What evidence do you have to support what you are saying?
5 Likes
If the best IDers were constantly being suspended by “jealous” curators, I and other IDers I rely on for help would spend a lot of time suspended. That has not been my experience.
The percent of errors is much higher in “Needs ID”, but there are are wrong Research Grade observations. That’s why any researcher using the data should check it over. There’s research gold on iNaturalist, but needs some smelting before use.
8 Likes
That’s what suspended users tell themselves to justify them being toxic enough to be suspened.
1 Like
I’m not a girl and I’m a patriot, not sure wht you tried to accomplish here.
4 Likes
Your plague is no loss to iNat. Bye bye
3 Likes
I started with iNaturalist as I was exploring the world around me and wished to understand more. I continue this exploration as I try my best to ply through the Needs ID pile and identifications. This is a learning process that reinforces “The more I learn, the more I realize how much I don’t know.” Sometimes this exploration leads me to explore more beyond the natural world. Today, I am exploring Logical Fallacies and Cognitive Biases. One day I will be seeing these fallacies and biases as fast as I can discern the difference between Triepeolus and Epeolus but as with everything it takes practice and time. I don’t always get it right off the bat, and need to go back and review. Time is a good cure for this.
4 Likes
I am currently dealing with the “Unknowns pile” in my county. It has been a quick job since I mostly try to assign a broad ID to the usually low quality observations (blured photos, subject far away, captive/cultivated organisms). I only give more specific IDs when I am familiar with the group.
Many of the unknowns I found are observations without photo, usually submitted through the app. It must have been discussed or mentioned before, but I believe it could be good to have the app display a warning when uploading such observations.
2 Likes
…then it’s time for November in Hong Kong (the inter-school challenge)! ;) I’m already warming up for that.
3 Likes
I try to always take into folks’ differing degrees of ability.
Some folks may only be able to do IDs and may lack the privilege to be able ( without impairment of health, or economical limitations, ect.) to perform field observations.
4 Likes
@bobmcd, read this instead https://forum.inaturalist.org/t/small-updates-to-the-community-guidelines/36410 . It’s all there, I’m not inventing anything. (yes, it seems that I got banned from the forum, QED)
Anyone who gets themselves flagged or suspended on this forum or iNat might want to reassess how they are communicating. It isn’t that difficult to keep things civil.
9 Likes
I’ve thought about the statement that a very high percentage of Research Grade observations are incorrectly identified. I doubt that’s true, but I know it varies from species to species and probably from place to place. Some of my experiences:
Phleum pratense – Needs ID are often wrong, particularly in spring when Alopecurus cause confusion. Lots of errors (https://www.inaturalist.org/journal/sedgequeen/55127-timothy-more-of-a-mystery-than-you-might-think). However, Research Grade is better and observers are becoming aware of the problems and doing better at labeling their own observations and those of others.
Oxyria digyna – nearly all the records (Needs ID and Research Grade) are correct, and were before I first checked them.
Gaultheria shallon – All out of range records have been corrected (except one from NZ where the error must be outvoted) and the nearly all records within the range are OK.
Gaultheria procumbens – Great majority are OK. There are older errors, but they’re vastly outnumbered by the correct ID’s. The plant seems to be just beginning to spread from cultivation in western Europe and in North America’s Pacific Northwest.
Arrhentherum elatius – a generic grass with the error rate among the Needs ID observations was so high I thought the RG records would be bad too, but all but one were correct.
Maianthemum in North America – “Needs ID” shows confusion among the half dozen species in the genus plus species in other genera. However, a couple good, persistent identifiers have “adopted” the genus and they insure that observations that get to RG are virtually all correct.
Carex obnupta – nearly all in “Needs ID” are correct, though there are some ambiguous or incorrect observations near the overlap with closely related species, especially Carex barbarae (a variable taxon that originated from Carex obnupta x C. nebrascensis hybrids). This problem has to be dealt with whether your study involves herbarium specimens, iNaturalist observations, or anything else.
Usnea longissima (Methuselah’s Beard Lichen) – Identifications are so bad that the Computer Vision can’t be correctly trained from these observations. Cause? Similar species (some in other genera, even another kingdom) plus U. longissima is the least common of at least three species with the common name Methuselah’s Beard Lichen.
Most really uncommon plant species that few people know – observations mostly correct because only really good botanists even think they’ve seen them.
So I still think that the data is mostly good, for species I’ve looked at. Lots of good data there, but the wise researcher will check it before analysis.
2 Likes
That’s a weird statement overall, locally we have dozens of professional botanists checking observations, wrong id rate is low enough to be seen as normal statistical error, and serious plant people are not avoiding iNat, they like it, because they get that you can change the data, it’s not set in stone, if you see mistakes, you can correct them. Also, mistakes are more common in common species, where mistakes don’t affect anything too much, mistakes in rare species are always corrected eventually.
7 Likes
I agree.
2 Likes
If someone holds, as a matter of dogma, that a high percentage of identifications are wrong, then that person would just take all your explanations to mean that you’re contributing wrong IDs. (Yes, I read hidden content.)
2 Likes
:-)
They could check – but where’s the fun in that?
A country - with all the obs at Research Grade to GBIF?
Two people have agreed for each one?
100% wrong?
ROFL - if you are going to troll, try to be convincing.
Nobody on iNat noticed gazillion mistakes?
iNat has no barriers to opening a profile on the socalled ‘iNat’ forum. iNat profile and understanding of how iNat works - absolutely not required - trolls welcome. All content from the forum is searchable on Google, when trolls are looking for a fight to pick.
1 Like
I was just responding to a commenter on my most recent journal post and I thought part of what I wrote might be interesting to discuss here. Here’s what I wrote:
“I find it hard to figure out where my ID time can be spent most usefully. Working with new observers, coaching them to make even coarse IDs? Moving as many observations to Research Grade as fast as possible? Learning obscure or complicated taxa (asters, anyone), so I can be an actual expert and clean those up? Organizing more ID-a-thons? All of the above?”
What are everyone’s thoughts on this?
3 Likes
You have to scratch where your itch is. If ‘that’ bothers you, and somebody should do something … then hop to it.
If it’s don’t know don’t care, then next …
I have built me an URL so I can catch those broad plant IDs that slip past my I do Unknowns not Needs ID. Not so many, if you catch them as they come in!
1 Like