wiki guidance on this is unclear. It does not offer criteria for distinguishing place=town from place=city and only leans on population thresholds. The problem is: population of what area, exactly? District’s population or regency’s population?
This is a large part of why tagging is inconsistent today. Two mappers can both be following the population guideline in good faith and still reach different place=town or place=city values for very similar settlements, because they used population from different administrative levels. The English wiki page itself acknowledges this weakness even at a global level, noting that population thresholds for place=town vary widely between countries and regions with no fixed number. A population criteria like this is ambiguous and cannot reliably decide between place=town and place=city.
Discussion (or Proposal?)
I do not want to present a finished, fixed proposal at this stage. I would like to open this up for discussion, but here are a few points I think are worth noting.
Tagging for regency centers. I think we need a fixed, agreed way to tag regency capitals that do not have city status, rather than leaving it to individual mapper judgement as it is now. My idea is using tag capital=* on the settlement node, alongside whatever place=* value honestly reflects its size.
Solution for place=town. To avoid colliding with district capitals, I think the same capital=* key can be used there too, maybe with capital=7 for district capitals. This way place=town stays available for both regency and district capitals when their size genuinely warrants it and capital=* is what tells them apart.
Other related things. A few things I think should also be part of this discussion:.
-
Whether population thresholds should keep being referenced at all in the Indonesian guideline, given the ambiguity over which administrative level the population figure should come from.
-
Whether this should live on its own wiki page or be merged into an existing one.
I would like to hear your opinion about this. I wrote this in English to reach broader audience, but you can reply in bahasa Indonesia too.