Back in May @kueda added counts for identifications and comments to thumbnails on the Identify page. (See also here.)
The comment count seems to be working fine, but the identification count is a little inconsistent. When there is only a single ID on the observation, no ID count icon shows up. When a second ID is added that does not disagree with the first ID, the ID count shows 1. However, if the second ID disagrees with the first ID, the ID count jumps directly to 2.
I’ve noticed that the observer’s own initial ID does not seem to be included the ID counts shown on thumbnails and in other places. Could that be part of what you are seeing?
I’ve thought about that, and if it were consistent I would find it to be an odd design choice, but I guess defensible to not count the observer’s ID. But if you go to the Unknowns and add an ID, you’ll find it doesn’t count either even though you’re not the observer. And having the observer’s ID not count as long as the following IDs agree, but counting the observer’s ID if someone disagrees still doesn’t make sense. It’s also not just the observer’s initial ID. If you put the first ID and they put the second, as long as they don’t disagree, the count is still 1.
I found a new wrinkle – an observation that has 4 active IDs, but the icon shows a 2. (I’d link it, but I’m afraid someone will add more IDs and disturb it.) Of the 4 IDs, 1 belongs to the observer, 2 are species level IDs, and the last is a genus level ID. Maybe the genus ID isn’t getting counted here in addition to the observer’s?
Could it be something to do with how CID calculates? No Cid for a single id, so that would explain the jump from 0 to 2 when a disagreeing ID is put as second
As I scroll down the page I see observations that appear to have no confirming ID like this one:
But when I open them up, they actually have been ID’ed, often hours ago:
Similarly (and thanks Cassi for changing title), observations with 2 IDs are showing as one, 3 as 2, etc.
I’ve noticed the same thing.
I think it doesn’t count the observers own ID.
It never has counted it. It isn’t counting the first ID from the first user other than the observer
sorry, I was going by your description, not the image!
The example given does indeed have 2 IDs and none counted :)
(sorry, I have had a bad day… should pay more attention)
I think something changed since I first posted about this, because it seems to be even worse. Back when I first posted, most counts were just off by one – it made it seem like the observer’s own ID wasn’t counting. Which was already weird. But now I’m mostly finding the count to be short by 2 – observations with 4 IDs show 2, with 3 IDs show 1, etc. This isn’t 100% true; I’ve found ones with 5 IDs showing 2, for example. But almost none show the actual number of identifications.
I think it might be showing the number of different leaf IDs, or perhaps just the number of differing IDs? (Guessing, as I am still trying to figure this out myself!)
Even more inconsistent behaviour now:
The butterfly is showing the correct thumbnail, as one additional ID beyond the original poster has been made. The bird, however, has 2 additional IDs but is showing one
Just a nudge on this, has anyone figured out what the count is?
@kueda @tiwane @carrieseltzer @bouteloua @jdmore @charlie etc
has also been raised in this post:
https://forum.inaturalist.org/t/functionality-searching-question/10665
Staff had some discussion about this recently and the intention is for that to be the count of active IDs on the obs not created by the observer, but it uses some older code based on counts that may sometimes be inaccurate.
How would you expect it to count identifications?
I think my preference would be to count every active ID, including the observer’s. But I wouldn’t be crushed if the observer’s ID didn’t count as long as the rest of the IDs were counted correctly and consistently.
Do you think this would ever be confusing in circumstances where there are several active maverick IDs that add to the count, or several more general IDs?
I would also expect it to count every ID, irrespective of the ID type.
