Explore filters updated

The ‘explore’ filters have changed. Couple of quick questions;

  1. Whats the difference between wild & verifiable?
  2. What does popular mean?

If you hover over these options, it shows:

That doesn’t work for wild. I assumed verifiable, or eligible for research grade was only for wild observations. Whats the difference?

I get different results if i select wild & unselect verifiable as opposed to default verifiable. Theres obviously a difference.

If the observation was wild but lacked a date, or had an incorrect location, or had no media, it wouldn’t be “verifiable”.

Wild means it’s not marked as captive. You can have an observation marked as wild that fails another entry on DQA, like this one of mine (which has no pictures or audio, so will always be casual, because no evidence): https://www.inaturalist.org/observations/309123106

So if you search for “wild” but explicitly “verifiable=false” ( e.g of mine, only that moth above: https://www.inaturalist.org/observations?captive=false&user_id=jamie-aa&verifiable=false ), you’ll get observations like that: marked wild, but missing dates, marked as wrong location, more than 1 organism between all the photos etc.

An observation can be wild, but not meet all the criteria for verifiable.

A verifiable observation is an observation that:

  • has an accurate date
  • is accurately georeferenced (i.e. has lat/lon coordinates)
  • has photos or sounds
  • isn’t of a captive or cultivated organism, or of a human

What is the Data Quality Assessment (DQA) and how do observations qualify to become “Research Grade”?

Checking captive / cultivated means that the observation is of an organism that exists in the time and place it was observed because humans intended it to be then and there. Likewise, wild / naturalized organisms exist in particular times and places because of other reasons (e.g. members of native or established non-native populations or released/escaped pets, hitchhikers, or vagrants).

What does captive / cultivated mean?

There are a lot of search queries that are not available on the website interface, but are available if people manually edit the url.

By default, iNaturalist verifiable is checked. verifiable is not in the url, but verifiable=true is sent to the API. The API I returns observations where verifiable is true.

When people uncheck verifiable, verifiable=any is added to the url. verifiable is excluded from the params sent to the API, and API returns observations where verifiable is true or verifiable is false.

The form doesn’t let people set verifiable to false. People need to manually edit the url to set verifiable=false. verifiable=false is sent to the API, and the API returns observations where verifiable is false.

observations previously treated as casual will still be treated as casual. wild=true observations have always included a mix of both casual and verifiable observations for as long as i’ve used the system.

this does pick up casual observations, however iNat has chosen to treat an unchecked verifiable box as verifiable=any, as opposed to verifiable=false. this is not new. you can see this here: https://www.inaturalist.org/observations?captive=false&verifiable=any

OK, I just didn’t pick up on that before The only other problem i see is that if I select Captive on the website explore, I have to deliberately unselect verifiable else I get no observations.

You cant select both wild & captive - verifiable should automatically be unselected if you hit captive.

i don’t really see any problem with this current behavior.

i think it’s less great that they implemented separate wild and captive checkboxes but treat them as a radio button set as opposed to actual checkboxes. (this violates the expected default behavior expected from checkbox controls.)

if they were going to have implemented more atypical behavior, i would have preferred that they implement tri-state checkboxes for wild and for verifiable, with a check representing =true, a blank representing =false, and a shaded box representing =any. then you could have better control over true/false/any, and you wouldn’t need separate wild/captive controls.

I’m really happy with the new collection project filter for casual observations. Is it possible to replicate that functionality in a regular query?
e.g.
Research Grade, Needs ID & Casual (exclude observations that are casual for any reason other than being marked as captive)

this would be the base for that: https://www.inaturalist.org/observations?verifiable=any&not_casual_excluding_captive=true

just note that the Explore page excludes spam observations silently by default, but the project page does not.

Excellent - thanks

Thanks for your help

Thanks