Show your work!

[Ian Toal]

I've seen this a few times, most recently with this observation http://inaturalist.ca/observations/1611148 . A major identification change was made a year ago, but the person who changed it did not offer any rationale for the change (even though the observer asked for clarification). It is really only courteous to explain why an identification change is being made, even if it is only a link to a source. It helps everyone learn and keeps the observation from sitting in limbo. I'm occasionally guilty of it myself when the ID is grossly wrong - a beetle identified as a butterfly - but I try to 'show my work' when I make an ID change. 

Sorry, it's a pet peeve of mine. 

Ian

[bouteloua]

It is nice to provide some info, but it's definitely at the expense of being able to complete more identifications.

Don't be afraid to ask for more information and don't be distressed if none is provided.

(user is busy, missed the notification, user is inactive for a multitude of reasons, user doesn't want to spend time repeating themselves over and over)

Community Guidelines, Things that are OK: "Requests for clarification. Requests to justify an identification in words are strongly encouraged. You don’t have to reply to such requests, but you also should not complain about them. Everyone has the right to ask, "Why?""

cassi

[Charlie Hohn]

yeah it's a hard balance. I remember at one point project noah made you cite a 'reference' for any ID and it's a main reason i stopped using the site. A lot of us just know a lot of species, and consult keys and guides for the ones we don't. And with hundreds of thousands of observations needing ID, many of which are from fled users I don't think requiring more of them is the answer. Instead like cassi says i think we need to just encourage more discussion when both people are inerested. If someone changes an ID on one of my observations, i may want more info, but i may just look again and realize they were right (or am sure they were wrong). Also in the example you posted someone DID post a very detailed description of the second ID.

It's a hard balance and i try to describe why i make a change but it depends on the user, whether i recognize them as still active, whether they responded to my input in the past, whether it's a coarse or fine level of disagreement, how easy it is to describe, and how any cases there were (for instance in one case someone made like 50 observations of buckwheat identified as the wrong species. It was awesome that they got all that data, but of course if you do that many in the same day, if there's a mistake it will affect a lot of them! So I posted an explanation for one of them but not all of them, and if you went and looked at one of the other ones you'd not see the explanation... seemed pointless to post it in all of them...

So yeah.

[Colin Purrington]

Might it be useful to modify the "Tell us why..." text to better coax details out of users who might have time to do so? Something like this, perhaps:

I agree with comments above, however, that it can be a complete pain to provide an explanation for the nth time to users who might, on average, not really care. Because I am truly grateful for any correction of ID or natural history tidbit, I sort of wish I could tweak the text myself to better solicit details. I try to make that desire clear in the initial post, of course, but if there was a toggle for stronger pleas I'd jump at it.

[James Bailey]

Cases where I won't justify an ID:

1. The user ignores all community input (although, for benefit of others, I probably still explain differences between difficult species).

2. Cases where the user clearly does not care or put effort in.

2B. A user identifies a pelican as a seagull, or two other wildly different taxa. In that case the ID is so obvious that I don't feel like I have to explain it.

3. When other users have already explained, or vouched for the "changing" ID.

Obviously, there is a lot of debate per person on what constitutes as "obvious", but I try to err on the side of caution. I explain a lot of IDs that I probably don't need to.

[Ian Toal]

Thanks for all the feedback, and I do understand why it's a PIA to provide justification all the time. I guess I need to work on the pet peeve of mine, and be a bit more flexible :)

[Charlie Hohn]

fwiw, if someone asks, or if it's someone i know is really trying to learn and wants comments, i'll pretty much always give at least a short one.

[tony rebelo]

I would find it handy if the reason copies across in the ID tool

Often I review a species or genus, and the same reason applies to half a dozen observations.  It gets a bit tedious copy pasting explanations, and I am afraid only the first few instances get explanations - they are not necessarily the most deserving.

My interest in providing explanations is strongly coloured by the amount of information (photographs and notes)  provided by the observer;  Just one photograph without any notes is not deserving of ID notes.  But if there has been a discussion about the identification, then surely adding a "show your work" should be compulsory if you disagree.

ta

T