Surely the solution to that is to contact the relative backbone and inform them. Rather than a deviations being made and months later the backbone picking up those data resulting in a deviation that does not deviate. With so much online and so many new journals, it is an impossible task for them to know everything immediately.
I think the first action should be to contact the relative backbone about an issue and if they choose not to make a change, then a deviation can be proposed stating that the relative backbone explicitly disagreed with the proposed change. This would be less work for everyone and more conversion of taxonomy globally.
2 Likes
Sounds like possibly more work for everyone.
Instead of keeping an eye on iNatâs curation moves only, it implies monitoring both iNat curatorsâ action and third-party taxonomic framework changes - in the hope of noticing (and report, and get corrected) any questionable change in the latter, fast enough so that no iNat curator has time to mirror that questionable change hereâŠ
Alterations at a 3rd-party framework (e.g. POWO for plants) are a thing, but they bear almost no consequences for the framework, they can be (and are) easily reverted, ten times a month if needed.
Meanwhile, iNat curators replicating such alterations with no safeguard or delay can produce huge, hard-to-revert consequences. (Cf. the Xanthium mess.)
2 Likes
Almost all of those are extremely in-the-weeds technical details required by the ICNafp â fixing gender terminations for species epithets (which is stupid and should be abolished), authorship where the author of the species is different from that of the paper, and citation of original articles for basionyms. Maybe the first should have been caught by a peer reviewer, but not necessarily; the genera in question have typical terminations so arguably both the authors and reviewers should have questioned them, but there are some where they are reversed (masculine words ending in -a and feminine ones ending in -us).
The problem with this is that it would totally exclude some significant taxonomic groups. Are we going to exclude all ~100 species of very distinct peacock spiders because theyâre worked up by amateur scientists in non-peer-reviewed journals?
Ok, I feel like this exchange reveals a lot about why much of this thread consists of people talking past each other. âTaxonomistsâ arenât a monolith, nor is anyone being âallowedâ to do anything. Someone can just publish something â even with peer review itâs not that hard â and as long as it conforms to the codes of nomenclature, itâs out there. That doesnât mean you have to accept it! The big problem with POWO, as I see it, is that it uncritically accepts any publication without evaluating the evidence behind it; and likewise, iNat uncritically accepts POWO.
This leads into a second implied issue here, which is the conflation of nomenclature and systematics. The codes of nomenclature solely govern what names have validity and priority etc. based on how and when they were published. They are neutral when it comes to subjective synonyms, i.e. whether two names should scientifically be considered two species or one.
The publication of names, synonymies, etc. are taxonomic hypotheses. You can read the paper, evaluate the evidence presented, and accept them or not. Many taxonomists, when they encounter a bad paper, simply ignore it and continue to use the previous systematic framework. There is a lot to be done and too few people and resources to do it. Even with things that are known problems â for example, when someone in my area published hundreds of spurious new species, sometimes describing one species from the flowers and one from the leaves of the same plant specimen â it took over 10 years for them to be resolved properly. I myself am working on a group that hasnât been revised in over 100 years.
6 Likes
At some point, we reach a situation akin to arguing about the difference between a pit bull and a Staffordshire terrier. Assuming the 10th of a mm business wasnât an exaggeration, surely every non-microscopic species has larger within-species variations than that? Are we going to erect a new species, say Sequoia microphylla, because the needles are 0.1mm shorther than those of Sequoia sempervirens? Iâm not sure that I understand what you meant by the 0.1mm reference. Unless it was alluding to:
If this is the case, then it shouldnât take measuring to less than 0.1mm.
Sounds like a good systematic review project.
1 Like
Well, my issue isnât that we should only allow institution scientists to describe species, my issue is when one person or research group splits off a new species without anyone else reviewing it, and then it finding its way onto iNat. Or someone rushing changes onto iNat only to find out the paper was rejected or revised or the authors of the paper didnât feel like it should actually result in taxonomic changes, and by then a bunch of issues have been created in the iNat data. This is a different issue from my issue with taxonomic splitting, itâs also true that changes sometimes happen on iNat before they should, and without much discussion beyond the curator who wants to make the change for whatever reason.
2 Likes
Two species may look completely different as a whole. The measurements between these two species may mostly overlap though, which makes it often very difficult to write a simple key to distinguish them. Often, the characters that are most distinct and do not overlap are small. Is it better to write a wishy-washy key with a bunch of larger characters that overlap or is it better to use smaller characters that mostly or completely differentiate the species? You could in fact have every single measured character overlap between species but the combination of them all makes them look completely different.
I think couplet 5 of this key was one of the greatest plant key WTF moments I have encountered. That is until I had to use it for a rare plant survey. Guess what? The pollen grains perfectly differentiated the species in question on my survey and, once I had keyed several, I could clearly identify them to species from several feet away. Overall, they looked pretty different but almost any other measure would overlap and not work well in a key. Having recently done a taxonomic treatment with a key, I have huge sympathy for anyone that ever has to write or use an identification key. I also have huge sympathy for any expert that tells you that they just look different and that is how to tell them apart.
8 Likes
To me it would be best, though, if such key couplets also included the syndrome of the other overlapping characteristics that give the taxa their unique field aspects. Not all things mentioned in a key couplet have to be mutually exclusive, if mentioned in explicit combinations. It makes for longer key couplets, but also for broader utility as an identification tool, which to me is always the first purpose of a key.
So my WTF for that key couplet still stands. 
2 Likes
My WTF still stands too but it is a double-WTF as it actually works. My guess is that, in this case, no syndrome stood out that aligned with the solid character. I suspect that it was just a solid character that pulled a few taxa away from the greater mess and made the rest of the key somewhat easier regarding the remaining taxa.
2 Likes
If the writers of the key only wanted it to be usable by people with access to a compound microscope, then I would agree.
Couplet #3 here is an example of what I have in mind (and with species still considered cryptic by many!):
http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=131832
3 Likes
Before I clicked on this link I said to my husband, âis this going to be the one about the pollen being 5-angled?â Then I clicked and laughed.
3 Likes
You can actually see it with a hand lens once you understand what you are actually supposed to be looking for, ideally a 20x rather than a 10x though.
The parts of the couplets mostly donât overlap in that one though. Many similar couplets overlap most values at least somewhat and the plants I key usually end up in the overlap area of such keys. Imagine writing a Stylocline key to 30 similar taxa. You got lucky that you only had 7.
Just for fun, below is the last couplet I got to in my Malacothamnus key when I attempted to focus almost entirely on morphology. I still had over half of the taxa to fit in the key and each couplet was getting worse. I ultimately decided to do a key starting with geographic divisions to make it much more usable. It wonât work for taxa introduced outside their current ranges but infinitely more useful than if I had tried to focus strictly on morphology. Note that in this couplet the only character that doesnât overlap is a trichome measurement to the nearest tenth of a mm and it is the average length rather any trichome as the extremes will cross the couplet boundaries. And, of course, you ultimately just learn that it looks different from the rest as that is way easier than trying to key it.
16 - Calyx bracts 6-13 mm, calyx lobes 6-18 mm, flower buds pointed, calyx glandular trichomes average ℠0.2 mm, inflorescence and leaves often sticky, Los Angeles and Ventura Counties ⊠M. marrubioides
16â - Calyx bracts †11 mm, calyx lobes †12 mm, flower buds pointed or not, calyx glandular trichomes average < 0.2 mm, leaves and inflorescence rarely sticky ⊠17
I think I would have surrendered and told those rare species: you are free to become extinctâŠ
1 Like
Fair enough - it was just the quickest example at hand where some creative use of âandâ, âorâ, âand/orâ etc. can be used to flesh out characteristics that otherwise wouldnât be as amenable to dichotomous couplets. When reviewing FNA drafts, I generally flag couplets like the ones youâve brought forward (especially when they are at high levels in the key) and ask the authors if they can add any other âfield characteristicsâ to help key users - âhow do you recognize these species when you see themâ etc.
That said, I think weâve probably sidetracked and dominated this discussion enough, so Iâm going to step back again. For me this was my roundabout way of agreeing that
and that keys can sometimes suck a little less when authors think longer about translating their recognition cues for a wider audience of key users.
2 Likes
Man, amen so hard to that last sequence.
Fungal keys are absolutely infuriating sometimes because theyâll jump immediately to microscopy - which I understand but thereâs definitely macromorphology that could have been used to differentiate species further before jumping right to needing a microscope.
3 Likes
Surely it is not the task of âPOWOâ to evaluate all published taxonomy (that is the last thing anyone would want). It is for scientists to evaluate and write a peer reviewed rebuttal if needed. POWO is merely the messenger not the peer reviewer in chief.
1 Like
If you write a nomenclatural paper, nomenclature being entirely ruled by the ICNafp, and then in 200 occasions not following the ICNafp and create a bunch of invalid names, then that seems quite fundamental to that publication. The ICNafp is all about technical details. A peer reviewer may miss a few, but 200?
Hi, you seem to have misinterpreted my comment. Surely you can only propose a deviation when you know there is a deviation, so there is no extra work âmonitoringâ anything as you have already looked to see that it is different and therefore you are going to propose a deviation.
The sequence of events, as of now:
(1) someone at POWO implements a questionable taxonomic change: Xanthium strumarium is declared a synonym of X. orientale (huh?! nonsense!!!)
(2) someone at iNat notices that change at POWO and (because it is âThe Reference Frameworkâ whereas curators are not expected to be experts or have a critical point of view on taxa they curate) decides to replicate it on iNat: after a few hours and with no further safeguards, a âtaxon mergeâ operation makes 1000s of obs of X. orientale renamed to strumarium
(3) after a few weeks, someone at POWO realizes how misguided the former change was, and reverts it in a snap: X. orientale is back to being a different species from strumarium
(4) meanwhile on iNat, everything stays a big mess that will take months if not years to revert and fix - photos, CV etc. got merged and confused too, new obs got wrong IDs, etc.
How it could be:
(2) someone at iNat notices that change at POWO and decides to replicate it on iNat: a âtaxon mergeâ operation is therefore drafted but, because a taxon merge involving thousands of existing obs is a lossy operation which could cause a big mess if performed hastily, a minimum delay of N weeks + a minimum requirement of N curators/topIDers is enforced - allowing enough time and eyeballs for experts to step in and declare that POWO was drunk and their change makes no effâing sense.
â
edit: to clarify a bit⊠not every curation operation requires cautionary measures and critical insights. I envision at least three cases: (A) Many ops are lossless (e.g. simple rename or typo fix, moving a well-defined species back and forth between two generaâŠ) and as such, donât really call for expertise or third-party input; (B) Other ops are inevitably lossy even if not dramatic - e.g. deleting a subsp. or var., merging two taxa⊠and therefore should require varying amounts of extra delay/care/input; (C) Some other ops, even if reversible and harmless on iNat, can still cause transient/larger consequences elsewhere - e.g. turning a âSpeciesâ into a âComplexâ, while largely cosmetic for iNat, changes the related observations into âtoo genericâ genus-level data on GBIF.
7 Likes
Yes, of course they are free to do however they prefer for their site. Anyway, if it happens that one or more relevant papers are missing among those that have been cited in the âpublicationsâ section, it is possible that someone will decide to try to make them notice this lacking.
As already written, this is not an accusation towards POWO. I think that, having to deall with the world vascular flora, it is understandable if in some cases some works have not been consulted.
3 Likes