Increasingly, my understanding also.
I don’t believe these things are fundamentally at odds though.
It seems if anything, just a gap in the “market” which will eventually be filled.
In UK, for Diptera at least,
Facebook groups = expertise & outreach but limited data collection.
iRecord = expertise & data collection but limited outreach.
iNaturalist = outreach & data collection but limited expertise.
I see no reason a platform won’t eventually tick all the boxes.
iNaturalist certainly seems close…
I like the idea of some sort of workaround similar to your suggestion though.
If observations are just annotated one by one even with “ID by so and so”, would obs with that annotation be possible to pull out from the GBIF side in theory?
As it stands, I´m not really sure how I will port my observations into the UK datasets without a lot of problems - imagining I will just have to investigate the API better at the end of the year…
1 Like
That is sad reality. I already have encountered a person here who describes himself an expert but is neither close to any expert nor behaves like one.
5 Likes
Interesting to hear. Are you based in USA?
I wonder how many scientists in UK use data from iNaturalist…and what their experience is…
I think a lot of the issues I´m encountering and the cynicism I´m hearing from UK experts are to do with conflict with existing recording schemes, processes and dataflow within UK structures. Probably also just some entrenched traditions of natural history recording in UK as well, which might not be the case elsewhere.
I can see heavily weighing the opinions of experts could potentially impact the community input. I think there must be lighter interventions to explore though which could increase accuracy without significant community cost…
E.g. something like @aspidoscelis suggests… or tagging recognised experts as @mtank mentioned…or other smaller actions…like a simple pop-up explaining better to new users what RG means when they first agree to an ID…
1 Like
With regard to GBIF–I have very little experience interacting with GBIF at all, so can’t say much beyond what I can see in the online interface. It looks like the only identification information that gets transferred is the community ID and the name / date for the first annotation using the same name as the community ID. It definitely is possible to get the complete set of specimens identified as taxon X by user Y via the iNaturalist API, but I only looked at it enough to verify that I could figure out how to get there if need be, not enough to actually do anything with it. The main sticking point for my potential use cases is that, while being able to work with particular users’ IDs rather than community ID would get me part of the way there, it probably wouldn’t be worth the hassle unless I could also overcome the limitation of IDs to names on iNaturalist’s “accepted” list. The best approach I’ve been able to come up with is to use the “observation field” functionality to basically tack an independent taxonomic database onto iNaturalist, and use the API as a means of populating that observation field (at least, for cases where it could be populated unambiguously without checking the images associated with individual observations).
1 Like
The thing is though is that professional experts have other ways of getting observations of the taxa that they are experts in into the broader research community - if they collected a specimen and housed it in a museum or university that databases it, then it more than likely will end up in GBIF.
If experts really want their observations of “their” taxa to make it to Research Grade, then I think part of it will have to be them being willing to teach amateurs how you can distinguish those species from a photograph. If they can’t actually be IDed down to species from a photograph and the expert who is posting it knows what species it is because they did the dissection/genetics/etc. necessary to get it to species, then it probably shouldn’t make it to Research Grade at the species level just from that photograph.
I think the best way “experts” can interact on iNat is to (1) post observations of things that are not their speciality but that they’d like to learn about and get help with IDs, like everyone else; (2) help with IDs in their taxonomic specialities; and (3) help others learn how to ID species within their taxonomic specialties.
8 Likes
Yes, absolutely. I was inspired by an expert identifier, @nathantaylor, to begin specializing in his field of expertise (Anisophyllum). With his generous help and a lot of independent study, some in the field and much on iNat itself, I’m now able to check/confirm IDs in Ansiophyllum with a high level of competence. Before, he was basically identifying (and correcting) the section all on his own. I’ve been able to bump a lot of observations that he IDd years ago to research grade.
10 Likes
The comment you quote was in regard to an “expert” making an ID on someone else’s observation but it remaining in the “Needs ID” pool for lack of secondary expert agreement. (…not in regard to an expert uploading their own specimen and being unable to have it verified …)
My comments here all relate to enticing experts to participate in the ways you mention in 2 and 3…and how we can encourage more of this.
2 Likes
I don’t think this thread has been linked here yet: https://forum.inaturalist.org/t/rename-research-grade-discussion-and-polls/590
(also, unrelated, my experience of Ontario fish IDs has been positive)
7 Likes
Maybe not so unrelated!
Most of my comments stem specifically from my experience with UK Diptera.
I think its really important we acknowledge that the issues being encountered and debated here can indeed vary wildly according to taxa and location.
Nice link. Another I’d not come across yet.
1 Like
I agree. It seems iNaturalist has never really settled on what it wants to be, despite defining itself on the “What Is It” page. Or perhaps, its diverse users haven’t.
If iNat is primarily a website designed to get people into nature and learning about nature, fine. For that purpose, we don’t necessarily need species-level IDs for most difficult taxa. For example, insect field guides generally focus on getting your ID down to family or genus and for the average person interested in nature, that’s usually good enough.
But if iNat is going to encourage species-level IDs for many obscure taxa, experts are needed and most experts also have a research interest. Maintaining the database’s taxonomy is a herculean effort in itself and that requires the input of individuals who are highly knowledgeable. Amateurs can certainly participate and contribute to the site and use it for learning purposes, but it also has to be inviting to experts who provide a lot of the quality control. If iNat is regularly presented as NOT primarily a database of biodiversity; where accurate IDs are not the main goal; and where Research Grade is not all that important … why should an expert want to get involved?
Getting people into nature is definitely worthwhile. But you also really need the people who are already into nature to help make the site functional and scientifically useful, which would attract others with expertise. I’m not sure how iNat can balance that based on its stated primary goal.
12 Likes
Yes, certain taxa definitely have more identification attention than others.
I’ve said elsewhere that when I started identifying, I started with birds since that’s what I knew. I quickly realized that tons of people were already identifying birds and there wasn’t much I could contribute there. After that I started learning hover flies because they didn’t seem too difficult but no one was identifying them. BugGuide, a recently published field guide, and other resources (e.g. guides in iNat journal posts) have since made them a lot easier to learn and now more people are helping.
I think one of the best things an expert can do to lessen the amount of work for themselves on iNat long-term is to make it easier and more interesting for others to help them. There is some onus on observers to try to identify their own observations, but that’s nearly impossible for many invertebrates. Making those resources does require more intense effort though (at least initally), and “please make guides for the taxa you’ve studied for years” is a much bigger ask than “please help us ID some of your taxa”. Even just linking to resources in your profile and responding kindly to requests for clarification can go a long way.
9 Likes
Flies are hard, probably one of the hardest groups (wasps are worse I’m sure). There are so many species, and they’re so ubiquitous that there’s over half a million observations that are “Needs ID”. Many can’t even be identified confidently to genus without a microscope. In addition, even in academia they’re understudied because of funding priorities so their taxonomy is often messy, making them even harder to learn.
5 Likes
the best way i can describe it is, i think most power users are most interested in data collection or curation (for various reasons depending on the person) and most ‘casual’ users are more interested in connecting with nature and getting IDs. There is of course some intergradation between the groups, but it seems to be a recurring theme. And not surprisingly, this forum is disproportionately populated by ‘power users’ though we certainly aren’t all on here. Most of the more casual users only end up on the forum if they have a problem.
8 Likes
UK has own long history of websites and databases, but iNat has a lot of info too, plus we do have some experts from diptera.info helping on iNat, which is really cool, so we’re not hopeless in identifications for that order.
4 Likes
Absolutely.
This is why I think we need more expertise…
Whilst I totally get that in European birds at least…there’s probably limited need.
1 Like
I use the description on their profile, and their helpful engagement on ID threads - to evaluate expertise.
‘I am a naturalist’ no, thank you unless you show that you are hiding expertise.
But - I am writing a monograph on the taxonomy of whatever then I will trust your species ID. I have seen obs tagged for species novum, once that description is published.
4 Likes
As one of those professional experts, I don’t think that’s accurate. For specimen records, yes, there are other platforms (mostly SEINet, in my case) which fill that data need. For non-specimen records, I am on iNaturalist because there isn’t a science-focused platform that addresses this data management need, and iNaturalist is so close to being the right tool to fill the gap.
I think there’s a perception that actual science isn’t compatible with citizen science, that we can have either citizens or science and citizens are the priority so science is kind of the red-headed stepchild. I think the concept of citizen science requires that we create room for synergy rather than approaching it as a zero-sum game. I see how iNaturalist handles identifications and taxonomy as the biggest obstacle to that synergy, and think I can see the path through. It looks like a zero sum game because the system is focused on the idea of the “right taxonomy” and the “right ID”. Those concepts provide very helpful mental landmarks for many purposes, so keep them around. If you’re starting to learn plants, you want a name to put on it, you don’t want to know who disagrees with whom for what reason. If you’re a taxonomist, the different possible answers and the evidence for each is the whole point. Putting a name on it is just bookkeeping. Being able to express the different possible answers is better than getting the right answer.
8 Likes
I don’t see the site as centred around the ‘right ID’ at all, if anything the exact opposite. It is focused around the ID that the community of people who have chosen to review and chime in with an identification coalesce around, which may or may not be ‘correct’.
If the site were centred around getting the ‘right ID’ it would have implemented an ‘expertise system’ a long time ago.
2 Likes
iNaturalist’s conceptual framework for identification and taxonomy is built entirely on there being a single right answer. I agree that iNaturalist does not prioritize arriving at that right answer above all else–but the system is still built on the “one right answer” concept. For instance, names not on the “accepted” list are prohibited from use. The opposite of “one right answer” is “one wrong answer”, the alternative I’m arguing for is “multiple possible answers”.
For research, either of the “one answer” opposites is downright toxic. If you have two Opuntia taxonomists, you have two Opuntia taxonomies. Collapsing that to a single taxonomy with a single ID per observation puts those two researchers into conflict. One must prevail and the other be vanquished, they have no room to maintain alternative viewpoints.
2 Likes
This site is not the ideal forum for those taxonomists to press their disagreements. The unfortunate tendency of vicious arguments to break out on North American snakes is a perfect example of this.
Having those debates in the context of observations is both confusing and far too often offputting for the vast majority of users on the site.
7 Likes