Recently I saw some examples when curators made swaps for taxa which are obviously outside their reported expertise, and which they have never observed or identified. All these swaps were made to follow POWO; however, they affected problematic taxa with unresolved status, and even their treatment in POWO may need revision. One of the swaps was committed despite two taxonomist experts calling for patience and indicating contrary evidence. The worst thing that could happen to the occurrences of these taxa is the instability due to always changing nomenclature, because they may be untrackably merged with one or another taxon. I think that curators should be more self-critical regarding their ability to assess if a swap leads to unnecessary instability.
I suggest that the curatorsâ right to make taxon swaps should be limited to those taxa that (s)he observed and/or identified at least a few times. The exact limits could be discussed below or wherever.
It could be nicely paired with some alerting system that notifies a curator if a taxon within her/his competence gets a flag.
How do you like it?
Related suggestion(s):
https://forum.inaturalist.org/t/implement-automatic-delay-for-proposed-taxon-swaps/38701/15
Enforce technical safeguards e.g. mandatory delay, before tricky/irreversible taxon changes.
I am not in favour. Limiting a curatorâs right to do taxonomic updates would just mean that certain suggested changes will never happen.
A responsible curator should trust the provided evidence that comes with the flag and commit a swap whenever the suggestion is supported by a scientific paper (genus revision, catalogue etc). And I have met curators on iNat that are willing to dive into taxa outside their comfort zone and checking the references etc prior to doing anything. Do not knock out these people who are running the system.
Please do not discourage these curators from doing this - iNat is in dire need of these curators. This is not the ideal world where there is a specialist curator with unlimited enthusiasm and time to take care of all flags.
If you have an issue with POWO - donât blame the curators, rather question the iNat policy of POWO being the declared authority for taxonomic changes.
I know there can be problems â Iâve seen one messy example â but by in large the curators do a good job and I donât see a good way to prevent the few problems without making life a lot harder for our volunteer curators â who donât have to do this if itâs too much a nuisance.
yup. some version of this and/or https://forum.inaturalist.org/t/notify-users-of-drafted-taxon-changes/38765, to allow other folks to provide feedback on changes, seems like a better approach than preventing people from making changes.
I personally disagree with making curators only able to curate taxa they observed/identified, there are a lot of rare taxa that are almost never observed, and in that way curators might not be able to make the change needed to update inat taxonomy.
Given the discordance between iNatâs current curator coverage (ie, where curators come from and their expertise) and the centers of biodiversity on the planet (ie, both geographically and the types of new taxa that are being described/revised), implementing this request would mean that large portions of the Tree of Life would be effectively uncurated. I think that there could be better guidelines, and curators may always make mistakes or not handle a situation perfectly, but the current system is better than it would be with this implemented.
What happen above species level is not that important, those changes can happen at any pace, and while it is not exactly always practical, these are all things solvable by searching for synonyms.
however:
splitting species should be much easier to do than lumping them together, because lumping erases information, splitting adds information (it is debatable how usefull that information is, but nonetheless, oversplit taxa are easier to handle from data point of view, and the existence of cpecies comlexes on iNat is a good soultion to the porblem of some taxa being difficult to ID to species from photo)
I think generally, speaking, lumping should be a longer process and some sort of voting should be implemented;
and I think we should have a way of registering, if lumping happens (esp. if the voting happend, and it was quite an even split), of what taxon would it be in the more split system (after it is lumped, it is not possible to assign the subtaxa, except in notes (which would require manual corrections in the case the sensu lato is unlumped), I think adding a category of âunofficialâ taxon, not necesarrily displayed anywhere visible, or burried in annotations (or DQA) somewhere, just to make a note in the system that the observation is that subtaxon, in case the sensu lato becomes sensu stricto again)
Hmmm, I would not be so optimistic â unless thereâs an automated way to correctly (re)assign all observations during the split, and to efficiently identify all new obs following the split. Otherwise it may just end up pushing back/leaving many obs at the genus level (or in some half-baked catch-all âcomplexâ, not much better).
Merging and splitting both have a potential for making things worse, if only temporarily - thatâs why there are feature requests for reasonable safeguards & checks.
If a curator is consistently defying expert consensus, it may be worth flagging their relevant activity. Otherwise, I do not think a system-wide change like this would be helpful.
What happens with rare taxa that basically no one has observed, like this https://www.inaturalist.org/taxa/428217-Aneuretinae
Can taxa not observed or IDed by active curators due to rarity not be curated?
I think this is heart of the problem. Havenât we had discussions before about undue haste in making taxon swaps?
species complex are a completely valid solution, and one that has bee used in vegetation science for decades, there is no reason to sneer at these. especieally as there are many taxa, where a species level ID from an avergae observation is not possible and should not be made.
Since these exist anyway, a couple more of taxonomic ones will not hurt.
I like complexes :) What I mean by âhalf-bakedâ: notwithstanding their usefulness and purpose, iNat complexes are a mixed bag of âtrueâ complex rank (as found verbatim in papers) and âsect./subsect./aggr./gr.â equivalents (as used in papers, but disguised as âcomplexâ on iNat). Moreover, many (all?) such âiNat complexesâ appear to be maintained by hand, rather than sourced from a reference taxonomic backbone (assuming that would even be possible - afaik POWO lacks infrageneric divisions); it certainly does not help with completeness and consistency.
I agree with the problem but not its solution. Its a complex problem with many moving parts. I donât think there is a simple solution. As a platform, iNat needs to get its arms around the entire problem before it spirals out of control.
The suggested fix to limit curators to observed/identified taxa is bordering on the level of âcruel jokeâ regarding invertebrate or rare species specialists. Plus, we should not discourage those who are taking their hand at dealing with taxa that were once out of their scope.
My understanding of curation is that iNaturalist prides itself on assembling its taxonomy via secondary sources only. Iâm of the opinion this generally hinders taxonomic progress and primary literature ought be the deciding factor, but iNaturalist identifies numerous reasons why relying on secondary taxonomy is more useful for their database;
In the case of accepting an ostensibly recent and contentious taxonomic decision, I would question if it is following iNaturalist guidelines of secondary source requirement anyways. It sounds like the curator in question is giving preference to primary literature, which could obviously create superfluous taxonomic change if it is not accepted by specialists in subsequent years. OR they are indeed following the secondarily sourced taxonomy. If the latter, I would question why it is being accepted to the greater scientific community but still suspected as wrong, a question not usefully debated over iNaturalist sorting quandaries.
Anyways, iNaturalist does a terrific job managing a system full of citizen scientists and pseudonymous identifiers/curators, and it sounds like all is preventable here with the secondary source requirement, which generally forces the acceptance of common practice over recent taxonomy or curator opinion. Moreover, iNatâs system supports laypersonsâ curation of taxa (that may be only recently known to them), because whatever we choose to call things on iNaturalist is sourced from highly accessible, not pay-walled, and not necessarily recent, sources.
There is already such a huge backlog of taxon flags that it can take years to get a response on one - I hate to imagine how much worse that would get with any added limitations.
I do understand your frustration with poorly-applied taxon changes, but Iâm not sure this is the way to address it.
How about some sort of mechanism to âobjectâ to a taxon change, that requires addressment / community overrule before a controversial change can be committed? That would at least prevent the unilateral decision-making that seems to be at the heart of the issue.
I expected the issue of too rare species to come up. For pragmatic reasons, we could define a frequency threshold over which this should apply. However, the core of the idea is to limit incompetent curators from making changes.
Can we please remember Curators are volunteers for whose efforts we ought be grateful.
If there are concerns regarding specific actions of individuals, can we agree those are best addressed via the flagging system / ticket system, and certainly would not belong here.
IIRC there was a feature request, to summon and Request For Comments the top identifiers whenever a taxon change is created. Or something like that, canât find it nowâŚ