Platform: Web
App version number, if a mobile app issue:
Browser, if a website issue: Firefox 145.0.2
URLs (aka web addresses) of any relevant observations or pages:
Screenshots of what you are seeing:
Description of problem:
Step 1:Observations that do not have an idenfication in the metadata are being labelled as Capsicum annuum annuum
Step 2:
Step 3:
What was the original file name on one of those photos?
It looks like your observations are from Chile, and chile is an English common name listed for C. a. annum. I’d guess “chile” is in your file name somewhere, as the system will try to automatically ID based on the file name.
It’s supposed to exclude taxon names that are also place names when auto-matching.
Also happens whenever I upload photos taken near the town of Lama (Corsica): the iNat upload page extracts photo metadata and deduces that some pretty flowers are big fluffy Andean mammals.
The one that annoys me is the genus “Area”, a moth. Since the computer pulls that up word as a taxon name, that space isn’t sitting empty so I can miss it and so can the computer feature that scans for empty fields. I’ve posted a diversity of things as Area, though I usually catch that error quickly.
I looked a bit closer and the issue is that when the list of problem place names is created, it’s only matching against scientific names, not against common names, so “chile”, the common name for C. a. annum, is not added as a match to “Chile”, the country. (If the scientific name were e.g. Chile annum annum, then it would match and correctly remove it. Edit: actually, it looks like it has to be an exact match, so if Chile were a scientific genus name, that would match to Chile the country.)
I think the solution is either to match on both scientific names and common names when looking for problem place names, or to manually add chile to the list of problem names.
There are two issues – first is that iNat doesn’t know about the town of Lama. iNat doesn’t query Google Maps or some other database of place names, it only uses what it has in the system, and it doesn’t have Lama. You can see the list of iNat places in that area in this screenshot.
The second issue is that iNat doesn’t filter out town names, only standard places that are state-level or larger.
You could ask for Lama to be manually added to the problem name list, but there are probably people who are legitimately uploading photos with Lama in the name who are intending to match to the animal, so the thing that improves your workflow is a detriment to theirs.
This should probably also get added to the problem name list – to be honest, I thought it already had been added.
Thanks for explaining that. I don´t have many observations from Chile and I have finished uploading them now so it will not happen again for me. The tag Chile was in the photo metadata, but curiously it sometimes picked Chile and sometimes the identification in the metadata.
People living in Chile might be irritated by it though!
The file names are all DSCnnnnnn.jpg, the tag Chile was in the photo metadata.
I’ve requested chile and area be added to the list of problematic words. Should happen later this week.
Seems like it would be easier to avoid future issues and just remove that feature?
OK chile and area shouldn’t trigger IDs any longer .
Very difficult to know how many people rely on the feature and are happy with it. The few questions/complaints we get here are a pretty biased data set. I suspect the tradeoff of confusion/irritation vs efficiency is probably still quite balanced on the side of efficiency.
This topic was automatically closed after 18 hours. New replies are no longer allowed.
