The Same Individual Keeps Posting CV-Suggested Misidentifications on Others' Observations, What Do I Do?

[eyekosaeder]

I don’t really like this idea. I think an equal value to being a database for observations that can be used in scientific research is teaching people about wildlife and getting them interested in it. So I think the pros of iNaturalist’s open approach outweigh the cons. (Besides, most IDs I get still are by people who seem to know what they are doing)
Personally, I have learned a lot by way of IDing that I likely wouldn’t have learned if there was an extra hurdle to becoming an identifier. I have made mistakes, of course, and I still make them. As I am constantly monitoring my inbox though, I tend to catch my mistakes rather quickly, and judging by the behaviour of most other IDers, the vast majority of them does as well.

@jeanphilippeb

This gives me an idea for a new feature request: allow the observer to disable computer vision on all their observations.

This may be a good compromise. I do like to check the CV to learn about similar species I was previously unaware of, though, so I don’t know if I would use that feature.

It could be an interesting experiment to display the identification accuracy on a user’s profile (though I am not sure it would be a good thing to add in practice). It might also change the “gamification” aspect previously mentioned in this thread into one where quality counts just as much as quantity because it’s not just about seeing the absolute number go up.

[jeanphilippeb]

@eyekosaeder

@jeanphilippeb

This gives me an idea for a new feature request: allow the observer to disable computer vision on all their observations.

This may be a good compromise. I do like to check the CV to learn about similar species I was previously unaware of, though, so I don’t know if I would use that feature.

Update: the restriction could apply only to other users, so that the observer could still use the computer vision on their own observations.

[cthawley]

I understand the frustration with users who may be blindly IDing with the CV. However, I don’t think allowing ad hoc disabling of a key feature of the platform on a user-by-user basis is a good solution. This is likely to create lots of confusion (many users wondering if CV is broken, etc.). The CV is also used as autocomplete by many high-volume, expert IDers. Disabling the CV for some observations is likely to reduce their efficiency. As such, I don’t think the benefits would outweigh the costs.

The best solution is probably to try to educate the users as others have suggested and then, if the problem is egregious and persists, to contact staff.

[jdmore]

Hi all, since this topic had been solved, and discussion had embarked in a new direction, I split off the new discussion, and am closing this topic.