Inconsistent or unavailable locations on explore and identify pages

This isn’t a bug, it’s how iNaturalist works, for better or worse. iNaturalist has its own “places” - either standard places like countries, states, counties, which iNat maintains, and community-curated places, which users make.

But because iNaturalist will never have as many place names as Google does, the location field on the Explore page searches Google Maps. This way you should be able to get the map to the right area. We then create a “bounding box” around the area that Google gives us. If you see “bounding box” here, then you’re not looking at an iNat place, it’s coming from Google.

image

So when you searched for South East Asia, you basically got what Google has for that term.

Now, if an iNat places exactly matches a Google place, eg Asia, then we show you the iNat place. That genrally only works for standard places, though. So when you got that Asia place, that’s iNat’s place for Asia.

Now, the Identify page doesn’t search “location”, it only has a “Place” search:

So it will only use iNat places, not bounding boxes around Google-sourced locations

What you can do is, when you have a result in Explore that you like, click on Filters then click on Identify and you can use Identify for the observations in that Explore result. And you can bookmark that if you want.

6 Likes

That’s good to know, I’ll maybe try the last part.

I’m mostly focused on Identify also. I see no Identify map, so couldn’t tell whether it uses a bounding box.

Anyway, if Identify locations are Google, can users submit corrections to Google (outside of iNat)? Because Google corrections “upstream” would reflect on iNat. Part of the confusion seems Google “should” know some locations better.

@thomaseverest When you talk about creating locations, do you mean for Explore, Identify, or both?

1 Like

You can submit suggested changes to Google, but they seem to take a while to review and I’ve only ever done things for very minor issues (such as a nonexistant address). Places can be used on Explore and Identify with URLs, but only show up in the search fields for Identify and Explore filters (not the main Explore search bar).

1 Like

What section/product of Google is that, do you know contact info?

I’ll probably look into both methods.


Screen Shot 2021-07-09 at 11.02.30 PM

1 Like

I remembered other examples, Northeastern United States and Canada, Northeastern US, the Northeast, Eastern US, etc. This makes me think similar issues affect most areas on the global map. I understand users can create individual maps, but that would take an overly long time now since it affects the map globally. That’s why I originally meant it might be something iNat itself looks into. Whether by contacting Google, or if anything can be done in site development. I could maybe also help make a few maps, especially if multiple people help, staff or non-staff.

Part of the problem is that those places aren’t distinctly defined. Take the Midwest for example, where everybody has a different definition (I know someone who thinks their home state of Minnesota is the only state in the Midwest). I doubt Google or iNat admins are interested in putting boundaries on something that subjective.

1 Like

I think some are more well-defined, like northeast states (https://en.wikipedia.org/wiki/Northeastern_United_States). Also, the maps already use most of the locations, so they can define them. It’s just they somehow defined some incorrectly, maybe unintentionally, like part of europe shown in asia search results, and india and taiwan shown in southeast asia results. So, I think it is definable, for at least most locations. It also affects a major function of the site, which is why I raised different possible solutions, and maybe see it as more of a priority. Asking Google would be a good first step though.

it seems like you are either disregarding or have completely missed some of the key points from earlier posts in this thread, and it’ll be really unproductive trying to return to this conversation or trying to reach out to Google if you don’t (or won’t) address those points.

specifically:

if you can’t differentiate between a place that’s coming from iNaturalist vs. a bounding box based on a location in Google, then you really need to go back and re-read some of the earlier posts in this thread or go read other similar threads in the forum so you can make that distinction.

please explain what you mean by "incorrectly includes Europe etc.). in your screenshot, i don’t see anything that looks like Europe is being included. i suspect whatever you’re thinking is a problem here is the same issue that you’re bringing up with North America in your new post.

again, i don’t see an issue. so if you think there is an issue, you need to provide more information about exactly what you think the issue is. screenshots pointing out exactly where the problem is would be helpful. a description of what you think should be displayed instead would be helpful.

4 Likes

what way?

i don’t see this. you really need to provide screenshots or something that explains in more detail exactly what you think the issue is.

i’m trying to help, but there’s only so much i can do from my end…

3 Likes

i don’t see this. if i look for all observations in North America, i don’t get anything in Hawaii, as far as i can tell:

sure there is: https://www.inaturalist.org/observations?place_id=1&not_in_place=6,11

for more information, see: https://forum.inaturalist.org/t/how-to-use-inaturalists-search-urls-wiki/63#heading--multiple--places.

again, i can’t reproduce what you’re describing:

so, again, i’m trying to help, but i can only do so much from my end. if you can provide screenshots or more details, that would help us help you.

4 Likes

i don’t see any true issues based on any of the discussion above. having the location/place filter next to the species box vs place filter in the advanced filters doing different things may not be the best design, but it is what it is. you may not like it, i may not like it, but:

the first reply to your original post tries to explain the distinction between the two filter boxes, and then tiwane tries again to explain the difference.

the way i usually try to explain what they covered is that you should generally use the place filter in the advanced filters. i would only ever use the location box next to the species box in cases where you can’t find the (iNaturalist) place in the advanced filters.

as far as i can tell, there’s nothing Google can do to address anything you’re talking about. so getting Google involved seems like wasted effort.

anyway, i don’t think anything i’m saying is making a difference. so i’ll leave the thread to you.

4 Likes

Which part exactly you have problems with? From what you demonstrated it seems as Google doesn’t have proper places for many places, e.g. has none for Asia, but iNat has a place for it (which is correct) and for most of other things, bigger places are usually correct, though because of polygons there are some troubles with them, still better than nothing. So what do you want to be changed from iNat end?

1 Like

From your post with screenshots there’re only a few iNat places mentioned:

  1. Mainland South East Asia https://www.inaturalist.org/observations?place_id=57241&subview=map
    Doesn’t include Maritime SE Asia as you say, but from Wiki " Maritime Southeast Asia (as opposed to Mainland Southeast Asia)". What is needed is to add another place for it or maybe it’s called differently? Then rename or prolong the name of it.
  2. Polynesia
    Just doesn’t have a place that is not of political borders.
  3. Micronesia
    This place is bugged or something as it doesn’t show obs where they actually are: https://www.inaturalist.org/observations?place_id=63039&subview=map
    There’s a different, but working place, I don’t know which one of them has more correct borders: https://www.inaturalist.org/observations?place_id=63045&subview=map
    Federated States of Micronesia is a place of political borders, it is correct and surely smaller as it’s not a whole region which includes more islands https://www.inaturalist.org/observations?place_id=10322&subview=map
  4. Explore: Asia as said no Google-defined place that you can see is abcent on Google maps without iNat:
    Asia - no borders

Though some transferring shows weird results, e.g. Google has defined borders for the city, but on iNat “explore” search uses user-made place with completely off borders (no ability to find out original one), but maybe this is another type of “place” from Google, I’m not an expert

1 Like

Probably need an answer from our staff if adding big places is possible now (and if it won’t affect the whole website work), but if you have a will you could open a separate topic and prepare list of needed places and maybe create kml files for them that you could send later, or just list probably is enough to know the face of your enemy.

The topic went slightly off the main subject, so I’ve condensed my comments to only this. The main point remains that the original examples of issues (in screenshots) were demonstrated, and it was made clear earlier which were defined by Google (boundaries on Explore) and iNat. They overall included wrong map definitions (vs most common definitions), missing locations/regions, and multiple locations/regions for what should be single ones. Both Google and iNat pertain to some, not only Google, and iNat Explore uses some Google locations so it is all related. To clarify:

“Asia map includes Europe.” Based on Turkey spanning Europe and Middle East. But, this example may not be perfect, if “Asia” is taken to include the entire greater Middle East as East Asia. .

Neither NA nor contiguous US can be searched by typing into the Explore box in the ordinary way (despite that search URLS can be used), which I consider to be a missing location/region. Re: if NA or contiguous US include Hawaii, I saw the HI orange outline show up, although it looks like no data points show in HI.

Anyway, even if we exclude the Asia-Europe and NA/US-HI examples as incorrect, the other examples seem to be correctly stated as they are above, which is what I’d like to focus on for the discussion.

@thomaseverest Just to update, I sent feedback to Google Maps for each of the applicable map issue examples (the ones with bounding boxes).