Suggestion: Change Obscured Observations Display

[Chris Kratzer (humanbyweight)]

First off, I'm sorry if any of these suggestions have been mentioned before - I haven't gotten the hang of searching through this forum yet. 

Anyway: I would like to propose that the way obscured observations show up on maps be changed. Right now they appear as a circle randomly assigned a coordinate within a box 0.2 degrees square that contains the actual location. 

I think this is a decent system, but it does have some drawbacks:

1. It takes a long time to load maps with a lot of observations
2. It makes maps much more difficult to read, and you just end up with huge blocks where active users live (see: example)
3. Inexperienced users could think you've trespassed on their property when you haven't (probably not a real problem)

Instead, I think obscured observations should be a faint outline around the town (or county) the observation is from. There could even be a scale, where the rarer the species, the more general the location.

Alternatively, I would be satisfied with a button that turns off obscured observations on maps.

I really just want the maps to look a little cleaner. What does everyone else think?

[Charlie Hohn]

you can filter them out of some maps, but i wish that were available for more maps. I also wish for range maps they didn't show up after a certain zoom and i wish that when zoomed further out they showed as a different icon. You have to zoom in pretty close to see what is obscured and what isn't, close enough that the maps get messy and weird first. 
Coloring in the grid, town, county, etc would be a nice way to do it, or perhaps more secure and easier to program just coloring in the grid with a color that gets darker with more observations, without the scattered points. like Ebird does when you are further zoomed out. Maybe that is hard to program. Another idea that has been proposed that could work is just setting up a grid and putting all the obscured observations for each rectangle at the same point. That way when there are 50 obscured observations in some place, like many users have around their home, it just puts one pin in the middle of the square or whatever. Probably easier to program. But that might be more readily read as the real location when it isn't?

--
You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist+unsubscribe@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at https://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

--

============================
Charlie Hohn
Montpelier, Vermont

[Chris Kratzer (humanbyweight)]

Considering all the other features the iNaturalist team have successfully created, I doubt it would be very difficult to implement. Obviously I don't know for sure.

I don't really like the idea of a single point in the center of the square. Like you said, that's just as easy to misinterpret. 

Another problem with squares is endangered species along large habitat boundaries. If a square only includes a tiny bit of land along the coast, and your observation is a plant, it's pretty easy to figure out where you found it (see: example). Using political boundaries instead could potentially help with this problem. 

In my opinion, the coolest way to address our issues with the map markers would be to give users some agency in which boundary they wanted their obscured observations to count towards. It could just be the dropdown menu of locations, and - like you were describing - those observations would show up as increasingly shaded regions, where smaller areas become opaque quicker than larger ones. 

On Saturday, April 7, 2018 at 2:06:22 PM UTC-4, Charlie Hohn wrote:

you can filter them out of some maps, but i wish that were available for more maps. I also wish for range maps they didn't show up after a certain zoom and i wish that when zoomed further out they showed as a different icon. You have to zoom in pretty close to see what is obscured and what isn't, close enough that the maps get messy and weird first. 
Coloring in the grid, town, county, etc would be a nice way to do it, or perhaps more secure and easier to program just coloring in the grid with a color that gets darker with more observations, without the scattered points. like Ebird does when you are further zoomed out. Maybe that is hard to program. Another idea that has been proposed that could work is just setting up a grid and putting all the obscured observations for each rectangle at the same point. That way when there are 50 obscured observations in some place, like many users have around their home, it just puts one pin in the middle of the square or whatever. Probably easier to program. But that might be more readily read as the real location when it isn't?

On Sat, Apr 7, 2018 at 12:24 PM, Chris Kratzer (humanbyweight) <ctk...@g.rit.edu> wrote:

First off, I'm sorry if any of these suggestions have been mentioned before - I haven't gotten the hang of searching through this forum yet. 

Anyway: I would like to propose that the way obscured observations show up on maps be changed. Right now they appear as a circle randomly assigned a coordinate within a box 0.2 degrees square that contains the actual location. 

I think this is a decent system, but it does have some drawbacks:

1. It takes a long time to load maps with a lot of observations
2. It makes maps much more difficult to read, and you just end up with huge blocks where active users live (see: example)
3. Inexperienced users could think you've trespassed on their property when you haven't (probably not a real problem)

Instead, I think obscured observations should be a faint outline around the town (or county) the observation is from. There could even be a scale, where the rarer the species, the more general the location.

Alternatively, I would be satisfied with a button that turns off obscured observations on maps.

I really just want the maps to look a little cleaner. What does everyone else think?

--
You received this message because you are subscribed to the Google Groups "iNaturalist" group.

To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist...@googlegroups.com.

To post to this group, send email to inatu...@googlegroups.com.
Visit this group at https://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

[Charlie Hohn]

yeah that makes sense... I live in Montpelier VT and i obscure stuff at my home for general privacy but i am OK with it all showing up as Montpelier which is more precise than what it does now. In other cases such as a species limited to only a few habitats I might want further obscuring.

To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist+unsubscribe@googlegroups.com.

To post to this group, send email to inatu...@googlegroups.com.
Visit this group at https://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

[Chris Cheatle]

I don't play this card too often but one possible issue in this idea is outside the US, it may be hard to implement. County or town level definitions for mapping are not available in the mapping tools used on the site for many nations. Even here in "modern" Canada the lowest level available to use on Atlas maps is the province.

obscuring something to show it was found at a province or high level seems fairly gratuitous.

[Chris Kratzer (humanbyweight)]

Okay, so what if there was also an option to use a square area in the absence of detailed locations?

[Charlie Hohn]

Chris, does Canada have an equivalent of 'counties' which are bigger than a city but have dozens to hundreds in something the size of a province? I suppose even your counties would be huge in the far North where population density is low.

--

You received this message because you are subscribed to the Google Groups "iNaturalist" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inaturalist+unsubscribe@googlegroups.com.
To post to this group, send email to inatu...@googlegroups.com.
Visit this group at https://groups.google.com/group/inaturalist.
For more options, visit https://groups.google.com/d/optout.

[Chris Cheatle]

Absolutely, there tends not to be a consistent name used. But Ontario where I live has 49 of them, called 4 different things.

[Jeremy Hussell]

Technically there are 49 census subdivisions in Ontario, but two of them combine two county-like things each, so there are actually 51 county-like things. It's confusing.

See https://en.wikipedia.org/wiki/List_of_census_divisions_of_Ontario

[Alex Rebelo]

I think the current system could be improved. At first glance it looks like no-one has bothered to give accurate locations (which is quite off putting). Only later does one realize that floating points are obscured. I think a grid system is the best (such as QDS or more refined).

Regarding " If a square only includes a tiny bit of land along the coast, and your observation is a plant, it's pretty easy to figure out where you found it":

You could do a similar thing with the current system, you just need to draw the box. (unless these boxes are not on a fixed grid?)

[Chris Kratzer (humanbyweight)]

That comment was about the current system, which is why I'm advocating for the option to use non-square boundaries where they exist.