#idtips — Public Fediverse posts

Another ID tip from a Real Human!

This is Violet Toothed Polypore, aka Trichaptum biforme. It grows on hardwood, and despite it's name it only has vibrant purple edges when fresh, and usually during more moist times of year, so don't discount this ID if it doesn't have the violet! Older specimens may have a green look almost like they have moss growing on them, and the edge will be light. Sometimes they are almost fully white/beige in color especially as they dry out. The underside will have pores.

When taking photos of mushrooms for ID, it is a good idea to take photos of the whole area, and then some closeups of the top, bottom, and any stem/stipe. Noting the tree (if on a tree) species is important if you can as well, as some are buddies :) For example, this was a Red Oak, so a hardwood. But, if it were a Conifer, it would be a very similar looking but different Trichaptum species! Fungi are like humans, super diverse in their varieties and genders : )

:boost_ok: And in ongoing iNat saga...
Zero response to any requests for an official statement that having our stuff fed to Google GenAI will be opt-in only. But the staff member Tony did manage to pop in to answer a question about when the live Q&A that many of us signed up for two months ago will be! He said and I quote: "We still plan to have one when we have something concrete to share and we’ll be in touch when a date and time have been determined."
(Source: https://forum.inaturalist.org/t/what-is-this-inaturalist-and-generative-ai/66140/890 )

I am just...sad-angry laughing over here.

#iNat #iNaturalist #genAI #AI #google #tech

#iNat #iNaturalist #biology #mosstodon #ecology
#nature #IDTips #ActuallyAScientist #fungiFriday

:boost_ok: I was gone for a few days and it seems the genAI and iNaturalist thread on the iNaturalist forum has...wow! I appreciate everyone who is making their thoughts about this project known, even though the iNaturalist staff is continually flagging to hide the comments you are leaving.

In the interest of fostering actual community who enjoys nature and learning what things are, here is one of the common tips I would leave for a cool lichen in this region that the computer vision almost always got wrong, so new people to the site were continually getting the wrong suggestion. A wonderful example of how even well trained models* get things wrong...and now they want to add genAI created ID tips? Ha! Instead, have a *real* ID tip from a *real* human scientist for this type of lichen!

"Despite the fact that iNaturalist suggested it to you, Usnea hirta doesn't occur in the Southeast US. This would need dissection and chemical tests to identify between the Usnea species that do occur here."

Did you like this post? If so I will do some more Actually A Scientist ID Tips!
____
*For those not aware, iNaturalist uses a trained "Computer Vision" model which is, I would argue, a good use of AI. Specific verified images of species are used to train the model so it can offer suggestions. The images used are cited for each taxon, so it is totally traceable, credits people, and follows the photo's usage rules that the person set for the image. This is much more accessible to many than expensive guidebooks or keys with jargon people don't know. However, it used to always be stated that this is a suggestion, and any ID you place should be from your own ability and not blindly trusting the model. Curiously, in the latest releases it is no longer Computer Vision, it is just being called "AI".

#iNat #iNaturalist #biology #lichen #mosstodon #ecology #nature #IDTips #ActuallyAScientist #LichenSubscribe