Other people have had similar suggestions in the past. They are unlikely to be adopted for several reasons.
-
the computational requirements for asking identifiers “are you sure” before submitting an out of range observations are significant. (It would also have the potential to be annoying, especially with poorly documented taxa or taxa in places without many iNaturalist observations—if I was a local identification expert getting started on iNaturalist, I would be annoyed by an “are you sure” popup on many observations because they’re “out of range. And sure, iNaturalist could invest time and money into making the model better, but “improving” a model always comes with tradeoffs, and iNaturalist already has a solution to the problem you’ve noticed).
-
the newly released geomodel tool already addresses this issue (identifications are “scored” on a geomodel, and observations outside of the predicted range for a species can be filtered to and corrected, if needed).
-
interestingly, sometimes the computer vision is ahead of the scientists. The CV consistently suggested Potentilla hebiichigo as an identification in North America, where it is considered absent by all authoritative sources. This lead some iNat identifiers, myself included, to “correct” these “incorrect” identifications, which turned out to be correct upon further study. You can read more about it here: https://forum.inaturalist.org/t/potentilla-indica-vs-potentila-hebiichigo/57037. But the point is, a popup like the one you’re suggesting would have made this less likely to happen.
As to your suggestion about ranking identifications based on the identifier’s expertise, it has been brought up on this forum several times before, and has been strongly rejected for several reasons. 1, there isn’t actually a good way to measure expertise. Many top iNaturalist identifiers of a taxon have no expertise and may have many misidentifications. Many distinguished taxonomists have rusty field identification skills. Even if we did find a metric that worked, it goes against the fundamental mission of iNaturalist. The iNaturalist organization has made it clear on multiple occasions that everyone is on an equal playing field, from a platform standard—no one’s vote counts more. This may be annoying to some, but I think it’s brilliant. Experts are too quick to pull the “I’m the expert” card, and forcing them to provide evidence like the normies is certainly a good thing (and probably keeps iNaturalist one of the nicest places on the internet).