One thing I see pretty often when I suggest ID fixes is that the observer responds with a comment thanking me for the info, but doesn’t change their own ID. This is almost never a case of “Thanks for your input but I prefer my ID” and instead it’s usually “That’s great! I’m glad to know what this thing is.”
It seems there are a lot of users who don’t understand the process of revising or withdrawing IDs. This happens often enough that I created a keyboard shortcut so I can prompt people to see if they want to make a change:
Did you want to correct the ID on this observation? If you’re OK with the <correct_identification> ID, you can click “Agree” next to that name above. If you’re unsure, but accept that it may not be <observer’s_identification>, you can click the pull-down next to that ID and choose “Withdraw”.
Pretty frequently, the observer does then come back and adjust their ID in some way. This is fine and I’m definitely not aiming to pressure users to “agree” with an ID that they don’t personally have knowledge of, but I think it indicates confusion among less-experienced users about how iNat operates and I wonder whether the community might come up with some user interface improvements that would make the process more intuitive.
A couple of things that come to mind:
When a user’s observation receives a disagreeing ID, the user sees a Review ID button. Clicking the button outlines the situation with options to address it:
"You and @confirming_user identified this observation as a Passenger Pigeon (Ectopistes migratorius). @dissenting_user thinks that it is actually a Rock Pigeon (Columba livia). Here are some ways you can address the discrepancy:
Leave your original ID in place, if you’re confident it’s correct
Accept the new ID as Rock Pigeon (Columba livia), if you’re confident that the new ID is correct
Withdraw your original ID if you’re no longer sure about it
Change your ID to Family Columbidae (Pigeons and Doves), the shared ancestor that includes all current IDs, if you’re unsure about any of the lower-level IDs
The presentation and wording should aim to give users an easy way to make the fixes they want to without implying that they need to change their opinion where there’s genuine disagreement. Maybe a graphical presentation would better convey the options available.
The notification also needs to be visible enough for users to take easy action without being so intrusive as to frustrate more experienced users. A “don’t show me this again” checkbox might be needed.
Highlight observations with dissenting IDs within an updated version of the user’s dashboard. There are lots of other threads here with suggestions for how to improve notifications. As part of that process, it would be good to highlight dissenting IDs and give users an easy way to fix them. Again, a Review ID button could launch the workflow above.
I focused here on observers who receive dissenting IDs, but the same guidance could be available for any user whose identification disagrees with a subsequent ID. That might help with the class project syndrome, where lots of observations get incorrect RG IDs because friends support each other’s identifications.
Another totally different (but not mutually exclusive) way to approach this would be if “needs ID” had two subcategories, observations on which there is disagreement and observations that are simply in need of refining/confirming. That way an identifer could choose to search specifically for observations in one situation or the other.
I think it’s a very good idea. Until recently I was one of these users, didn’t know the function of ID withdraw; I found out after a more experienced user told me “Please withdraw your first ID (Attulus sp.), in order for the observation to update itself, according to the correct identifications given.”
I like the part about making ‘withdraw’ easier to find and even encourage people to do it.
I don’t think agreeing to the changed ID should be encouraged in any way. It’s already used way too often by people who just use it bowing to the (assumed) superior knowledge of that other IDer. This basically sabotages the 4-eyes principle for getting to research grade.
Newbies may simply ‘agree to the ID’ as it is the only visible way to ‘Like’ (as on FB)
They can also see agreeing politely as a way of saying thank you.
I’m glad it’s not just me experiencing this issue! Thanks for the comments and suggestions.
You’re right. I tried to edit the category, but got the error “You are not permitted to view the requested resource.” Can a forum admin help me?
I agree. I’d also like to easily find observations with dissenting IDs. (Using combinations of ident_taxon_id= and without_taxon_id= sometimes works.) If development resources are limited I’d recommend prioritizing UI fixes that might prevent this issue from occurring.
Wow! I didn’t know. I really hope the fix for that can be adding Withdraw ID to the iOS app.
I think it’s important to phrase the text describing the Agree option clearly. Any iNat user may see this message, and we want people to understand that their ID should reflect their level of knowledge (and not just use Agree as a Like button). When I suggest that Agree is one option, I’m usually doing that after I have laid out the reasons for my dissenting ID. So if the observer clicks Agree, they’re (hopefully) agreeing with my logic. For the general situation with dissenting IDs, we definitely should not imply that Agree is expected.
I’m glad to see this discussion and to read some of the clarifications. As a user, I’ve struggled with these. I find it ironic - I’m a 30+ year researcher, grad students etc., but this particular element of iNat has thrown me.
Awhile back, I finally figured it out, somewhat, with the help of one of the reviewers. I understand it better now with this discussion so again, thanks for bringing it up. I’ve shied away from providing ID’s because I wasn’t sure what I was getting into, in terms of the “accept/withdraw” elements.
I also work full time, and do volunteer work for several other groups. So I lose track of my observations, & which ones need more attention. Any sort of “easy filtering” will be nice, someday. I realize it’s all a “work in progress”.
I would suggest letting this excellent discussion continue for a bit, to help flesh out any additional nuances. Then when you feel like you have a finished feature request targeting the specific parts (there are several) of the interface where ID activity occurs, go ahead and submit a companion Feature Request post for consideration.
And yes, it is good to to keep in mind that there will always be a push-pull between educating users and making the interface and workflow too slow or cumbersome. Your idea of allowing a user to turn off the reminders is one good approach. It may also be that the preference will end up being to include this in the long-awaited “on-boarding” improvements for new users. But don’t let that stop you from proposing other solutions!
As a ~ some times here ~ sometimes not ~ user, I have been confused. Am I withdrawing the posting, or my suggested ID? Will my comments and that of others stay when I withdraw mine?
I have made an ID based on the mapping of possible species when a better visual match is not seen near my location, therefore I try to rely on trained people to help.
It was not clear to me that I had to withdraw before I could re-guess. Thanks.
Me, too. But I think it may be a long time coming.