On the Identify page, which source do you usually use for suggestions?

General
1
  • Observations
  • RG Observations
  • Captive Observations
  • Checklist
  • Misidentifications
  • Visually Similar
  • I donā€™t know what youā€™re talking about

0 voters

2

This is a follow-up to this post: https://forum.inaturalist.org/t/on-the-identify-page-why-does-the-source-for-suggestions-default-to-observations-instead-of-visually-similar/12292

1 Like
3

For the people selecting ā€œI donā€™t know what youā€™re talking aboutā€:

This is a page intended to make it easier to do identifications:
https://www.inaturalist.org/observations/identify

If you click an observation, then click Suggestions at the top right, under Source, you will find the options on this poll. Some ideas on what the observation might be.

Visually Similar gives you what the AI thinks, similar to the app. Iā€™m honestly not certain what all the others do, and I donā€™t know of documentation of them. Is there documentation?

2 Likes
4

If Iā€™m IDing ā€˜needs-IDā€™ I use Visually Similar the most, If Iā€™m reviewing a research grade taxon that i know is commonly misidentified I use Misidentifications because usually thatā€™s the fastest way to get the thing Iā€™m looking for

6 Likes
5

There isnā€™t, although this would be a good idea for a Tutorial.

3 Likes
6

I never used it, if you id you need to know the subject, this tab is probably useful for searching something, as @alexis18 stated.

7

For the Great Basin, where I do most of my identifying, many of the plants have few observations so there arenā€™t enough to meet the minimum requirements for the AI. We get many observations that are incorrectly identified as California plants because they are similar and have are so many more records in iNat. So Observations work a lot better in sparsely populated/observed places.

3 Likes
8

I think there is value in the principle of least surprise. And applying that here, I think it would be best to default to the suggestions tab of the identify page first presenting the species that is the best guess this website can provide, not filtered by the taxon of existing IDs which may be wrong. Which would be Visually Similar, with the taxon cleared. As the app does.

The thread on default filtering by taxon: https://forum.inaturalist.org/t/compare-suggestions-source-visually-similar-filter-by-taxon-by-default/673/12

https://en.wikipedia.org/wiki/Principle_of_least_surprise

9

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.