Leading subspecies IDs should change the obs taxon like leading IDs of other ranks

[jwidness]

Currently subspecies IDs only move the observation ID forward if there are at least 2 IDs for that subspecies (unless the only ID on the observation is a subspecies level ID). Can we change the behavior to something like “move the observation ID as if the subspecies ID were at the species level”?

Current behavior:
Left at class level because there is a class level ID followed by a single subspecies ID.

Moved to species level because there is a class level ID followed by a species level ID.

At subspecies because the subspecies ID is the only ID.

Desired behavior:

I think some would actually prefer that a single subspecies ID move the observation ID to the subspecies level, the question is how does that interact with Needs ID vs RG? If an observation is RG at species level, I think many/most don’t want to move it to Needs ID at the subspecies level with a single subspecies ID (which is why the current behavior exists in the first place).

A couple reasons to want to advance the observation ID: to be able to find the observation via species filters, and to have the auto-obscure work appropriately at the species level.

[charlie]

i think this ties into a broader issue of how subspecies are treated. Some myself included don’t think people should add a subspecies ID by range only (don’t call it Sierra lodgepole pine just because it’s in the Sierras, if you can’t see the characters that define the subspecies) because it’s self-reinforcing and at best tells you nothing and at worst masks variability in subspecies range. Maybe letting a user choose for themselves, but it may bee too complex or specific an ask. I do agree the system could be tweaked to be less confusing.

[tiwane]

@jwidness

Can we change the behavior to something like “move the observation ID as if the subspecies ID were at the species level”?

Forgot to reply to the thread about this in the Google Group. I talked with Ken-ichi about it and yes, we think this would be ideal but we have not been able to figure out a way to do that and prevent situations where a sequence like sp1 sp1, ssp1 results in the obs taxon being ssp1 and the obs being Research Grade. So I think this is an issue of how to code it rather than this being something we don’t desire.

[kueda]

FYI, I changed the subject of this thread to be more specific, since it wasn’t clear to me what was being proposed here while perusing the thread titles. And yes, like Tony said, I banged my head against this problem for a while when we initially separated the obs taxon and the community taxon (which I still don’t love) and didn’t come up with a good solution.

[henrydelange]

@tiwane

Forgot to reply to the thread about this in the Google Group.

For reference, the link to the Google Groups post: https://groups.google.com/d/msg/inaturalist/KnDFoW6BNok/dOXSr4GsDgAJ

(I though you guys didn’t see it )

[tiwane]

@kueda should we close this topic as this is already on our radar but have not determined a method for it? Then anyone who voted for it will get their votes back.

[kueda]

Works for me.

[dan_johnson]

I just ran into this again today. It’s been something like 9 months and still no fix. To me this is a terrible bug and I’ve just resorted to not identifying something to subspecies to avoid it.

[grace_sibley]

Yeah, I just ran into the same problem. For now, I am withdrawing a couple of subspecies IDs and identifying only to species level to avoid this.

[dan_johnson]

I think a solution to the research grade problem is perhaps to add a category to research grade. For example there could be “research grade (species)” and “research grade (subspecies)”. So, in the “sp1 sp1 ssp1” case, it would be “research grade (species)”. But another ssp1 ID need to be made to achieve “research grade (subspecies).”

[sessilefielder]

@tiwane

prevent situations where a sequence like sp1 sp1, ssp1 results in the obs taxon being ssp1 and the obs being Research Grade

Isn’t this already an issue? A sequence of ssp1, sp1, with the second user answering “I don’t know but I am sure…” to the Potential Disagreement prompt results in a Research Grade observation with an observation taxon of ssp1.

[matthias22]

I was directed to continue this conversation here (I started a separate thread, not knowing there was one already for this issue: https://forum.inaturalist.org/t/identifications-that-include-a-subspecies-ignored-in-observation-id/8926)

I think this could be solved computationally without being overly complicated:

1. If the designation “research grade” is restricted to the species level as the lowest possible taxonomic level the subspecies could simply be ignored when calculating whether an observation is research grade or not. In the definition of terms it would have to be made clear that the designation “research grade” is only applicable to the species level (or higher), and that any subspecies identifications are an independent, additional piece of information.

2. The community ID at the subspecies level can then be computed independent from research level calculations. It can still be done based on the same rules but it would be independent, and it would only take place when the observation ID has reached the species level (regardless of research grade or not). A subspecies ID can then be added to the observation ID unless it is contradicted by one third or more of subspecies IDs that apply to this species (subspecies IDs pertaining to other species can be ignored if there are any in the previous determination history).

In a nutshell: subspecies calculations are kept separate and research grade is a designation that applies to the species level as the lowest possible taxonomic rank. I think this approach is logically sound. I am not a computer programmer but I don’t think this would be complicated to implement or draw excessive resources to calculate.

Thanks!

[birdwhisperer]

To add my little two cents worth. As explained above, subspecies ids should change the obs taxon like taxons of higher ranks. To explain the issue going about, as I explained here.

If someone submits a obs with the id Class Aves or birds and I submit a Red-shafted Flicker (Colaptes auratus cafer) id, the obs will remain at class level. If I were instead to submit a species id or in this case, just simply Northern Flicker, the community id will be lowered to species level. And it’s my personal belief finding a flicker obs in the “Identify” page is easier than searching for simply “bird”. Those obs may fall in-between the cracks because subspecific ids do not lead.

I can see how this having status quo can help prevent subspecific ids for sole fact of subspecific ids. If we change the current implement, if an observer submits a species and I submit the subspecies, it would become RG at subspecies level without a supporting id. I see two solutions to this, one let subspecies ids lead only if the original id is genus level or higher, or after a subspecies id, the obs goes back to “Needs ID” in order to get supporting subspecific id.

And if subspecies cannot lead, why can a subspecific obs reach RG without a supporting id? Ever since March 2019, if an observer submits a subspecific id and an identifier puts a species id, like every other possible disagreement id you have to decide whether you disagree or not. Here’s the results of those choices.

I don’t know but I’m sure it’s (species taxon) – RG as the subspecies
No, but it’s a member of (species taxon) – RG as the species

Contrary to a species-genus id.

I don’t know but I’m sure it’s (genus taxon) – Needs ID at species level
No, but it’s a member of (genus taxon) – Needs ID at genus level

So I personally believe that we need to decide sooner than later whether or not subspecies ids get the same attributes and characteristics as other ranks or none at all.

[kiwifergus]

@birdwhisperer

if an observer submits a species and I submit the subspecies, it would become RG at subspecies level without a supporting id.

No, this could never happen, even if you change it. 2/3+ concensus is a requirement for RG.

@birdwhisperer

I don’t know but I’m sure it’s (species taxon) – RG as the subspecies
No, but it’s a member of (species taxon) – RG as the species

No. The first choice here does NOT give RG as subspecies. It gives RG at species with a Display ID of subspecies:

The second option in both cases is an Explicit Disagreement, and effectively nullifies or forces the other ID to be treated at the agreement level (species for the first, genus for the second) as far as calculating CID is concerned. They are the iNat equivalent of being able to vote in an election and also claim that the person who voted before you is an idiot for voting for someone else, and that that vote should be disqualified. [Apologies for expressing my own personal dissatisfaction over how explicit disagreements are handled!]

Another requirement for RG is that the CID must be finer than genus (ie sub-genus or better) OR be at genus and have “ID is as good as it can be” set in the DQA. The reason the second example goes to Needs ID is because the CID becomes genus, so doesn’t meet this OTHER requirement! You can test this out by setting “ID is as good as it can be” for the second example, and it will become RG at genus.

To be a “double standard” it would have to be a difference of how two otherwise equal things are treated. Subspecies as a taxonomic level is not the same as species or genus… every taxonomic level has it’s own set of “rules” that relate to it. A double standard would be having the same taxonomic level treated differently under animals vs under plants, for example. If you consider “taxonomic levels” to be equal things, then it’s a double standard to say two species level IDs = RG when two family level IDs can never be RG. The RG label is all about the differences between the taxonomic levels!

[jwidness]

@kiwifergus

It gives RG at species with a Display ID of subspecies:

Well… that kind of depends how you look at it. Here’s an older example that’s had time to export to GBIF:

The Observation Taxon [what you’re calling Display ID] is at subspecies (with the RG banner next to it), while the Community Taxon is at species only. If I understand you correctly, you’re saying this observation is research grade for Sciurus niger only, not the subspecies. However, it’s been exported to GBIF at the subspecies level:

And if you search for research grade observations at subspecies or lower, there it is:

And looking at its score, though species level is highlighted in gray, the subspecies level is given the same score of 1.

So there might be some sense in saying it’s only research grade at species, but I would argue there are lots of ways in which it’s being treated as if it’s research grade at subspecies with only a single subspecies ID.

[kiwifergus]

The “two 1s” show that there are 1 ID at each level, the next column is the cumulative, and it clearly shows 2 at the species level but only 1 at subspecies.

However, that is the only fault I can see in what you are showing here. To me it is clearly exporting and treating the Display ID as if it were the CID!

[edit] Actually, in the search box it doesn’t state it is searching on CID or Display ID, and it would make sense to search on Display ID…

@tiwane @kueda @loarie @carrieseltzer Is Display ID meant to be used in these cases that Jane brings up? I would have thought CID should be used…

[bouteloua]

Yeah, it’s been this way since Jan 2018. See also https://forum.inaturalist.org/t/search-by-observation-taxon-or-community-taxon/3620

[kiwifergus]

@birdwhisperer It would seem (going by @bouteloua’s and @jwidness’s comments and links) that this is going to be addressed in the explore revamp, it might be best to wait see what comes of that.