Identification algorithm buggy - not accounting for disagreements


This is popping up in several observations according to a friend of mine, where the identification is not changing after he corrects misidentified spiders. Here is an example where two corrections has not changed the ID from an incorrect species:
(note: edited link, original link led to an observation where user had opted-out of community IDs)

The algorithm did not set up the math correctly, as it didn’t include the ancestor disagreements at all:

On the observation page, the community ID is listed as Ground Spiders (correctly), but the observation heading is still the species:


1 Like


He has opted out of the community ID, feedback from anyone other than him will not be reflected. He has to agree with the updated ID. Sorry, it is unclear which is your friend, but the submitter has done this, so they need to agree.

1 Like


Ah, sorry, the link for the first one was wrong (I didn’t double check, thought I was linking to the observation the screencaps were from,’m aware of the observation that had opted out)

Is this just a matter of if the user selects the green or the yellow button when they make their ID? That adds another layer of complexity to the algorithm, I suppose! I was trying to explain to him how it worked, and he’s trying to understand it, but I don’t quite get it myself.



Yeah, deanna54 must have hit the green button and kschnei agreed with her “non-disagreeing” ID. Re-typing in Gnaphosidae under Suggest an Identification and clicking the yellow button is necessary to enter an explicitly disagreeing ID.



Interesting. So if somebody disagrees, they are best off typing in the taxa themselves even if it’s already represented? It seems counterintuitive. I wouldn’t have thought to do that, personally.



Right, because hitting the agree button is agreeing with a “non-disagreeing ID”.There isn’t another way, that I’m aware of, to trigger the yellow/green pop-up than manually typing in the name again.


closed #8