Let's Talk Annotations

Not sure if I read the chart right, but I’d like to see an “evidence of presence” for plants.

I don’t believe the current “Leaves” and “Flowers and Fruit” annotation are fulfilling these rolls. They aren’t exclusionary, in the sense that both categories of annotation don’t delineate whether the observation has a tree or is just the seed, or leaf.

I believe it would be beneficial to have evidence of presence as an annotation for plants. Some presence annotations I thought of off the top of my head are Organism/Tree, Leaf, Seed. I’d thought of including ‘bark’ as well. Though I’m not sure if it’s going to be as useful; ideally, I’d like it to describe observations that only contain a ‘bark’, but I don’t know if that’s appropriate.

Additionally, I would love to have a type of ‘alive/dead’ as well. I don’t imagine it would have nearly the use as presence, but I see a lot of downed trees or remnants of trees that had fallen some time ago. People take observations of decaying logs or stumps, felled trees. Some annotations: Alive, dead, decaying, dead (Unnatural). Unnatural being more niche, and for obviously cut or harvested trees, and only really useful for determining human intervention.

I use “trees” colloquially, but I think it could be broadly applied to Subphylum Angiospermae and Classes Pinopsida, Ginkgoopsida.

Apologies if I missed some sort of formatting or rules!

Since many plants die back in the winter or during dry seasons, determining whether a plant is actually dead is a non-trivial problem. The option to select “no leaves” and “no flowers/fruit” is intended to provide a way to indicate this without having to decide whether a plant is really dead or only temporarily dormant. (It is less useful, say, for certain parasitic plants that don’t have leaves at all. I don’t recall whether this annotation has been removed for non-applicable species.)

I definitely agree that the scope could be narrowed, and I appreciate the feedback!

Likely, it would be easiest to apply it to the many genera that fit the colloquial term for “tree”. If required I could come up with a handful of taxa to apply it to, and it could from there be added to taxa that require it. However, it would be less complete than broadly applying it.

I feel there’s an argument to be made for the part of the plant that remains after dieback being “dead” itself. I’m not sure if there are plants that regrow into the parts that “died” from seasonal dieback, so there might be some of those cases. Maybe the alive/dead could apply to “stems” only?

Honestly, this does feel like more of a philosophical question that I run into with Populus clones. Is a clonal stem still an ‘individual’? If a clone in a forest of clones falls, did a tree actually die? Similarly, does a felled tree that still has its greenery still count as alive?

I don’t recall whether this annotation has been removed for non-applicable species

It’s still on a number of plants that don’t have “leaves”, yes. One being Monotropa uniflora as an example. Which is why I think the annotations could be applied more broadly. I do have level of trust or faith in users who annotate, as they seem to more often than not know what they’re looking at.

The difference between a tree that has died and a tree that is merely dormant in the winter is fairly significant. It is not safe to assume that users will only use “dead” for the former and not for the latter, or that they will necessarily be able to tell whether winter deciduous trees are alive or not.

There are a number of problems with the current system of plant annotations (e.g., the meaninglessness of the “sex” annotation for a large percentage of the plant taxa for which it is currently available), but I do not see the lack of a dead/alive annotation as one of the more major issues.

Most flowering plants and ferns are not woody trees or shrubs. Very many perennial non-woody plants die back in the cold or dry season, surviving by roots, rhizomes, bulbs, or tubers. The above-ground parts grow back when conditions improve. Assessing whether these are alive or not is difficult to impossible. Therefore, if alive vs. dead were an available, it would be misapplied so often we couldn’t trust it even for those cases when it is correct. Therefore, the idea of making an alive vs. dead annotation for plants was considered and rejected – as I think it should have been.

@spiphany

The difference between a tree that has died and a tree that is merely dormant in the winter is fairly significant

I’m not denying it! I’m more so looking for a way to delineate a tree that is living from one that is clearly dead.

@sedgequeen

Most flowering plants and ferns are not woody trees or shrubs.

I did discount ferns, and would have been more than willing to build a more narrow preliminary taxa grouping! While you are correct in ‘most’. It’s near half of all known angiosperms that are considered ‘woody’. Though I can imagine the definition varies from group to group. Working in forest research, I’ve been stunned many times by what is considered woody.

Therefore, if alive vs. dead were an available, it would be misapplied so often we couldn’t trust it even for those cases when it is correct.

I haven’t been on iNat very long, but in my experience people don’t usually annotate things they aren’t confident in. ironically, to what @spiphany said about sex and plants. Populus are one of the few species that are monoecious. It is incredibly rare for me to happen upon an observation that has sex annotated, that I didn’t place there.

And again, I feel differently, and have a level of faith in people who annotate. There is room for improvement. Perhaps “dieback” has potential for an annotation itself. As I said, I don’t think the current annotations are as exclusionary as they need to be.

With all that said, I now understand this has been brought up before and discussed thoroughly! I won’t push the matter now that I’ve stated my case! I do however believe strongly in the other aspect of my request; that plants should have an “evidence of presence” annotation. That honestly matters more to me than alive/dead!

The reason this annotation exists for animals is that they move. An animal no longer present can have left marks such as scat, tracks or a burrow. There is no obvious reason to generate a parallel category for an organism that is still present.

I’m sure that’s a reason, but I always figured there were several others.

If someone is trying to ID scat or tracks of an organism, it seems useful that they can filter for this. If someone has no idea how to ID scat or tracks of an organism it’s nice that they can filter those out. I imagine something similar could be true if we had e.g. “evidence of presence: seed” as an annotation.

The reason this annotation exists for animals is that they move

As opposed to leaves and seeds which do not move? You’ll have to forgive me, I was under the impression the annotation exists to delineate observations that have the organism in it, or just signs of it. Certainly other information can be inferred, but is their primary value not in allowing filtering for those searching?

An animal no longer present can have left marks such as scat, tracks or a burrow.

I don’t mean to sound offensive, but I feel this is incredibly narrow-sighted. What I’m suggesting is just the inverse of that. A seed is the same as an ‘egg’ in terms of annotation. A leaf is the same as feather. Leaves and seeds can be places the actual organism is not. That is the reason.

I’ve seen observations of leaves on water. Are you suggesting that is the same level of information as an observation that has the actual organism in it? I don’t believe observations only containing a leaf or a seed are nearly as valuable. It’s the same for animals and insects. An observation of an track does not have the same intrinsic value as an observation with the actual animal.

How is a seed not “the organism”? I routinely ID seeds or seedlings.

Fallen leaves are a marginal case, but you can add a species-level ID if you want. Much as I do with “ripe fruit”, you can use observation fields to indicate the level of evidence if it is useful to you.

Classify it as the organism’s age, or a sign. I would love to have the life stage for plants encapsulate it. Certainly that solves the problem. I see it’s requested, so if you know how I can contribute, I’d love to know!

Fallen leaves are a marginal case

That is an understatement. There are entire months where the bulk of observations are just leaves. If you need another use case to justify it. There are plenty of observations are unattached twigs or downed branches. I imagine there’s a place for tree stumps to be used similarly ‘bones’ no? Bones under evidence of presence is used to indicate the organism’s bones.

These primarily apply to “Trees”, but I imagine if I or some one else took the time to think of better terminology, it could fit a large number of taxa.

What is the use value of recording fallen leaves if you can record the tree in a bare state? Or the use value of recording twigs that you may or may not be able to identify? Both of these forms of evidence become worse if they are at any distance from the source organism. The use value of animal tracks or scat are that the animal was at that place. If the wind took a leaf from the foot of the tree to the next hill, the tree was never at that place, and this starts to move beyond the realm of what iNat is for.

As I said, if you feel it is important, create a custom observation field and start adding values to your observations of interest.

Both of these forms of evidence become worse if they are at any distance from the source organism.

If the wind took a leaf from the foot of the tree to the next hill, the tree was never at that place

That has been my entire point. The value of the observations is far less than that of the actual tree. I’m not sure how you can agree with exactly my point, but come to an entirely different conclusion.

You’ll forgive me, are you some sort of arbiter for this thread?

Oh, I think most of us think and act as if we are arbiters for whatever topic we’re commenting on. :grinning_face:

Over the years, many people have suggested to add ‘‘Immature’’ as a life stage annotation for birds. Reasons for this addition are plenty and have probably already all been discussed. I could write a whole essay on the topic, but won’t. Rather I have a question. I cannot find the reason why, after all these years, it is still not implemented. Could anyone explain why iNaturalist has still not added ‘‘Immature’’ for birds, or why iNaturalist may even be against it? We have ‘‘subimago’’ for mayflies, ‘‘teneral’’ for dragonflies but somehow we still not have ‘‘immature’’ for birds? What is holding iNaturalist back from adding perhaps the most obvious life stage annotation?

Here are all the discussions I could find in this topic about this:

https://forum.inaturalist.org/t/lets-talk-annotations/627/126

https://forum.inaturalist.org/t/lets-talk-annotations/627/219

https://forum.inaturalist.org/t/lets-talk-annotations/627/301

https://forum.inaturalist.org/t/lets-talk-annotations/627/496

https://forum.inaturalist.org/t/lets-talk-annotations/627/820

It looks like part of the reason is because of disagreements on how it should be added:

Should

Or should iNat follow ebird and

And if the later, should the Immature annotation also include Subadult, or should it be its own separate annotation?

If a new annotation(s) is added, a lot of observations will have to be reviewed and reannotated (looks like there’s currently ~440,940 observations annotated as juvenile). They would also have to be very explicitly defined:

I oppose adding annotation categories for plants in an “evidence of presence” group because I don’t think the cases involved (e.g. fallen leaves carried by the wind) are useful enough to justify the change in the platform. If that is agreement with your point, I’m happy to hear it.

I’m not sure why you see me as treating myself as an arbiter for the thread. Each of us presents their best arguments for or against certain actions and responds to the arguments of others. I would call that a normal discussion.

I think some to greater extents than others! It takes all kinds though! :grin:

@afdexter

I’m not sure why you see me as treating myself as an arbiter for the thread. Each of us presents their best arguments for or against certain actions and responds to the arguments of others. I would call that a normal discussion.

You speak in certainties, and are incredibly dismissive of the points I’ve brought up. I’ve brought up valid use cases, and reasoning for those use cases. You’ve provided nothing but criticism, with none of the constructive part.

You’ve essentially said “No”. I’ve provided counterpoints for the concerns you raised, and you’ve even agreed with me. Even in the paragraph above this, you provide no actual reasoning for your point beyond “I don’t feel it’s useful enough”. Again, despite me providing the reasoning for their usefulness.

If that is agreement with your point, I’m happy to hear it.

Are you trolling or just being facetious? We both clearly agree that the value of one observation can be greater or lesser than another based on evidence. The only difference is you don’t feel the annotations are needed.

For what little it’s worth, I’m the primary identifier for an entire genus of trees. I’m also some one who does, and could use the annotations to filter observations. I believe the annotations are required. I think that should be more than enough reason for it to be at least considered. But what do I know, you feel differently.

Edit: The incredible irony of this ‘discussion’ is that I found this observation while identifying just after posting. I suppose that’s a whole tree to you.

Thank you, Brenna, for summing that op and providing the link. Saved me a lot of time browsing through the whole topic!

It is still not clear for me however why iNaturalist has not updated its annotations for birds at this point, after such a long time and so many good arguments. Is it really not doing anything because of disagreements on what to do? As @swampster points out, technically speaking, we can call all birds that are not adults ‘‘immatures’’, but the same is not true for juveniles. The juvenile phase is actually rather short in most birds, and the fact that some species may retain juvenile feathers even up to three molt cycles does not mean they are still juvenile. So the easiest solution is to actually just change the word ‘‘juvenile’’ to immature, or even better and more inclusive, as a slash, to ‘‘juvenile/immature’’. My question is, why is this very obvious solution dismissed by iNaturalist, and why is the term ‘‘juvenile’’ kept in place even though in many occasions the term is plainly wrong, as pointed out by many users?

‘‘Subadult’’ is an old-fashioned word for ageing birds that are almost adults but not yet fully molted into their definitve plumage. This is not a really helpful term, not a different life stage and is now largely avoided in precise molt terminologies such as the WRP-system (Wolfe-Ryder-Pyle), the HP-system (Humphrey-Parkes) or other authors on molt terminology like Howell. ‘‘Subadult’’ can easily be included in the term ‘‘immature’’ and understood as such. The same is not true for juveniles: juveniles and immatures are two different life stages, which is precisely what the annotations are about, right? A slash may work just to separate them all together from fully grown adult birds. We now have many bird ID’ers who don’t feel comfortable annotating immature birds, because the bird they are looking at is neither juvenile nor adult, and so as a result you miss annotations from them. What is holding iNaturalist back from making a very easy change that would suit everyone and result in more precise and useful annotations?

Joe @joecoolbrew says many species breed in their juvenile plumage, but there is a difference in plumage and plumage aspect (how plumage looks). Strictly speaking, juvenile birds cannot breed, and when birds breed they have replaced at least the majority of their juvenile (body) feathers, but immature/adult birds may have retained juvenile feathers or a plumage aspect that resembles that of juveniles more than that of adults (for example delayed plumage maturation in some American warblers). It is not just the plumage aspect that defines our terminology, it is the life stage of a bird that defines how its plumage looks first and subsequently how we can call it based on the plumage aspect.

If the term is changed from ‘‘juvenile’’ to ‘‘juvenile/immature’’, not a single one of the 440000 annotations would have to be changed, and all of them are (unless wrongly identified as juv./imm.) automatically correct, too. Again: what stops iNaturalist from making this simple change, one that eBird indeed has already implemented years ago (and rightly so)?

(Edit: Please note that I am not critizing iNaturalist at all and love the platform! Just asking questions trying to understand the ‘‘why’’, and making a suggestion for what I think may improve the database. My direct and honest writing style can be a personal or Dutch culture-based thing where we say things as they are. My apologies beforehand if I may come across as rude towards iNaturalist at any point in my post.)

You’re welcome!

At this point in time, I think this is probably the best solution.

I can’t really answer this, besides maybe due to the disagreements I listed.