I don’t want skin be separate, I want fur and scales to be, similar to feathers. And it was proposed ages ago.
I would also like the AI to help match transverse sections, longitudinal sections, surface cells, reproductive features etc… in microscopy work such as algae. Currently the AI would probably not be able to determine that the organism is sectioned and photographed under the microscope.
I add ‘Series’ of observations showing Underwater photos’ accompanied by microscopy photos. How can I train AI to recognize that these are of the same organism?
Could iNat assign matata data to the observation such as Photo1 = annotation value?
Therefore the most essential option at the moment is acknowledging that there are multiple photos included in the submitted observation.
- Photo2 - Microscopy - Plants (Including Phaeophyceae) - Surface Cells
- Photo5 - Microscopy - Plants/Algae (Cross Section)
Other = - Microscopy - -plants - Reproductive features
- Microscopy - -plants - longitudinal section
- Microscopy - Bacteria - Colony / Individual
iNat should recognize that reordering or deleting photo’s from the observation should not impact the annotations.
There was a summary already, but it was getting harder to find, so I moved it to the top of the topic.
I want to add a “fragment” annotation to mollusks. (I don’t think I ever said that outright so I’m doing that now)
concerning “shell”, I am content without it. Shelled mollusks are almost never found still rotting so “dead” basically means “shell” for most intents and purposes.
Are people going to interpret “shell” as strictly mollisc shells though?
They might see it and think it fits for a cicada husk, egg shell or exoskeleton molt.
For those, it might have value and not automatically denote dead.
I don’t understand your concern, sorry. what do you mean
I assume people will interpret it in the context of whatever taxon to which the observation belongs.
just to clarify,
I am advocating against “shell” (which has not been implemented) exclusively for mollusks in favor of the use of the present function “dead” instead and referencing the summary at the top where it has “shell” under mollusk requests.
No, I understood that. 
I was just thinking of shell as an annotation for multiple taxons, nd wondering if it’s generality would make it a good annotation category or not.
I was mostly just spitballing, and meant to post as a reply to the thread, rather than a reply to you, specifically.
Well, given that only a relative few plant species are dioecious, the “sex” annotation should have nonbinary options, too 
There’s at least one feature request for that.
I’ll see if I can find the link(s).
It would be very useful to be able to annotate anamorph and telemorph stages in fungi.
I guess I didn’t realize how many viable “fragment” mollusk observations there are cause now I REALLY REALLY REALLY wish I had the button to click. I’ve been making a list in case it gets implemented and then I thought to myself “I should post some of these on the forum so people can get an idea of what I’m talking about”. so here is some of that list (actually this is the whole list 
in my defense I’ve only spent 20min looking for them and most of them are different species) I easily could have made this list in 2min if I only used channeled duck clams cause 80% of the time they are broken, quite often to the point where you could call it a fragment.
https://www.inaturalist.org/observations/153576030
https://www.inaturalist.org/observations/109047104
https://www.inaturalist.org/observations/151608238
https://www.inaturalist.org/observations/150367089
https://www.inaturalist.org/observations/153334124
https://www.inaturalist.org/observations/152452466
https://www.inaturalist.org/observations/128151900
https://www.inaturalist.org/observations/149679971
https://www.inaturalist.org/observations/103714538
https://www.inaturalist.org/observations/94030656
https://www.inaturalist.org/observations/149178417
https://www.inaturalist.org/observations/142767918
https://www.inaturalist.org/observations/150252949
to repeat my request
term: “fragment” taxa: Mollusks values: goes under “evidence of organism”
I’ll stop pushing this now unless someone else talks about it, I guess I just really wanted it to happen
I would add the “cocoon” Evidence of Presence. There are multiple instances in which you can only see de cocoon, and not the chrysalis or even if there is something inside it .
I was going to request an annotation for photos which are underwater as opposed to specimens out of water (e.g. low tide or caught fish) until I saw this:
These are not photo-level annotations, so the annotation should describe the observation as a whole.
My reason for requesting it is that soft-bodied creatures like sponges, tunicates and anemones look much different out of the water. They are more likely to be observed and photographed by a beach comber than a diver, which means scrolling through a lot of observations when I’m trying to ID a species I haven’t seen before.
Maybe what we need is an annotation for expanded/collapsed or extended/retracted. Whatever the preferred scientific nomenclature is. Or is this something best handled as a project? Some examples are below.
Chain tunicate:
Submerged - https://www.inaturalist.org/observations/153419422
Dry - https://www.inaturalist.org/observations/26550796
Suberites aurantiacus:
Submerged - https://www.inaturalist.org/observations/123393039
Dry - https://www.inaturalist.org/observations/151549523
Giant Plumose Anemone:
Submerged - https://www.inaturalist.org/observations/153295443
Dry - https://www.inaturalist.org/observations/141413285
I can DM you at least 2 if you want more for your list.
That’d be great. 
I stopped working on the list after a few pages when I came across a boatload of giant Atlantic cockle fragments and couldn’t decide if it’d be easier to dig through the species list to find them or use a list of url links. Isolated examples are good for that though.
concerning the poll below, I would have shut up a long time ago if I had know how to do a poll.
- FOR Mollusk “evidence of presence”: “fragment”
- AGAINST “fragment”, it is not necessary or worth implementing
- undecided
0 voters
cons:
- you’d be more likely to use “fragment” to exclude poor-quality observations than ever look for actual fragments.
- I roughly estimate It’s only applicable to at most few thousand observations. (there are 2.02 million mollusk observations on iNat)
pros:
- cephalopods, gastropods, and bivalves all have fragments. (some species are more prone than others.)
- Personally, I know I’m going to click it many hundreds of times.
- certain shells are notoriously hard to find non-fragment specimens. However, most people don’t post fragments so the ratio on iNat is more like 50% for those species.
does anyone else have reasons it should not be implemented? I assume the main reason against it is it might not be worth the amount of work it takes to implement it for the number of times it is used.
How can I mark in the metadata that I’ve uploaded a bird’s song / voice, not a trace or a photo of an organism? I can’t choose “voice”.
Users can already filter for observations with photos and/or sounds, so this type of evidence is already available. Whether a file is a photo or sound is determined on upload and doesn’t need to be entered by the user.
If you’re referring to the Evidence of Presence annotation, you would still select “organism” since the call is evidence of the organism. It’s probably not worth spending the time doing that though since all calls would be evidence of the organism. https://forum.inaturalist.org/t/what-evidence-of-presence-annotation-for-audio/40854