I apologize if this has been addressed elsewhere. I came across an observation of what appears to be a life-like toy and I am wondering if/how to use the DQA in an instance where this is the case, like if someone had posted a plastic plant. Thanks!
3 Likes
The recommend action is actually an ID of Homo sapiens. (Man-made objects are evidence of humans.) This will also make the observation casual.
17 Likes
Ok, thank you!
1 Like
Two more examples were something IDed as a waterbird of some sort, but a seam-like line and writing proved that itâs actually a decoy; and a toy insect, with joints that are thick and unlike any real arthropod joints. Both were identified as human. I put âconstructionâ as evidence of presence on the decoy before IDing it as human.
5 Likes
If it is not IDed - yes homo sapiens is better.
but if it is wrongly IDed to some other taxa - then flagging no evidence, since then adding Humans will make it require two more IDers effort and not worth it (for me).
i also mark it as no evidence of organism when things like sunrise or clouds or such are uploaded - sometimes they have placeholders like meteorite, sunrise, ⌠- and IDig it with technically correct human ID loses that valid placeholder intention.
3 Likes
Clouds and sun are properly marked âno evidence of organismâ. Fake organisms should be IDed as human. Both of the observations I mentioned went to human; the decoy I tipped to human. If an ob has 1 ID as duck and 1 as human, the community ID is vertebrate; if itâs insect and human, itâs animal.
1 Like
Maybe Iâm off base, but I just donât see an issue marking observations like this as no evidence of organism. Sure humans are organisms and can be identified, but iNat is not for humans. Identifying as human just doesnât really accomplish anything in my opinion. Sure itâs technically accurate. But a little haze in the sky of a sunset picture is probably evidence of humans too. I know the value of human observations is often debated on here. But itâs pretty clearly against the goal of iNat. Iâm not convinced marking no evidence of organism for human made items, etc, is not appropriate.
5 Likes
Homo sapiens = Casual without any further effort needed.
This was super useful, thank you. Hereâs a related question: what about observations of a human holding up a phone that has a photograph of, say, a fungus or an insect on it? Probably the posterâs intention is to identify what is in the photo to satisfy their own curiosity - in this case, is the best practice to go ahead and make the ID to the best of your ability, to help satisfy OPâs curiosity, and then select date/time or location as being inaccurate in the DQA?
Marking as âno evidence of organismâ still leaves the picture incorrectly labelled as the organism the observer thought it was and the record listed under that taxon. While âcasualâ records are excluded from a lot of searches and observations with DQA other than ânot wildâ are not used in training the CV, it nonetheless seems undesirable to leave these observations with a misleading ID.
This can be corrected by IDing as âhumansâ. For those who might be interested in the sorts of human artifacts confused for living creatures, this also means all the observations are in one place, albeit lumped together with observations of people (while annotations have been disabled for observations of humans to prevent users making assumptions about the age and gender of human subjects, I think there might be a use case for an annotation of âconstructionâ or âhuman-made objectâ to distinguish these from observations of people). There is an observation field âfake organism taxonâ that can be used for the organism it was confused with: https://www.inaturalist.org/observation_fields/17235
2 Likes
This is the point I was going to make. If I look at photos of a taxon (as distinct from the up-to-12 taxon photos), the default quality grade is âanyâ - which I assume means that it will include any photos made casual by marking âno evidence of organismâ? (Just as one use case where they might be found, misleadingly.)
I always leave a comment explaining that man-made artifacts are âhumanâ on iNat, in case it was an honest mistake by a newbie, or a joke. If one or more identifiers have actually been fooled, I give a longer explanation, as in this example that I came across recently:
https://www.inaturalist.org/observations/161703640
In this case the original IDer (who was not the observer) has withdrawn their ID. You can see from the thumbnail how easy it was to be fooled!
p.s. itâs still sitting at âlifeâ
1 Like
The only case helpful for is when users manually search or data download of casual set of such X taxa, whereas filtering such wrong casual data within X with no evidence flag is as simple as adding a filter of &fails_dqa_evidence=false
and if one is to be exhaustive in this, most casual observations have high error rate anyway in their initial taxon picks and no IDers correct them as furiously anyway, yes ofc some IDers do ID and correct casual observations, but its safe to say there will always be lot of wrong IDs everywhere in that set than normal verifiable observations, so i see it still as marginal benefit of correcting these human obs IDs (which i honestly see zero value for that human ID specifically) and making two more IDers time than simply using flag.
here is a long discussion on it https://forum.inaturalist.org/t/when-observations-are-photos-of-photos/5610
i dont know if there is consensus on that from iNat itself.
1 Like
exactly. if one is to pick another correct ID, then I can say picking bacteria ID on those observations people adding it as human IDs is still valid because yay all humans have bacteria. so, it is also technically correct ID and there is no reasonable strong defense then whether that ID should be human or bacteria eventually especially when there is no prior/wrongID from observer.
marking No evidence is hassle free and simply escape this paradox lol.
but i thought of a wacky reason: when some suddenly got curious and wants to analyse what kind of non-life+human evidence things does iNat users upload, then adding human ID will readily filter this set 
maybe ignobel worthy lol if there is clear patterns and analysis?
Thanks - now Human - and tipped out of Pre-Mavericks.
Thereâs a similar case about this but the IDs are borderline trolling? I did ID it as Homo sapiens though
[link removed by moderator]
For what itâs worth, I donât necessarily disagree with the vote âno evidence of organism.â Indeed thatâs probably what I would have done if I hadnât read on the forum that ID of human is the accepted method.
1 Like