I seem to remember this specific use being explicitly discouraged by staff elsewhere on the forum, but I can’t find the post so maybe I’m imagining things. My recollection is that “improved” means “made more specific” in this case, not “corrected”. So using the DQA to mark that you think an ID is incorrect is a misuse of it- “can be improved” should be reserved for cases where the ID is at a higher level than it could be (for example, it’s RG at species but you want it back in Needs ID to get subspecies input). That being said, I can’t find the previous discussion on this, so perhaps I made the whole thing up. lol
i don’t blame anyone for not understanding how the function is supposed to be used because i don’t think even the staff understand how it should be or even how it is currently being used (ex. https://www.inaturalist.org/blog/120950-revisiting-based-on-the-evidence-can-the-community-taxon-be-improved). my opinion is that it’s just bad design to begin with, and they’ll never get it right without significantly changing how it works.
By the way, it doesn’t bother me and I’m not saying it’s positive or negative. It’s just something I noticed. I can understand that this bothers some people (“what is the point of yet another ID?”), but it does have a clear benefit: if earlier identifiers, for some reason, disappear from iNat or just retract their ID, the observation will still be RG thanks to your extra ID.
Yes, to some extent it’s kinda pointless (I’ve said as much in the forum more than once). In my case, there is a very specific reason why I do it which has to do with how I extract data from iNat. If iNat provided better tools for keeping track of work queues, I might not have to do it this way. But that’s a different discussion.
I think you might be misinterpreting what “improving the community taxon” means. Or maybe I’m misinterpreting it, I’m open to be corrected here. How I have always interpreted it, is as a question of taxonomic specificity. So if you have an observation, where the community taxon is at the level of a genus, and it can’t be specified down to species (because there are no characteristics visible for distinguishing different species in the same genus), then you can check “the community taxon can’t be improved”; improvement means giving a more specific identification. By this interpretation correcting the wrong species level ID isn’t an improvement, because it doesn’t make the ID more precise, only more accurate.
I would also be careful with “the evidence shows species A”, it’s always your interpretation of the evidence that suggests a certain species, and clearly in all cases we are talking about, other people have interpreted the evidence in a different way, so it seems wrong to give anybody the power to singlehandedly remove RG from an observation, even when contradicting an arbitrary amount of other people.
PS: I just saw that paul_dennehy made the same point, but I think having this second explanation still clarifies the point. And this also makes it unlikely that we both fabricated exactly the same wrong explanation.
I think your interpretation sounds right. Once you see it this way, it seems plausible this is what was originally intended. So it’s more a matter of refining rather than improving.
And you’re right about the rest, too. Yes, it’s my interpretation of the evidence against three or more others’ interpretations. But I still think the pros outweigh the cons here. Adding a new ID would alert earlier identifiers, if they’re still active and have their updates turned on. If they agree with the new ID, great. If they don’t, they can just vote “no” and the observation will return to RG status. If they’re not active anymore or don’t check their updates, at least the observation is more likely to be found again by other identifiers.
I hope in the end everyone is trying their best to make IDs better and would use this feature, even if it’s not specifically intended for this, with good intentions, just like they’d add IDs to the best of their knowledge with good intentions.
I can’t agree with that. If the RG status is so flimsy that one disagreement can overturn it, that is entirely how it should work. If you are convinced the consensus is wrong, you shouldn’t hold back from disagreeing because it will take away RG status.
I don’t think the quote above is referencing disagreeing IDs (folks here don’t seem to be disagreeing about how the CID incorporates disagreeing IDs) but the good as can be box in the DQA. On most observations, even those with any arbitrarily high number of agreeing IDs that support an RG CID, a single user can send the observation back to Needs ID (i.e.,
by ticking the “Yes, it can be improved” box which will not generate any notification for the observer or other IDers in and of itself.
I see. Thanks.