| Flagger | Content Author | Content | Reason | Flag Created | Resolved by | Resolution |
|---|---|---|---|---|---|---|
| dctropics | dave_holland | Tree ivy (Fatsia japonica x Hedera helix) (Genushybrid x Fatshedera lizei) |
ill-advised taxon merge |
Mar. 7, 2023 17:23:20 +0000 | rynxs |
not accepted by POWO |
Comments
As stated in my PM response to you:
The name is listed in POWO, but not accepted. Names listed as "artificial hybrids" are not accepted by POWO - you can verify this by looking for × Fatshedera on the list of accepted groups in Araliaceae here: https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:30001539-2
An important aspect of iNat taxonomy of which few users seem to be aware is that taxa lacking taxon framework relationships and not accepted by our source for taxonomy for the related group can and will be swapped without warning. Due to the sheer number of artificial hybrids in cultivation, POWO does not accept any of these names that are not naturally occurring. Since iNat follows POWO for vascular plant taxonomy, we also reject these names.
In essence, whether or not the name is validly published or observations of it currently exist simply does not matter. What matters is that it is not accepted by POWO, and for almost nine years no one bothered to add a taxon framework relationship marking the taxon as something that should not be swapped. So, in keeping with POWO, the taxon has been merged with the lowest common denominator taxon between its parents.
iNat does not prioritize observations of hybrids, especially not hybrids that are not naturally occurring. If you wish to track observations of artificial hybrids, I would recommend using observation fields, as they are not subject to taxonomic curatorial actions. Let me know if you want a list of observations previously identified as × Fatshedera.
You can still search for IDs of this taxon: https://www.inaturalist.org/identifications?taxon_id=1442272¤t=any
@rynxs Ryan, again i do recommend not to do such, or similar changes, which are counter-productive, serving no purpose.
There is no need to expect that iNat would/should reject hybrids of any kind, no matter if natural or artificial, or that iNat's taxonomic concepts necessarily needed to be exactly the same as presented by POWO.
Changing all reports/observations of × Fatshedera lizei to Araliaceae is certaily against the intention of iNat's users, that's what counts at last and needed to be respected by you. Thus in my point of view the merge is not only counter-productive, but respectless as well.
Erwin, we do not accept artificial hybrids. Users are more than welcome to post them, but our identification system is sourced from POWO, which does not accept artificial hybrids. Therefore, we do not either. Respect has nothing to do with taxonomic sourcing and maintaining our taxonomy in a manner than follows POWO, as dictated by staff. If POWO were somehow able to accept and list every artificial hybrid ever conceived, then there would be a basis for accepting them. However, POWO does not.
Artificial hybrids do not fit into our taxonomy. They should not be added, because when added, they need to be maintained. The curators that add these taxa neglect them, which is how I have been able to single them out and find them in the first place.
These changes most definitely serve a useful purpose, so I have no idea what you mean. Following POWO is the primary objective of vascular plant curation, which is why curators and staff need to link every taxon to POWO in some manner.
Clements and MDD don't include hybrids, but iNaturalist recognises bird and mammal hybrids. Mules are an artificial mammal hybrid, recognised by iNaturalist. They cannot reproduce, and therefore cannot establish feral populations. Tree ivies are also sterile.
And following frameworks is not the primary objective, improving iNaturalist is. Following frameworks is just a means of doing that.
Excluding tree ivies is actually more problematic than excluding mules. If a tree ivy was growing in the wild, it would have to be IDed as Arialiaceae and marked as having a Community Taxon as good as it can be. The result is that it would be casual, as iNaturalist considers any observation that cannot be IDed beyond family to be casual. A similar situation for a mule would make the observation Research Grade.
The removal of tree ivies has made tree ivy observations that weren't marked as captive/cultivated Needs ID, which is inappropriate. The ID is perfectly known, it doesn't need to be IDed further. I have been marking such observations as having a Community Taxon as good as it can be, which made them casual. Now I suppose they were cultivated anyway, so they would be casual anyway.
All right, i got the meaning. It's obviously very important and useful to you to do likewise changes, so there is no need for proceeded discussion to me.
@cyanfox some exceptions are made for hybrid taxa well-known enough to cause more trouble with the community if omitted than included, such as mules. Wild hybrids are included because they are naturally occurring.
If × Fatshedera is capable of occurring naturally, you should send any available evidence to POWO or write a paper about it to help POWO accept it as a natural cross.
@rynxs
Given that the example of an intergeneric hybrid used by the Curator Guide—an example that's lasted unchanged since it was added to that section of the guide (and to the iNat taxonomy itself) by one of iNat's co-directors in 2014—is itself an artificial hybrid (with several hundred observations, which isn't an outlier numbers-wise—I just stumbled across another artificial hybrid with nearly 20,000 observations, and I'm sure that's not the only similarly observose example), the iNat leadership evidently isn't 100% across-the-board opposed to any and all inclusion in iNat of validly published artificial hybrids (which in certain cases may be important to track as potential invasives, if they have the potential to expand out of cultivation; I admit I don't know if that's a possibility for ×Fatshedera).
True, it wouldn't be very productive to add every published artificial hybrid under the sun to the iNat taxonomy (the majority of which would never be encountered outside of cultivated settings), and iirc staff have very strongly recommended against doing anything like that (instead recommending only adding hybrids if necessary and generally sticking to wild organisms)—but occasional judicious use isn't flat-out prohibited. Rather than indiscriminately merging these into their parental taxa whenever you encounter them, it's probably better to assess them on a case-by-case basis in the future, and at least in cases like this one (with a not insignificant number of IDs), rather than simply merging with a parental taxon on sight, the best course of action would likely be to set up a deviation (as the other of iNat's co-directors has done for both artificial hybrids I alluded to in the previous paragraph), or at the very least discuss with frequent IDers of the taxon first to get a consensus before merging. (After all, if deviations from iNat's taxonomic references were never allowed, then there wouldn't be infrastructure on iNat specifically to support them—and if artificial hybrids were fully prohibited, as you claim is the case, then iNat admins wouldn't be using deviations to include artificial hybrids.)
Adding to the confusion, it's not clear that x Fatshedera is truly an artificial hybrid. "In 1912, when Lizé Frères of Messer’s Nursery in Nantes, France, showed their plant of Aralia lizei in an exhibition (see fig. 196 in ref. 4), they stated that the specimen had grown from seed set on a plant of Aralia moseri (Fatsia japónica ‘Moseri’), which had been pollinated accidentally by a neighboring specimen of Irish Ivy (Hederá helix L. var. hibemica)." (https://journals.ashs.org/jashs/view/journals/jashs/112/6/article-p1053.xml) So this is possibly a spontaneous (i.e., natural) hybrid, albeit one arising in cultivation.
BTW I entirely agree that artificial interspecific hybrids (of which there are probably thousands) have little value in iNaturalist (many nothospecific names are misapplied anyway, e.g., Begonia rex-cultorum). But these hybrids will always be identified to genus, which provides some degree of precision. I believe that intergeneric hybrids are qualitatively different (as well as far less common), and allowing their identification only to the level of family is very imprecise.
@maxkirsch taxa not aligned with POWO can and will be swapped without warning to force alignment with POWO. It is up to people who care to find and request deviations for taxa lacking connections to the framework.
I did assess this specifically. I looked it up on POWO, looked around a few resources, and found that the hybrid has never been observed outside of cultivation. If every curator waited for "consensus" on every change they put through, we would get nothing done.
To be clear, the addition of this genushybrid to our taxonomy went against curator policy. It should not have been added, but since it was, it needed to be dealt with. As I stated prior, staff do capitulate to users who demand certain artificial hybrid taxa, but I have no such reservations. When you leave a taxon without a relationship, expect to have other curators fix the mess you've created in whatever way they see fit. It's not as if this was an established, heavily used taxon, either - it was added January 23rd of this year.
To be fair, identifications of this taxon go back to 2017, but the earlier ones are obscured by a later taxon merge. Also, are we talking here about recognizing the nothogenus x Fatshedera, or the species within that nothogenus, x F. lizei? The two seem to be conflated (i.e., the name of the hybrid genus is not x Fatshedera lizei, it is x Fatshedera) but are entirely different taxonomic ranks. (I can no longer find the nothogenus, even as an inactive taxon.)
EDIT: found x Fatshedera (presumably added on or period to 3017) but see the rather odd taxon change in January 2023, which swapped the perfectly legitimate nothogeneric name for a hybrid formula: https://www.inaturalist.org/taxon_changes?taxon_id=415500
@dctropics the nothogenus and properly formatted hybrid name were created in 2014, but this specific taxon was created in 2023, alongside a hybrid formula taxon. It looks like the properly formatted hybrid names, alongside the hybrid formula taxon, were swapped into the improperly formatted names in an attempt to preserve capitalization of the genus name, which is an issue with name formatting on iNat (I had assumed the opposite would have been true, so I didn't check the properly formatted name's creation date, my bad).
We are talking about both the nothogenus and the hybrid itself.
All around a complete mess, and I couldn't ask for a better demonstration for why unsourced hybrids are unwelcome here.
I'll just add one last comment here reiterating that your merging of ×Fatshedera into Araliaceae, not considering [that when the majority of the feedback you're getting is that this decision may have been problematic, there's maybe a slim chance that may be the case], isn't fully consistent with how staff members themselves have dealt with similar cases in the past (and the Curator Guide doesn't imply that curators should be more aggressive in stamping out taxonomic differences and interfacing with other iNat users than a staff member might be in the same situation). I'll second the observation above that iNat's main purpose isn't to "force alignment with [taxonomic sources]" no matter the case—if the iNaturalist community (of which you, I, and all other commenters here are a part) sees fit to deviate from a source in specific cases (as does seem to be the case here from comments here and elsewhere), there's infrastructure set up to allow that. The issues people have with your merging of ×Fatshedera (which had lasted nearly unchanged since 2014 on iNat) into Araliaceae primarily revolve around the fact that this discards a significant amount of prior ID precision. "staff do capitulate to users who demand certain artificial hybrid taxa, but I have no such reservations" seems slightly ...unfair [in its assessment of both staff and other iNat users]? over-intense? unbecoming? idk. It might be good to get staff input here. (perhaps @tiwane ? no need to respond if busy)
Given that your view on ×Fatshedera doesn't appear to be in the majority here, I wouldn't be entirely surprised if someone reactivates or re-adds it in the future [note—this isn't me implying I'm going to run off and do that now—I'm solely noting that it's a possibility that someone might]. If that were to happen, and if someone were to add a deviation rather than leaving its relationship unknown, would you likewise simply re-merge it into Araliaceae contra nearly all feedback you've gotten so far in this case? (Or is it solely the lack of a previously set-up deviation that's the issue?)
I'll mention again that a staff member (and a founder of iNat at that) both 1. added ×Chitalpa to the database (and there's no indication this wasn't of his own volition as you imply) and 2. happened to pick, out of countless possibilities, ×Chitalpa, again of his own volition, to illustrate a nothogeneric name in the Curator Guide, the same guide which you quote to support your stance which, if carried to its logical endpoint, would result in ×Chitalpa being merged (with a significantly less precise parent taxon).
(As an aside, that section on intergeneric hybrids could potentially use some expansion and clarification—in its current wording, it's unclear that that explanation of how nothogenera are formed only applies to taxa whose nomenclature is governed by the ICN rather than any intergeneric hybrid across all life [it doesn't apply to animals, for instance], and perhaps incorrectly implies that that explanation is a guideline for curators to use to come up with compound names themselves when necessary [7–8 years ago that issue came up briefly with e.g. this, for example]—but that doesn't change the fact that the section in its current wording, as written by an iNat admin, uses an artificial hybrid as an example)
Also, just to clarify re: your last comment, ×Fatshedera isn't an unsourced hybrid, it's an artificial one—the two aren't synonyms. As dctropics mentions, ×Fatshedera lizei (and ×Fatshedera) is a validly published and legitimate name, not listed as a synonym of something else in POWO but rather as an artificial hybrid (but per dctropics, whether it should even be labeled artificial by POWO may not be fully clear-cut). How exactly this situation "couldn't [be] a better demonstration for why [hybrids like this] are unwelcome here" isn't fully clear to me, given that 1. pushback only started due to your unannounced unilateral merging of ×Fatshedera (which had been used seemingly without issue on iNat since it was added by another curator all the way back in 2014) into a much less precise parental taxon and 2. the mixup around the date these were added, suggesting to you they weren't long-used, resulted from an unnecessary swap of ×Fatshedera to a hybrid formula with the genus and species of one of the parents erroneously reversed in January (plus a followup swap by the same curator who added the first swap resulting in ×Fatshedera lizei having the wrong rank on iNat), which could happen with any nothotaxon (or even any other taxon involved in superfluous or erroneous taxon changes), not just artificial ones, but you aren't campaigning to, say, ban all nothotaxa just because similar confusing swaps involving them are a possibility. tl;dr: I'm not sure much of the confusion here is related to the artificial nature of the hybrid in question, but rather specific to the history of this iNat taxon
People who do not have an issue with a change tend to not comment on them. Four people do not represent the wider iNat community.
I did not say the purpose of iNat is to force alignment with POWO, I said that alignment can and will be forced for taxa that are unsourced and not accepted by POWO. Please do not misrepresent my comments to portray a point I never made.
Staff have made clear to me, on multiple occasions, that they "have no love for maintaining hybrids." Take that as you will, but I see that as not wanting them, as is reflected in the curator guide's language.
An example is an example. Staff using a well-known hybrid cross to illustrate how hybrid taxonomy works isn't evidence that we need to maintain any and all artificial hybrids that are added to iNat. As stated prior, staff make some exceptions. As it seems to me, these are done for the community, because the community wants them. Not because staff want them.
Correct, if someone were to have bothered to create a deviation, I would not have been able to just swap it. In fact, I would have never found it. Obviously, I don't want anyone to re-add it, but if someone did recreate it with a deviation, I would not have license to spontaneously force alignment with POWO. I think it's more telling that the taxon, in some form, existed for nine years without any curator bothering to link it to our taxon framework. If it had been linked, that relationship was destroyed in the convoluted swaps that happened before. If you care about any other taxa without relationships, I would recommend helping me shrink the list of relationship unknowns, at which point you may understand why artificial hybrids are untenable: https://www.inaturalist.org/taxon_frameworks/10/relationship_unknown?page=1
Yes, × Fatshedera lizei and × Fatshedera were unsourced. They were not given a substantiating source in the framework, therefore they were unsourced. Just adding a taxon does not mean it is sourced, you have to actually link it to the framework, with a source, which creates a relationship. This is the crux of the issue here. Additionally, I looked at the link, and the specimen says "accidental," not "spontaneous," which, to me, implies human error rather than a semi-natural occurrence, not that this matters anyhow.
The issue with this taxon is that, since it was not linked to POWO, a bunch of confusing, unsourced changes could occur that should never have happened. As a curator, you are not supposed to inactivate "matches" without significant community discussion and a consensus, but × Fatshedera can never be a "match," so a bunch of confusing, unsourced changes occurred that were not caught and corrected by other curators who could have fixed these errors, until me. Ideally, one would be able to catch these changes by looking through relationship unknowns (any "wild" taxon this is done to would have previously been linked, and could be referenced with POWO), but it's currently bogged down with a bunch of other taxa that cannot be linked to POWO, such as this one, which I have been working on remedying. These artificial hybrids are more trouble than they're worth. Just create an observation field if you want to track observations of houseplants, greenhouse specimens, or garden plants.
@rynxs You didn't say that the purpose of iNaturalist was to force alignment with POWO. But you did say that the primary objective of vascular plant curation was to follow POWO.
@cyanfox because the primary objective of vascular plant curation is to follow POWO. What are vascular plant curators doing with vascular plants, aside from helping to follow the taxonomy accepted by POWO?
@rynxs But have you considered why iNaturalist follows POWO? Surely it isn't just doing it for the sake of it? Isn't it to improve iNaturalist? So isn't that the primary objective of vascular plant curation?
And since the primary objective of vascular plant curation is to improve iNaturalist, not follow POWO, there may be cases in which following POWO is actually bad, as it may conflict with the objective of improving iNaturalist. And I'd say omitting an intergeneric hybrid that could potentially occur in the wild, forcing its observations to be casual, is probably one of those cases.
× Fatshedera lizei observations: https://www.inaturalist.org/observations?place_id=any&verifiable=any&field:Unrecognized%20hybrid%2Fintergrade=×%20Fatshedera%20lizei
× Fatshedera identifications: https://www.inaturalist.org/identifications?taxon_id=1442272,415500¤t=any
The removal of both × Fatshedera taxa has lead to cases like https://www.inaturalist.org/observations/36857668 in which one person suggesting a different ID lead to that becoming the observation taxon. The two × Fatshedera lizei identifications are treated as agreeing with the Fatsia japonica ID.
I agree that iNaturalist should not bother adding all artificial hybrids, however, I think that any hybrid, even if artificial, even if not recognised by the framework, that has been observed on iNaturalist should be recognised by iNaturalist. iNaturalist certainly shouldn't remove hybrids that it has recognised for many years when it would mess things up.
@cyanfox I'm not going to continue arguing this point as it isn't really relevant, but the curator application for becoming a vascular plant curator requires that you state what changes you want to make, according to sourcing from POWO. You are not able to become a curator through vascular plants if you don't agree to follow POWO. Sure, one can broaden the goal of vascular plant curation to whatever degree one likes, but the lowest common denominator of all vascular plant curation is and will be, for the foreseeable future, following POWO. I can make the statement that iNat's goal is to connect people with nature, therefore the goal of vascular plant curation is to help iNat connect people with nature, but that doesn't tell you anything about what vascular plant curators actually do on the site.
You can correct the IDs that have been moved up, there isn't an overwhelming number of them. I know how iNat's ID system works, this result was expected. The framework cannot "recognize" anything, it's what taxa are linked to manually by a curator, or automatically by processes run by staff for taxa already aligned with POWO. "Mess things up" is subjective. There were many users who were disgruntled by the removal of horticultural hybrid waste bin taxa ("Hemerocallis hybrids," for example), but those also should never have been added to the site. If you can find a checklist of every vascular plant hybrid that also follow's POWO's taxonomy and individually verifies each hybrid, then staff may able to consider horticultural hybrids.
If you wish to look through every ID of × Fatshedera, all 53 of them are listed in these two links here: https://jumear.github.io/stirfry/iNatAPIv1_identifications?taxon_id=1442272&taxon_active=false¤t=false&per_page=200
https://jumear.github.io/stirfry/iNatAPIv1_identifications?taxon_id=415500&taxon_active=false¤t=false&per_page=200
POWO is another matter entirely. I’ll save my comments about that for a different discussion, and will only say here that forcing iNaturalist and its users to conform to the taxonomy of POWO (including identifiers who may be experts in their respective groups) misleadingly suggests a taxonomic consensus that does not exist. POWO is one taxonomic authority, not THE taxonomic authority, and represents the taxonomic opinions of a relatively small number of botanists (some of whom have very controversial views on plant classification).
@dctropics POWO is the standard we needed that also works best, just because iNat chose to follow their taxonomy does not mean we are attempting to pass them off as taxonomic consensus.
"the specimen had grown from seed set on a plant of Aralia moseri (Fatsia japónica ‘Moseri’), which had been pollinated accidentally by a neighboring specimen of Irish Ivy (Hederá helix L. var. hibemica)." (https://journals.ashs.org/jashs/view/journals/jashs/112/6/article-p1053.xml) So this is possibly a spontaneous (i.e., natural) hybrid, albeit one arising in cultivation."
I quoted this because I noticed that it was glossed over in reiterating the dogma that this is an artificial hybrid. If the whole basis for excluding Fatshedera is because it is purportedly an artificial hybrid, then this point becomes crucial.
"They were not given a substantiating source in the framework, therefore they were unsourced. Just adding a taxon does not mean it is sourced, you have to actually link it to the framework, with a source, which creates a relationship."
So it looks like to solution is to link the above-named JASHS article as a substantiating source.
@jasonhernandez74 I believe POWO excludes × Fatshedera lizei for being an "artificial hybrid" not because it is artificial in origin, but because there is supposedly no evidence that it occurs in the wild.
Here is an artificial hybrid that POWO accepts: https://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:30117681-2
It is accepted because there is plenty of evidence that it occurs in the wild.
But iNaturalist is not POWO. Evidence that an organism occurs in the wild should not be a prerequisite for inclusion. iNaturalist is a platform to post evidence of organisms occuring in the wild. Also, we have observations of organisms, like × Fatshedera lizei, that are only known to occur in captivity/cultivation. I know iNaturalist is supposed to focus on organisms in the wild, but it's not always obvious whether a plant is cultivated or not. So someone could see a tree ivy and not realise it's cultivated and post it to iNaturalist. And with iNaturalist omitting tree ivies, it might never cross their mind it's a tree ivy, with tree ivy never being a suggested ID. I wouldn't be surprised if it ended up as a Research Grade common ivy observation then.
I agree with @maxkirsch and think a case by case approach is ideal here.
Can users and curators add the occasional artificial hybrid if the situation calls for it? Sure
But you also won't see me manually adding every commercially sold orchid hybrid to the taxonomy either
I'm confused by the merge of this taxon (along with the nothogenus xFatshedera) into "Araliaceae" as doing so loses significant information. While an artificial hybrid, both the nothogenus and the hybrid species are validly published, legitimate names (and are accepted as such in POWO). With the merge, there is no longer any way to either identify observations, or to search for this taxon.