I am not a curator. People tag me on flags sometimes, since I have a lot of IDs. Most of the time I have no knowlege about the taxonomy and wouldn’t even know what/where to begin reading to educate myself. So I imagine the same is true for a lot of users, but still it couldn’t hurt to automatically notify the top 5 identifiers or something. It would probably be educational for me.
This should be feasible yes… but solves only part of the issue with “reverted taxon changes”.
In the same example I mentioned earlier, the reversal at some third-party taxonomic DB (the one used as reference by iNat) took “only” several weeks. The situation evolved like this:
- Month 0: a TaxoDB merges SpeciesA and SpeciesB for whatever deeply flawed reason
- Month 1: iNaturalist takes notice, drafts then quickly commits a ‘taxon swap’, merging NNN obs of SpeciesA and NNN obs of SpeciesB into a new unique all-encompassing ‘SpeciesA’ (…voiding hours of past identifier work)
- Months 2-3-4: observers and identifiers (grumpily) assign new observations of these diverse organisms to the only choice at hand, the new all-encompassing ‘SpeciesA’
- Month 5: the TaxoDB comes to its senses, and reinstates the good old SpeciesA and SpeciesB
Now, what to do at iNat?
– for the old observations that were merged during Month 1, your idea should work: go back in time, reinstating the former SpeciesA and SpeciesB, then reverting the taxon swap here by withdrawing en masse the recent IDs.
– what about the (dozens/hundreds of) new observations added during Months 2, 3, 4? the issue is, they were all assigned to the dreaded all-encompassing SpeciesA but there’s nothing to revert here! and you can’t split things magically either.
For these “recent” observations, the only reasonable option is moving everything back to genus – rendering moot the ID work done during Months 2, 3, 4 – then calling on (grumpy) identifiers to revisit these obs they’ve already reviewed, and resume/restart identifying “SpeciesA vs. SpeciesB” as if nothing ever happened. Not very nice.
3 Likes
Ah I see. Thank you for the example.
That’s also where two Feature Requests (not accepted yet, alas!) could have helped:
- enforce some mandatory cooldown delay, at least several weeks, giving plenty of time for any taxon specialist and top identifier to come back from a holiday, log in, and evaluate the situation, eventually contacting the upstream taxoDB to inquire about (and maybe prevent) the change
- publicize more broadly the drafted taxon change before it happens (in proportion to its numerical importance, potential for disruption, etc) thus giving anyone interested a chance to chime in and offer remarks, caveats, silly opinions or bright ideas, suggest some extra delay or deviation from the reference, refine things, notice and fix tiny errors etc.
7 Likes
These two items have been on my wish list for years on iNat. I’ve now even forgotten what reasons have been given (e.g., by staff) for not implementing them.
2 Likes
This should be a very easy thing to implement and I don’t know why they haven’t done it. There should just be two name fields for everyone’s ID. There should be the ID taxon that someone provides, which should never change unless the IDer changes it. There should also be the currently recognized synonym of that name which is automatically changed depending on whatever taxonomic concept inat follows. In the situation of a taxonomic lump, the previous IDs would be retained but presumably displayed as the lumped names, at least for community ID. If/when the folly of the lump was realized, it would easily revert back to the previous ID if the IDer didn’t change their ID to the lumped version.
2 Likes
Can you please give an example of a questionable taxonomic change with a big impact for ecologists? What impact exactly do you mean?
1 Like
I am not a taxonomist nor a specialist.
That said, recently I uploaded an Observation of a butterfly species known to me, via the app. There was no alert, nothing different, however when I opened the same Observation on the computer to annotate, there was an alert on the desktop version that there was a taxon change pending for the species. It was the first taxon change thingie I had seen.
- I wonder: would I have seen it had I not opened the desktop version?
- I wonder: would I have seen it had I not been looking for that species?
I opened the taxon change, being curious by nature, and tagged a BAMONA specialist for Mexico within the taxon change and also thanked him for reviewing it with an eye toward Mexico, because, as noted, I am not a taxonomist nor a specialist so certainly I am not the one to assess it.
When I saw this thread, I went looking for that taxon change among all the taxon changes .
As I could not recall which butterfly species it was, I simply entered Butterflies, thinking surely that would bring up all taxon changes within Butterflies.
It did not, it only brought up the taxon changes that dealt with Superfamily Papilionoidea.
- Does this mean that the BAMONA specialist for Mexico would need to be subscribed to every species in Mexico to possibly be made aware of all taxon flags for all species?
I might question how aware the necessary parties are being made so that discussion, if needed, can take place.
As I said at the outset, I would not be one.
edit to add:
This feels like it could lead to a debate of a particular change rather than a general exploration of the topic at hand.
4 Likes
yeah there are plenty of individual changes I have gripes about but my understanding is we aren’t supposed to call out particular actions or users on the iNat site, on the forum. So that’s why i’m being vague. I will tell you that the change this post was inspired by involved a three way split and some were concerned the science only supported one of the splits. But i haven’t looked in detail in that case because i don’t think iNat should adopt any of them right now.
1 Like
Yeah, best not to delve too deep into specific [pun intended] cases. The general idea behind this thread (in a nutshell: gather more and broader input before stringent action) appears sensible enough in and of itself, without the need for dropping names.
(Incidentally, the hasty taxon change I had in mind earlier has rendered iNat data much less usable for some conservation purposes: a near-extinct native taxon was merged with its abundant invasive relative; that unfortunate move and its messy result still echo to this day.)
3 Likes
there also used to be a LOT of emphasis on how the main point of iNat was to be accessible to anyone with access to technology. A lot of things myself and some others wanted to implement like stricter controls on location precision were rejected because “inaturalist is was created to connect people with nature, the data is just a byproduct, it’s great to get data, but it’s secondary to making sure inaturalist is usable to a broad audience”. Or something like that. At that time i felt it was too far the other direction and led to actual problems with the data like locations not being usable. Somehow at some point it flipped and taxonomy changes that in my opinion are wholly incompatible with broad usability and access became a major part of inaturalist. A huge amount of focus and energy now goes into changing taxonomy to match the current state of the field which seems to be very rapid changes, often not fully vetted, and extensive splitting. I know people have different opinions on this, and some people like it, but broadly here i am more wondering why the main focus of the site changed so much and whether it was a conscious change by the people running it or culture drift by influential curators that the staff just haven’t pushed back on for whatever reason (maybe just being overwhelmed). I know I’ve gotten really riled up about this and have irritated some of the more influencial users of the site, but truly a lot of it could be solved with some communication. I’ve got over 60,000 observations and have been here over 13 years so i’m very invested in the community. It would be nice to be told if the shift of the site is from broad usablity and outreach with kind of self curated data use to trying to make a perfect taxonomy database that won’t match most people’s field guides or datasheets. Like, i get it, it’s not my website and they have the ‘right’ to do that if it’s what the people running it decide to do or if a huge chunk of the community wants to. But it would just be really helpful to get some communication so those of us who it doesn’t fit for are able to gradually look at other options or change how they use the sit erather than just having ti explode in their faces one day. Because that’s really hard.
So yeah there’s no point pointing to just one thing becase there’s been hundreds (thousands?) of these incidents and really it would just be nice to get some staff guidance. Because i can’t tell if a few curators went roge and it’s one of those social norm things where i am just supposed to be made to feel unwelcome and shut up or leave, which might be worth appealing to staff to deal with, or whether this is an actual directive where the people who run the site want this to be a totally current taxonomic clearing house, in which case i’d be looking at alternatives and scaling back my use a bunch to just a lifelisting tool for my personal enjoyment.
4 Likes
I agree with others who have suggested that the process of managing the iNat taxonomy is lacking. Actually, I believe the iNat taxonomy is the platform’s weakest link. It’s probably not productive (for me) to recommend specific changes so I’ll offer an analogy: If the iNat codebase were managed the same way the taxonomy is managed, we would have chaos. Put another way, the integrity of the taxonomy is nearly as important as that of the code.
2 Likes
As I see it, this is a consequence of the weakness of the processes upstream by other entities. If there is no single authoritative source for us to align, very little we can do here – apart from being extremely cautious with anything that cannot be reverted easily. Primum non nuocere – first, do not harm.
3 Likes
I agree that there should be more communication about taxon changes, and I think the suggestions above regarding a waiting period before commitment and broad publication are excellent. At present I don’t even know how I would find proposed changes, there could at the least be a page listing all pending transfers that could be filtered to find taxa of interest.
But after the discussion somebody still needs to decide whether to make a change. Who gets to be the decider? It seems to me that the problem is not that a Shadowy Cabal of Evil Taxonomists is in charge, rather that for the vast majority of the branches on the tree of life, nobody is in charge.
The official position of iNaturalist seems to be that where possible taxonomic decisions are outsourced to other providers such as the Clements checklist for birds and POWO for plants (to mention a couple that I am familiar with). This works pretty well for birds: there are a relatively small number of species and a mature, fairly complete list, and there is a well defined process to make changes where formal proposals are submitted and voted on by a standing committee. It does not seem to work as well for plants which have 1-2 orders of magnitude more species, and decisions on change to the checklist appear to be made on an ad-hoc basis based on whatever publications have come to the attention of the compiler most recently.
Managing taxonomy is a core requirement for iNaturalist, and given the size of the job I don’t know how that can be managed best. The one general suggestion I would throw out is that it seems like it would be technically feasible to make more use of synonyms. Unlike with a physical book a database does not need to specify a single name for a taxon. The unique descriptor could be a number, and the display name could be whatever is most useful for the end user. The default display name for a plant could be whatever was listed in POWO, but I since I mostly make observations in California I might specify that I want to see the names listed in the Jepson eFlora for California. But this is probably drifting off topic…
2 Likes
To be very clear, i know i get heated about this, but i want to clarify that i don’t think any taxonomists are evil or intentionally harming conservation. I think taxonomic research is valuable and should occur, but i just disagree with the extent and rate to which species-level changes are made, the relative rate of divisions at species vs subspecies, whether it is a good conservation strategy to split out more species to push that more of them get rare status and protection, the validity of creating paraphyletic groups after splits to ensure data integrity and communication (I support it, many do not). While the views i advocate may seem too far one direction to some, i’m definitely open to middle ground and things that work for everyone, i just have felt like any effort to do that is dismissed and deemed ‘unscientific’ (which i disagree with) and portrayed as only one person’s crusade against taxonomy, which isn’t true (i like taxonomy, i just disagree on how it’s implented, and also i get messages and posts from many people as well in real life who agree with me or have more ‘extreme’ views on taxonomy change than I do such as “they are all a waste of time and none should happen ever” which isn’t actually my view.). I do think it’s possible this view will just be dismissed, i’ll get mad and leave the site, and i do believe others will leave or not join, inat will attain more of a strictly academic standing, and it won’t be a good result for conservation or the community. If it’s a priority to the iNat community and admins that taxonomy changes are frequent, splits are fine-scaled and at species level, and holding bins are strictly monophylletic, etc there are ways of improving communication and transparency that might still make the site more usable to those of us who would struggle with that. Clearly getting mad and arguing in every flag isn’t productive, and clearly it isn’t just me getting mad about this, but I don’t see that getting better unless a broader policy is fleshed out and actually followed. This will need admin involvement. I’ll accept that splitting is law of the land if admin says so, and choose whether to stay or not, but right now it really does feel like a lot of it is a few curators acting without full admin support and outside the guidelines. I’ve reached out to POWO and am talking to them too, and have gained some valuable info that way as well. Maybe we just need more involvement of ‘lumpers’ within the broader system and not just iNat. But going back to school and doing a thesis on taxonomy is not accessible to me, so that may be a dead end on my part. Also ultimately iNat isn’t a democracy, and whatever the devs decide to do, we have to conform to or leave (or fight until we are kicked out).
I think synonyms are a good idea, and obviously aren’t impossible in the database given how many languages the site uses. But it does get hard to use synonyms with splits and lumps.
1 Like
Try this and insert the taxon of your choice :)
2 Likes
I have not thought much about how use of synonyms would work in detail, it is just something that seems like it could be technically feasible. More flexibility for end users might make decisions about individual taxon changes less consequential and maybe less contentious.
1 Like
Yeah, something like this but for actual splits or lumps that have been created rather than for all flags.
There was a proposal in the old Google Group, by Patrick Alexander if I remember right, to allow multiple active parallel taxonomies to exist on iNaturalist. This could solve some problems, but would also increase the burden on taxonomy curators, and result in a proliferation of messy many-to-many relationships, so it seems unlikely to be implemented.
2 Likes
This would be the place to start:
https://www.inaturalist.org/taxon_changes
You can filter by taxon, by the kind of change, and by whether it is pending or already committed, among other things.
1 Like