A “case-by-case” approach has emerged, allowing “what the local people say about their land” to evolve into many different approaches. This is often a single boundary to represent “the land and the people” (together), tagged boundary=aboriginal_lands, with additional specific tagging on any specific datum regarding language and other local tagging practices. Other possibilities could include multiple (multi)polygons, and/or “a node to represent a village,” or many others. Again, it is best practice for OSM to represent what the local practice is as specified by “the people of the land.”
I find this very vague and I don’t feel that “it is best practice for OSM to represent what the local practice is as specified by “the people of the land.”” is backed up by the actual best practice guidelines - it seems to contradict Verifiability to me.
I previously started a discussion on the talk page but think this is a better place for it.
We’re not robots and I think vague rules sometimes do have an advantage because they can be interpreted with the necessary caution and respect for a concrete case. Trying to write concrete rules in this case would risk that someone without any tact would overrule a local community with some “you are wrong, here’s the wiki proof” and cause a grievance, all just in the name of rather pointless uniformity.
Is there a concrete, or even imagined, case for your thoughts?
The language on this page is not particularly clear. The U.S. community does want to have clearer guidance on how to model reservations and other native lands in the database, as well as their subdivisions. However, as with any other kind of boundary, the definitions aren’t going to be as cut-and-dry as a simple interpretation of the “on-the-ground rule”.
The actual current practice in the database is not as squishy as the documentation may make it seem. Basically, every formally designated boundary is mapped as a boundary=aboriginal_lands relation, including reservations, Oklahoma tribal jurisdictional areas, Hawaiian home lands, and off-reservation trust lands. So are all of their formal administrative subdivisions that don’t correspond to populated places, such as agencies of the Navajo Nation and districts on the Pine Ridge Indian Reservation.
Conflating all these concepts into a single undifferentiated boundary=aboriginal_lands tag is far from ideal. Unfortunately, we haven’t reached an agreement on what to replace it with, if anything. If you have a particular case or use case in mind, maybe you could help move the conversation forward?
I think you might be somewhat confused about what my point is. I’m not saying anything about how we should tag something, I’m saying that when it comes to what we cover we should restrict ourselves to what’s objectively verifiable (which would include anything with formally designated boundaries).
As far as I know, the boundaries we’ve mapped are formally designated boundaries. Most of these boundaries are national boundaries established by treaty. Signage varies for reasons outside the tribal government’s control, but it’s very important to them that their boundaries be respected, including on maps. As these are all enclaves, omitting them would misrepresent the land’s jurisdiction, even where signage is lacking.
One caveat is that off-reservation trust lands are technically private property lines, which starts to get into a gray area for OSM. However, many are governed together with reservations in practice.
I can see how someone would read the passage you quoted from the wiki and come to the conclusion that it advocates for mapping traditional homelands or something subjective about local identity. That’s not what we’re doing. We’ve been leaving that to other projects like Native Land Digital that can handle the topic with the necessary nuance.
If you’re accurately describing policy, then I have no issues with the policy, but I do think the article in question should be rewritten to make it clear that it’s solely about formally designated boundaries.
Our wiki are malleable. I welcome more-accurate updates to the passages in the wiki to which you refer. Minh calls them “squishy,” and that’s about right. So, “tightening up” the language would be good. Full disclosure: as the author (sometimes original, sometimes with the collaboration of others, in other states / jurisdictions where things can be wildly different than what I’m used to, yet we strive to accommodate all in OSM) of much ado about admin_level wiki (in the US), the text there regarding Native American / Indian reservations is what might be called an early draft: a dart that hit the dartboard, at least. I think I hear @eldomtom2 saying “let’s improve this.” If you can hit closer to the bullseye with a re-write, that’s certainly OK by me!
Depending how this discussion about political subdivisions of reservations goes, there may or may not even be a place to discuss the reservations in this article that’s purely about “administrative level” tagging. The broader article about boundaries in the U.S. could have more to say on the subject, but for now it just links right back to the article about admin levels.
Would that be support for the idea I floated for boundary=indigenous_administration with admin_level=*? How do we reach consensus on replacing the status quo with that, so that we can fix the wiki? Or is the suggestion for the wiki to prescribe a change this time around?
I also offer support for Minh’s “floated” idea. We’ve gotten a lot of excellent, much-more-clear work done with tagging distinctions apart from admin_level=* using boundary=* tagging. Similarly, boundary=indigenous_administrationis a logical extension to this key which can push ahead such syntactic growth into a semantic space where it can offer similar distinctive clarity. Tag well!