On another thread, I came across a mention that adding certain observation fields will affect how these records are handled by external tools such as GBIF:
I wasn’t aware of this - learn something new every day! Now I’m curious which other observation fields act like this and what effects do they have? I’m already aware that some of the interaction fields are picked up by GloBi. Any others that similarly affect where and how iNat data is used?
I believe fungi with the “DNA Barcode ITS” observation field get pulled into MycoMap for blast searches, but that’s the only one I know about.
that’s probably because it’s not true.
you can pick some random observations using this field, and see if they and their associated observations make it to GBIF or not. i checked a couple, and i don’t see any evidence that that observation field keeps any observations out of GBIF.
Correct, it does not affect whether something is in GBIF.
Aw, bummer, and here I thought we finally had a solution to offer for duplicates.
@annkatrinrose I don’t see any substantial issues with duplicates as it makes little difference whether a single observer shares the same observation mutiple times or multiple users the same organism multiple times. iNat isn’t a platform generating highly structured and quantitative data, so it’s mostly an issue on the side of users harvesting GBIF-data to account for observer biases, of which there are many.
There is an issue for identifiers.
Both duplicates, and ‘single subject’ spread over many obs - make more work for identifiers.
If you need a few fieldmarks to ID that sp of bat - and the field marks are all on iNat - but spread over a few obs, you need to spend extra time on each one - till you check the timestamp and realise this is One Bat.
iNat is not about distribution numbers - agreed.
New DQAs that push to Casual, as we have for Single Subject - would resolve the problems for identifiers. Casual would also work for GBIF?
There are lots of ways observers make unnecessary work for identifiers - photos of distant trees, photos of a single simple dicot leaf, beautiful close-up photos of a flower in a genus with many species with similar flowers, photos of a small blurry rodent half under leaves (I did just that), photos of blurry distant spiders on the ceiling, and so on. Duplicates are a minor pain to identifiers, in my judgment. I don’t think structural changes to iNat - things like a new DQA checkmark - will really help; what is needed is better education of observers, particularly when they first start using iNat.
Carrot and stick? If we can tip the balance from Don’t DO That to them helping to resolve the issue, and pay it forward - then the quality of iNat obs will improve, and the work load will be spread over many more iNatters.
I made an observation field to combine these into sets, e.g. group pictures of leaves, stems, and flowers of what seems to be the exact same plant with the same time/location of observation. Unfortunately I only had limited time to work with it when I created it and then sort of forgot it existed again. I need to put it into more action. I see somebody else used it to combine pictures of the same organism posted by different people, so that’s another good use for it. Often people on our hikes will take pictures of the same plant so everyone has a record. Combining them potentially leads to having more features/angles to look at for IDs.