search

MushroomObserver / mushroom-observer

A website for sharing observations of mushrooms.
https://mushroomobserver.org
MIT License
81 stars 26 forks source link

lat/lng based Location searches and maps #2294

Open JoeCohen opened 3 months ago

JoeCohen commented 3 months ago

Location-based Maps and Observation Searches should include all Observations whose lat/lng is within the bounds of the Location.

Tasks

nimmo Aug 7, 2024, 11:02 PM (5 hours ago) to webmaster, Nathan

Yes - this seems right.

For sure, when someone is looking for obs w/in an area, they want all of the obs that lie within the bounds, not just the ones where the observer chose that as the “location”. TODO

On Aug 7, 2024, at 12:56 PM, webmaster mushroomobserver.org webmaster@mushroomobserver.org wrote:

The emails from Huafang raise some interesting issues about the new Obs workflow, particularly the default Locality and the Locality drop-down list. I think we've made reasonable choices. Probably Huafang's issues are due to mis-defined MO Locations, or are of her own making.

But it does make me think that we need different "Observations at" links on the Location page:

  • Observations with this Location == the current "Observations at this Location": Lists Observations whose Location is this Location.
  • Observations within this Location. Lists Observations whose lat/lng is in the bounding box for this Location.
  • Maybe something that combines the above.
  • Probably something similar for the Search UI(s).
  • And maybe something similar, except #3 for maps of Observations.
JoeCohen commented 3 months ago

Nathan Wilson 5:06 AM (19 minutes ago) to nimmo, webmaster

This is probably obvious, but it seems to me that the combined result is at least close to the right option. The problem observations will be ones where the recorded lay/long are outside the locality bounds (which I think should be included) and the cases where the bounding box overlaps the bounding box of another locality, but the actual boundaries don’t overlap (e.g. the state of California and the state of Nevada). We could take out observations whose locality are not totally inside the requested locality. There would still be a problem with distinct localities that are contained in the bounding box (e.g., Reno, NV is entirely inside the bounds of California). There might be a way to get Google to help with that, but I suspect the only real solution would be to switch to polygons rather than bounding rectangles which would be great, but probably a lot of work.