27 June, 2019

Fungus ID frustrations

edit again almost 4 years later --- it's not worth your time reading this really; it's nothing groundbreaking. Most examples are no longer pertinent or you're already familiar with them if you've spent any length of time here. i may delete it eventually


Edit: It's been 7 months and I realize as I was writing this in my overzealous newbie haze I was being a bit harsh...plz don't take anything personally...we're all working together to smooth out inat's kinks. But even so, many of my concerns still stand:

I am going to use this journal to vent about the frustrations of IDing fungi on Inat while pretending that people are actually listening to me. Here are some problems I've noticed lately:

Not enough identifiers - Animals seem to get identified the day they are observed. Plants might take a few days. Fungi will sit unidentified for months or years. I am trying to fix this, at least in Ontario where I live and am most comfortable with the local funga (think "flora" or "fauna"). If I stumble upon a fungal observation and don't immediately know what it is, I will click around on mycoquebec https://www.mycoquebec.org/bienvenue.php until I find it (I'm tellin' ya, I have no life. I consider this fun). Usually I can get it narrowed down to a few species, but then the dilemma starts. If I am still unsure of the species, do I identify it as that species? Or do I do nothing? If I do nothing, I've wasted my time and made no progress in slashing the pile of unidentified fungi and it will fall to someone else to do that, which may take a very long time or not even happen at all (If I can't figure out what it is, there really aren't too many people who can). But if I identify it as my uncertain species, then the observer might agree and promote it to research grade even if none of us are entirely sure. And if it is indeed wrong, the chances of someone coming along who can correct it are very slim as well.

Taxonomy - This is a classic problem: Inat recognizes three separate synonyms of Hydnochaete olivacea. I just reported mine as the aforementioned one because it seemed to be the most-used. Another problem: I have always called my honey mushrooms with no yellow anywhere (which is most of them) Armillaria ostoyae. It has 124 observations to A. mellea's 2,696 (Jun 27 2019). It is obviously underreported, or is something else at play here? Is it going by a different name? A lot of the time I see observations that I would call A. ostoyae that are IDed as A. mellea, and some of them are even research grade. However, most of the Armillaria observations are just sitting there as Armillaria sp. Can we not do better? Have I been identifying these wrong my whole life? Or are people just unaware that A. ostoyae exists? A similar problem I just had to deal with is the Gyromitras, G. gigas, montana, & korfii. Are they the same thing? Are they different? People are reporting G. montana from Ontario even though supposedly it is a species from western N. A. That is not the fault of the observer, who obviously is just taking suggestions from the computer, but for me, the identifier, I don't know what to identify them as because I'm confused about the status of these species. There are a lot of other examples like these that I can't think of right now...I'm not suggesting anybody "fix" these problems because I don't know what the solution is.

Computer vision - Seriously what is up with winter russulas? Half of the red russulas I see are labeled as "winter russula, Russula cremoricolor". I have NEVER EVEN HEARD of winter russula, Russula cremoricolor. Mycoquebec has it, so it must occur around here, but they don't even have any photos, so you know it's rare. Plus they say it's white. I think there are red forms on the west coast, and maybe what happened is so many people started reporting it there that the algorithm decided it was the best option for red russulas everywhere. This started a positive feedback loop in which the more people reported it everywhere, the likelier it would be to get suggested. I have been thinking of ways to solve this. Perhaps, instead of suggesting winter russula, the computer says, "This is genus russula but it is a very large and confusing genus so don't even try to identify it to species until you've got yourself a nice microscope and a bunch of fancy chemicals".

I only joined Inat 3 months ago; my views will inevitably change as I gain more experience. I suppose I will have to update this as I see fit. More to come :)

(If somebody is actually reading this, thank you, and feel free to leave your opinion in a comment)

Posted on 27 June, 2019 19:21 by n_russell n_russell | 6 comments | Leave a comment