How to Treat Observation Where Organism ID'd or Described by Observer is not in Photo, but Other Organism is

This is clearly well-meaning, but it is clearly not what the observer was intending to represent by the photo. I understand the feeling that “no evidence of organism” isn’t literally true in this case, but I think that just reflects how hard it is to come up with exactly the right label for that checkbox. I think of it as meaning something like “no evidence of the organism that the observer claims is here”.

5 Likes

Hey @sgene, we were just trying to figure out what to do about such an observation yesterday, and with the suggestions above, I’m still not sure what to do with that particular observation. I identified an obs. that included 2 photographs of trees. One of the 2 photos also featured a wooden pole with a hole in it. The OP identified the obs. as a Red-bellied woodpecker. I identified the trees because it was the only organism in both photos.

However, the OP has now indicated that he/she was identifying the creator of the hole on the pole in one of the photos. Now, granted, that hole could easily have been made by a red-bellied woodpecker, but obviously, that is only one of many possibilities. So, can I still mark that as “no evidence of organism?” That doesn’t quite seem right because the hole technically could be evidence of the presence of a red-bellied woodpecker, it’s just not sufficient evidence to narrow it down.

Here is the observation in question:
https://www.inaturalist.org/observations/23617456

This is a rare moment where an OP of an ambiguous observation actually responds, so I’d definitely like to make sure I do the right thing for this user and any future such scenarios I encounter.

3 Likes

No you can’t, the nest hole is a clear evidence of the bird, and those shots are not the same as a random tree the bird was sitting on. The second shot was done to understand which tree it was, cause it probably may affect the bird species that can use it, though I’m not up to add it. Also, don’t link examples, please, it’s not encouraged on the forum to do that to avoid bashing, etc.

2 Likes

I think it would be better to retract the ID of the trees, since you now know that this isn’t what the observer was trying to identify.

3 Likes

OK, done.

1 Like

OK @fffffffff, thanks for letting me know.

So, should I just leave this unidentified and unreviewed then? There’s really no category I could use to identify that hole.

1 Like

Well, I personally would choose can’t be improved or leave it as that and mark as reviewed in id tab. Some people want to have their stuff to be in “id needed” rather than casual, but it’s easy to find out if the user does respond.

The question that keeps running through my head is whether I should go back through everything I’ve ever put an ID on and remove it if the observer had ID’d as a different organism that wasn’t in the photo. Or, alternatively, assume that if they didn’t complain about it they didn’t really care.

I do the “no evidence” thing without contradicting the id when it seems clear from context that the photo is being used as a “memento” of the observation. I consider that it is showing the general habitat of the reported encounter, rather than some specific other lifeforms in that habitat.

Context can differ. In some cases, the observer actually says, “it flew away but”; in other cases, their obs might be part of a bird count project even though they are only pictures of the trees that the birds were on. In rare cases an observer may make hundreds of these habitat shots such that their “standard operating procedure” becomes the context to help mark their subsequent observations.

For this particular case, it is often someone taking a picture of “a bush” and then when they type it in, that’s what comes up. So, the issue again comes down to context, this time the context of “typical mistake”. Then we can just add a clarifying, “I think you meant this instead, let me know if not”.

As for going back though, it’s not worth it IMO, just deal with any that come back around with a later id, if you see them.

4 Likes

Nests, tracks, scat etc are all acceptable evidence of the creature that created it assuming they are distinctive and can be identified.

4 Likes

This almost always happens because the observer tried to enter an ID of “bush”. I usually interpret this as the observer trying to get the plant identified, not a non-existent bird. So, add a plant ID, do not mark it as “No evidence of organism”, and if you’re feeling really patient and helpful, add a comment welcoming them to iNaturalist and politely explaining how “bush” went wrong and how to withdraw the “Bushtit” ID.

In cases where the organism the observer wants to record is clearly not in the photo, I think the correct thing to do is to comment that observations without photos are permitted, point out that irrelevant photos attached to an observation cause trouble, and explain how to delete photos from an observation.

At both extremes the observers are usually new to iNaturalist, and not really familiar with how to use it. It can be tiring to have to explain the same things over and over, but usually you only have to do it once per person.

Fundamentally, you should try to encourage people to make each observation as good as possible. In many cases the “No evidence of organism” button should only be used after an observer hasn’t visited iNaturalist for a few months. It’s essentially a vote for “this observation can’t be rescued, so identifiers don’t need to look at it again”. The cases where I would use it immediately are the ones where the observer is obviously trying to get an ID for something genuinely non-living such as a rock or a toy.

4 Likes

This is pretty much how I had understood it before.

2 Likes

I totally agree with this, but I just never thought of it including situations where the observer’s choice wasn’t in the photo.

In such situations the observer confuses the artificial intelligence of the system, after all. And thereby spoils other definitions of AI. This should not be forgotten either.

1 Like

and now I guiltily see an old obs coming up on iNat - where the observer saw a bird, and I IDed the plant. Now it is a confirmed plant obs. Oops.

2 Likes

Maybe the observer added that later?
I’ve gone back and added descriptions after the fact.
Also, I know that iNat unfortunately doesn’t notify you when some things that may affect your ID are edited (like the description, the number of photos, etc.)

3 Likes

You can withdraw it and put a comment (or even an @ mention) to the other identifiers.

3 Likes

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.