I think this is a really good way to express my ID philosophy as well…thanks for writing it out!
I definitely disagree with “If you can acknowledge a possibility that it could be sp. ‘xxxx’, then you should not explicitly disagree.” In many cases, there’s a very small chance that almost certainly incorrect ID’s could be correct (released pet, etc.), but that very small chance is not a good basis for making an ID. It could apply in almost any situation.
I think a good suggestion as to how to disagree comes from the wording iNat itself gives: “Is the evidence provided enough to confirm this is sp. ‘xxx’?” If there isn’t enough evidence, people should feel free to disagree and should provide a decent reason why. In many cases, the RG genus level record is still quite useful.
I do think it requires a high level of expertise in a given taxon to make that call though.
Here’s another… we have a National Great Kereru Count, to log observations of our native wood pigeon. Iconic bird with little chance of being mis-ID’d as something else (assuming it is seen within 50m), plenty of literature and media to promote the week long event and it is for only the one species, so I would be fine with confirming on those.
Yeah the problem with this is it only really works for either high level experts or really obvious cases like super blurry ones. A lot of times I can’t tell and someone else can. And people have been getting weird and preachy about photos that aren’t pretty but are still perfectly valid. For that reason i think an explicit disagree based on “insufficient evidence” should only be done in very certain cases.
Folks have been surprisingly pleasant about the Junco clean-up - I think part of that definitely comes from giving some explanation as to why the taxon is bumped back up to species :)
I’ve unfortunately been blocked by two other users for correcting their species-level IDs (not bumping them back to genus but actually correcting an incorrect Research Grade ID to the correct species). Another topic entirely, but I still think those abuses of the ‘block’ feature are a short-coming of iNat that allows users to hack the community ID process by shielding their observations from correction.