it would be nice if iNat at least allowed users to opt out of taxonomic change. At the current time you can choose not to auto-accept changes, but people still come in and try to force the observation into the new taxon. So i reject community ID but then it never can be research grade or else it gets forced to maverick and taken off the map. I think forcing everyone into constant taxonomic revisionism is unfair.
1 Like
I think laypeople almost as widely understand the taxonomic concept of a ‘family’, but typically use it in a narrower sense closer to the Linnaean rankings of genus or possibly tribe. For example I could say ‘the allium family’, ‘the onion family’, or ‘the garlic family’ to basically anyone who gardens and expect them to understand what I meant, but perhaps not ‘genus allium’. They would also not assume that I was including amaryllis in ‘the allium family’, even though genus amaryllis is in the same ‘family’ that contains genus allium.
5 Likes
No, you don’t completely understand what I’m trying to describe. I’m only interested in observing and IDing to genus for Taraxacum. This is fine, but there are 80,000 outstanding obs of Taraxacum. Part of it is the officinale snafu, but even if there were a bunch of species… most wouldn’t be able to ID to them. If you identify a core group of experts here like in Europe, you’re convincing them to focus on tons of obs, many by new users, instead of relying on people like me to make it go “green” at genus.
If it automatically went “green” at genus… at least people like me made a first pass at it. And a pro could try to ID it further (to “gold”).
To make it even more sophisticated, you could prompt for additional data when a Taraxacum obs is made by someone who has the node un-unbundled… this might improve the data collected and ID’d at “gold” level.
But this requires storing additional data to manage the prompting… which would make it complicated for developers. The only reason it might deserve passing consideration is that it would achieve these two goals…
1 Like
Do you mean similar to what I suggested here for species complexes, but chosen manually at different levels (complex, genus, family…) for different taxa?
Sorry, I hadn’t read it, but yes. Seems similar. The thing I like about the “gold” grading is it tells the user they need to be more savvy. The problem is that my depth of knowledge isn’t sufficient enough about any taxa to not frighten the expert on that taxa. Even so, consider the chickadee example from above***…
If someone makes an obs of a chickadee feather (and it’s observably a chickadee feather), and they have that species exploded (meaning they want to try to reach “gold” level id). The expert on chickadees could have created “gold” level obs fields to define what is required to determine “black-capped” vs. “carolina”. Maybe the prompt is for a link to DNA sequencing of the feather. Most observers at this point, will just ID to Poecile. Some will be excited to go for the gold. They’ll pay the $35 bucks for sequencing and send that feather off. This will (should? might?) excite the observer and the taxonomist.
But it’s also why we should think about how “gold” encourages users to behave. I want the sequencing done more on Trillium and Chickadees than on Taraxacum… because of the region I’m in. Defaulting of exploded nodes should be site specific (but adds more complexity).
*** I have no idea what it takes to Id chickadees to species. Maybe it can be done by observation alone? Maybe it takes sequencing? I’m only using them to demonstrate a concept. I am the one who needs the prompting. I apologize in advance for my ignorance on chickadees. If you would like to give me a really stern look, I’ll allow it. The example of the chickadees was actually really helpful, so I’m using it even if I butcher it. The thoughts on Linnaean rankings being partially motivated by biblical teachings was also helpful. Software requirements and design is an iterative process that depends on subject matter experts (SMEs). Always has been. It’s interesting to me to explore the possibility of “open source” requirements and design. But maybe just this once!
I think what you are describing sounds like some version of a feature request that @aspidoscelis has suggested several times including here:
https://forum.inaturalist.org/t/improving-inaturalists-nomenclature-taxonomy/36143
I thought there was also a regular feature request somewhere but I’m not finding it
Gentle warning, the migratory bird act in the USA makes it illegal to possess native bird feathers without a permit.
I realize this is just an example you’re using, but I wouldn’t want anyone to run afoul of the law by doing this.
https://www.fws.gov/lab/featheratlas/feathers-and-the-law.php
4 Likes
Sorry. Had no idea. Better to change it to droppings?
That one seems to be more about when two people have different opinions about what the taxonomy should be rather than a user choosing as a default option to Id and observe some ranks as bundled and some as un-bundled.
Where they become a little similar is… what to do if the bundling can’t be elegant. It’s simple if it’s always just bundle all species and below up to genus. Or bundle all sub-species and below up to species. If it is more complex than this… it still might be worth considering for cases like Taraxacum. 80,000 needs-id and most obs by new users. We need an army of people plowing through them at first glance unless the observer picked the non-default option to allow prompting for their gold level Taraxacum obs. The prompting would suggest what data they should gather for say, red-seeded or something where careful data collection is needed to positively Id to the lower level.
I don’t know what the data looks like in POWO or who has update access to it… but I assume it’s taxonomists. It would be cool if the observation fields required for “gold” status were defined there and associated with a rank. And then the presence of the “observations fields” would tell iNat when to collapse the bundle for someone who didn’t self-identify as a “gold” user.
Trying to help my peeps.
Ah, that would require very intricate programming and curatorial discussion.
For example Black-capped and Carolina Chickadees have hybridization and identification challenges, but all the other species in the genus in North America are easily separated (although not from a single feather). Willow Tit and Marsh Tit in Eurasia are extremely visually similar, but I don’t think there’s any question about them being different species. If they were invertebrates or plants they might be considered cryptic species, but being birds their differences in ecology, morphology, vocalizations, reproductive separation, or whatever else is important is apparently clear enough. It looks like they both even have a lot of subspecies described.
Taxa below species (more granular?) include sub-species and variety. People have been using these more granular taxa since Darwin’s time and their use or application has nothing to do with new technology. Some applications of new technology have shown there’s no basis for designating sub-species of a particular species (and so scientists stop using subspecies designations in that particular species). Other applications have shown that a particular species does have two or more distinct populations that could (if one wanted) to designate as subspecies.
There’s no such thing as an increasingly granular species. A species is a species regardless of the technology used to identify/delimit/discover them. The criteria we use to decide whether a unique population should be recognized as a species is unrelated to the technology being used.
We have moved on beyond the original rankings devised by Linnaeas (for example, we have subfamilies and superorders). So we’re not using “ancient” rankings.
3 Likes
Excellent point. This is why the observation fields for prompting should be stored with the taxonomic meta data.
i don’t know what taxa you focus on, but with vascular plants this is untrue. The definition of ‘species’ used to involve only discrete units but now tiny variations that genetically intergrade and used to be subspecies, varieties, or not described at all, are now considered separate species.
Taxonomists have been very consistent that they alone (with input from taxa SMEs) define the traits necessary at each rank. And so they should make the first attempt at deciding what prompting and for which ranks. It could happen slowly. For taraxacum only at first for e.g. The initial release would be update to the taxonomic meta data only.
One thing I’d love to know is… do any of the ranks or sub-ranks… section, sub-species, sub-genus, etc… Do any of those support the concept of a foreign key. A rank that has a parent but also a second parent identified by the foreign key (foreign key being a db concept… but maybe taxonomy too?) Or are they all truly hierarchical.
Of course they are. but it’s up to the rest of us whether we choose to accept their proposed terminology or not. In many many cases people already don’t. The more they are willing to work with other users the less this will happen.
Depends on which taxa, and probably where?
As for the diversity of local hawkweeds (Hieracium, like dandelions DeLuxe) it went from 90 distinct taxa in the 1970s (90 subspecies, grouped in ~31 species), down to only 22 now (all at the ‘field-recognizable species’ rank 
with no subspecies anymore).
But who knows what new taxonomic craze will happen in the next 50 years… 
2 Likes
Oh, for sure. That’s why i think the field visibility level is useful for vascular plants albeit itself hard to draw the line on. There isn’t any distinct genetic or phenotype related way to declare a group a ‘species’ since it’s all a human construct, albeit one based on things that are legitimately different and worthy of knowing and describing in some regard.
If I’m interpreting you correctly, despite all the movement they still remained hierarchical which wouldn’t be a big deal.