shapironick
commented
4 years ago Quick update is that I've figured out how to use the Native Lands API
I've uploaded the code here: https://github.com/Carceral-Ecologies/Carceral-Proximity-Analysis/blob/master/Native%20Land%20Scrape.Rmd
I emailed them to see if it was ok if I pummeled their server with a lot of requests (6,703) and how much sleep time they want in between requests. I figure blowing up their server is not good data sovereignty practice.
So the things for y'all to think about (beyond is there better data source, which is still a good question) is to begin thinking about how you plan to do the analysis assuming we can get the names of the traditional Land care takers. what stats would you want? Some ideas: tribe with most facilities on their land? most facilities per unit of territory?
what data resources would you need to make these arguments?
Also there are limitations with how the API works, as it only give one name and often its multiple overlapping territories. we should figure out how they decide who to list.
Step one: Data source
Figure out what data source you will use to determine whose territory the carceral facility is located in. This may mean using the snapshot that is https://native-land.ca/. (here is their doc on how to think critically about their map). But its worth checking in with your networks to see if there is other better data.
Step two: Join the lists
If you are using Native Land it actually might be easy. They have an API here and you can enter in the lat and long and you get back the name(s) of who's territory it is. here's an example https://native-land.ca/api/index.php?maps=languages&position=42.553080,-86.473389
So with a couple weeks of work I think you could do this for not just CA but the whole country because you can do it all with programing and not do it one by one.
You can use the lat/long from the final columns in this table.
You can start reading up on how to Query APIs in R here and then ping me when you're done and I'll help you get to the next stage :)