I’m not sure how you conclude that subspecies don’t matter in the new system. In the above-cited Trillium example, although the community has not yet gotten behind the subspecies ID (which they might or might not–that’s democracy), that subspecies ID is still there and available for lifting to RG if the community concurs. That said, it is the valid–and community supported–ID at species level which is being denigrated by not rising to RG with demonstrable and suitable community support. Something is amiss if a valid species ID cannot reach RG just because either one subspecies ID was suggested or because the Observation ID was initially as broad as “Angiospermae”.
4 Likes
By the way: the title of this thread (presumably “solved”) is severely misleading.
It should rather read “Not Research Grade with two IDs at that rank”
1 Like
This thread is six years old at this point, and the original bug report that only one ssp. ID with a supporting sp. ID made it Research Grade, and that’s what was flagged.
Maybe I misunderstood what you said in your first comment, but based on what you said here:
I thought you were implying that because there is four sp. level IDs with one ssp. level ID that it would never reach Research Grade. In the old way of things, if we had one sp. or even ssp. level ID as the first, original ID, and four other people provide coarser, but not disagreeing genus level IDs, it’ll still be Needs ID. However, with the sp. and ssp. IDs, the original ID being at ssp. and the rest at sp. in the old way, it would still reach Research Grade, but now this isn’t the case:
This is why I think it’s a huge oversight from staff that this was missed during testing, now even if there is only one ssp. level ID, should it still be allowed to reach Research Grade because someone agreed it was the same species? Probably not, but that’s why I think a better solution would’ve been to split grade levels across taxons, because it can also work with coarser IDs in both a genera and higher level. I’m going to use salmonids as an example, but if someone identifies a photo as being salmo trutta but I come and say it’s an oncorhynchus species but I’m not entirely sure what species it is exactly, and one other person says it’s oncorhynchus mykiss and another person says it’s oncorhynchus lewisi, then we can all confirm that it’s at least an oncorhynchus species, and can technically be “Research Graded” at the genus level if someone used the DQA to represent that, because the community agrees it’s one of its species.
If I did end misunderstanding you, Chuck, please do correct me if that’s the case.
2 Likes
This is getting too complicated for my simple-minded thought processes. My perspective and calculation ignore any DQA influence. It’s another reason why I’ve never been comfortable with the “as good as it can be” DQA and it’s effect on RG status. I’ve long argued that it should be scrapped or reworded.
1 Like
An observation was labeled Taraxacum officinale. We know that the plant is a dandelion and that whatever it is, the name T. officinale is probably inappropriate (for technical reasons) and that unless the observation has excellent photos of the phyllaries, the styles (for pollen), and the leaves that overwintered, species identification is probably impossible. In fact, it’s probably impossible with all that because of controversy over species limits. Therefore, I added the ID Taraxacum and clicked “No, it can’t be improved” because it can’t be improved, as those of us helping with Project Dandelion have been doing. IT WENT TO CASUAL!!! Note that the two ID’s agreed at the genus level – it could reasonably have gone to RG as Taraxacum. But it didn’t. This is beyond stupid. Surely with all the names at genus level or below the observation could have gone to RG. (In this case, I removed the “No, it can’t be improved” and it reverted to Needs ID. where it will be until someone agrees that it’s Taraxacum officinale, which we can be confident it is not.)
A lesser problem: A friend is a very careful, thorough, knowledgeable botanist. He often ID’s his observations to subspecies/variety taxa that almost nobody does. I often don’t know how to ID the subtaxa, so I agree to the species and then push the observation to RG. Now if I do that the observation is supposed to sit there until TWO other people ID it to the species or to the subtaxon, although we already have a shortage of people IDing plants. In this case, the solution is obvious, since I am certain that the subtaxon ID is correct. With other people’s ID’s, though, I feel uncomfortable IDing at subspecies unless it’s something I know myself or can easily look up. (Trillium ovatum var./ssp. ovatum is the only T. ovatum subtaxon in Oregon.) So what do I do? I can simply not put an ID on it. But what will the next IDer who comes along do? Probably add a species ID. Then the observation needs to wait for two more ID’s (minimum) to get out of Needs ID although we all agree what species it is. I can “follow” the observation so I can add one of those two ID’s, forced by this new change to deal with an observation twice which could have been removed from Needs ID the first time – and it still will won’t be out of Needs ID until a third person comes along.
I understand why this change was made, but I think this is worse.
16 Likes
I agree that this shouldn’t go to Casual in the scenario you describe, but you don’t mention making a disagreement. It sounds like you can easily get it to behave correctly by using the orange button to disagree (because “the evidence provided [is not] enough to confirm this is Taraxacum officinale”), which should change both the Community Taxon and the Observation Taxon to Taraxacum.
Yes, of course. However, a number of us are trying to reduce the number of Taraxacum observations hanging in Needs ID (where one IDer after another will have to waste time seeing them) by marking them “No, it can’t be improved,” thus sending them to RG with two ID’s at the genus level (Taraxacum) or as Taraxacum officinale plus an ID of Taraxacum. Basically, Taraxacum can’t be ID’d to species from iNaturalist photos, with rare exceptions. We figure any researcher wanting to work on Taraxacum will know to search on the genus name because of the many unreliable species ID’s. We’ve moved a few thousand Taraxacum out already. With this new change, it will be harder.
I don’t see how this change makes it harder for you? If the observations can’t be identified to species, surely you should be ‘hard’ disagreeing back to genus anyway, in which case ticking the box will make it RG as before.
2 Likes
Bizarre! Using the hard disagreement puts the observation in RG while using the softer one puts it in Casual! I don’t think that makes any sense, but it’s a useful path forward. Thanks.
5 Likes
The reason it’s being made Casual when you soft disagree is because the observation and community taxon are different (species and genus respectively), so once you hard disagree, they’re the same (genus) and it’s okay.
2 Likes
I’m sorry, but that still makes no sense. I know how to make it do what I need to, though, so that’s good.
4 Likes
Do you want us to do that for all dandelions? Would I do that for your project in South Africa too?
Please no! The main problems are in section Taraxacum (used-to-be Ruderalia). Other Taraxacum sections are also difficult, but much less impossible for experts.
2 Likes
We’re doing it for the common weedy dandelions so abundant in the northern hemisphere. These are beset with taxonomic controversy and then there’s the fact that most photos don’t show the details needed to ID them. We’re trying to get the dandelions out of Needs ID, since they can’t be ID’d anyway. I don’t know what the situation is in South Africa.
2 Likes
This is why I think the community taxon is such a bad addition to the site, because in doing this it will result in thousands of observations that shouldn’t be in casual grade to be in casual, when they have accurate data in every other category. Location good, date good, wild organism, actually has photos or audio, and evidence or recent evidence. Observations that have all of these green checkmarks should not be lumped with observations that are missing one of these boxes because they aren’t equals. To put it into perspective, you wouldn’t throw away perfectly good food that has absolutely no flaws that you bought the day before, but you would throw food away that’s in your fridge if it molds or goes bad or has been in there for a long time because they are not comparable. To say a hard disagreement is okay when there isn’t full certainty that the organism actually is what it’s identified to will create many gray areas that are going to cause problems, and they will not be good problems.
Community taxon is not new. Observations have always had a community taxon and an observation taxon; the only thing that has changed is the status of the observation when community taxon and observation taxon are not the same, and then only under certain circumstances.
(I am not saying that this change is not a problem; as I have made clear, I absolutely think it is. But it is also important to be clear about what exactly is happening.)
Isn’t this precisely when a hard disagreement should be used? If there is not enough evidence to ID an observation more specifically, it should be pushed back to a level where it can be securely identified. This again is regardless of the recent change. I do not see how it is helpful to anyone to add a soft disagreement (leaving the observation listed under a too-specific ID) if one knows that this specific ID is not justified.
(I realize the dandelion issue is a bit more complicated, where a soft disagreement may seem to be a reasonable compromise between people working with different species concepts, but it still leaves the observations listed under Taraxacum officinale and thus does nothing to inform users about the existence of taxonomic debates.)
7 Likes
In the case of dandelions, I think there’s no real difference between leaving the observation at Taraxacum or at Taraxacum officinale. Both are meaningless below the level of Taraxacum. Anyone seriously research dandelions will know that and cope with it.
I view the hard disagreement as appropriate for cases where I know the ID on the observation is wrong. I hate to use it to distinguish meaningless from meaningless (in the limited context of providing a meaningful species name to a dandelion observation), but if it gets dandelions out of Needs ID, I’ll use it.
3 Likes
I logged this as a bug, but it was closed.
It seems this is a bug which needs to be resolved and staff agree as such…but it’s not yet logged on Github?
I only see this thread, which is also somewhat confusing.
I’m not sure I fully understand the intended behavior here, but here’s another instance that I would argue has gone wrong.
I had an RG hairy woodpecker observation,
https://www.inaturalist.org/observations/324405336
The recent taxon change auto-updated all the IDs,
It seems to have proceeded in reverse order, so the most recent Dryobates ID became the first Leuconotopicus ID. For some reason the first is ssp. orius and the rest do not designate ssp.
The community taxon is now villosus, no ssp., with seven IDs. But due to the initial orius ID, the observation is still marked Needs ID.
I’m not sure whether the taxon swap of the community/observation taxon rules have gone wrong, but this is at best a usability problem.
2 Likes
I am guessing that the reordering may be due to the subspecies taxon changes being processed first but not sure.
1 Like