I saw some comment about it somewhere on the forum but hadn’t got around to investigating. This comment made me wonder what I was missing, so I went looking, top, bottom and middle. I still can’t see anything, though? (iNat AU, if that makes any difference?)
Probably makes a difference, yes.
My understanding is that blog posts are not regularly displayed for all users – people who use iNat in a language other than English or one of the network sites instead of .org do not see all blog posts. This is apparently intentional. I vaguely thought there was some other staff post where this was talked about, but this is all I managed to find in a search:
Blog can otherwise be accessed by scrolling to the very bottom of any page on the website and clicking on the link in the footer, but this requires that one remembers to check it regularly for new posts…
I find it concerning users are treated differently due to their language choice or their choice of network.
I don’t think you meant it in this way, but your response comes across a bit passive aggressive. The overall point was that things like that announcement need to be more prominent and somehow pushed out by iNat to users who want that information.
I’m on the Aussie site and it’s been quite obvious on the dashboard page.
Thanks - very helpful.
I can understand that if people don’t use English, pointing them to an English blog post may not be useful (though it does ignore people who prefer another language but still know English), but it seems an odd choice for English-speaking networks sites. I guess I’ll have to remember to check the blog at intervals. Good to know, though, thanks.
I don’t know what the difference is, but I’m only seeing the ID-a-thon results post linked there. Very odd.
For what it’s worth, I didn’t find it passive aggressive, I found it helpful. Sure, it doesn’t answer the overall point about lack of information, but it did point me to a post I was interested in looking at (and I assume there’s no way he could have solved the background communication issue anyway).
From the December 2025 Identifiers Survey Blog Post
”Look for our product roadmap blog post next week, which will include information about product improvements that address some of these issues. In addition to product changes, we’ll continue to improve messaging and engagement with both new and existing users, and create more help documentation in 2026.
Thank you so much to everyone who helps out others with identifications, and to those who took the time to fill out this survey. We plan to conduct more surveys of the wide range of iNaturalist community members in 2026 to get a better understanding of where things stand and where they can be improved.”
Thanks for sharing this — folks can see the full blog post with results here and discuss more specifics about surveys here! Like it says in the post, we’re planning to continue doing more surveys of the iNat community and sharing results as we can, so your feedback is very much appreciated.
11 Likes
Agreed with afdexter and, frankly, if iNat had changed in the manner he seemed to be envisioning, I think that would have jeopardised my own membership. Trying to have this schism between “casual users” and “true” enthusiasts; well who is a true enthusiast? Someone who really likes knowing the names of what’s around them but doesn’t care for or understand the vagaries of scientific taxonomy? People who really like posting pictures of pretty things they saw in their yard somewhere, which is still potentially useful data even if that isn’t the intent? I work as a taxonomist, but not on anything that really gets coverage on iNat; does that make me more or less “true”?
Part of the selling point of iNat is precisely that you can be whoever and get the same benefit: a repository for sightings that’s super-intuitive to use, with an ability to find out what you saw and interface with people who, to varying levels, know what things are. That’s it. That’s all it ever should be, and it’s alarming that one of its own creators seemed to miss that point and want to turn it into something that sees two tiers of users, only one of which has more right to use it; the others get sent onto, presumably, a different app. Why do “casual users” need a different app? Whether your intent is casual or not, the one correct ID answer will be the same for everyone.
And iNat’s interface is already so easy to use; I frankly didn’t even notice much difference between the classic and the new one and just presumed they’d done a big software upgrade, improved a bit of the utility (like zooming in on maps when geolocating your observation) and made it shinier. It didn’t annoy me and I knew nothing about any of the kerfuffles until I came here and read this.
As for AI, well iNat is already AI. The predictive ID function based on what’s “expected” around you is already a form of AI, and it already carries significant problems (which I’m sure tons of users have already covered) to the point where I constantly tell people not to trust any natural history app’s predictive function, including iNat, unconditionally. The idea that they’d take money to improve their existing AI capability doesn’t surprise me in the slightest and I really actually have no problem with it if it means better precision in its ID functions and info delivery. It doesn’t come at the exclusion of the human component, the moderating capability of which is actually AI working as intended, i.e. through the corrections and validations of the users, the precision of the algorithm improves.
I find it bitterly ironic that, through all the stupid, stupid attempts of society trying to find new uses for AI, the ones that consistently work best (not perfectly, just better than everyone else) are those who always found viable uses for it: the science sector. Yet we’ve gotten so Luddite about it that we’re now rejecting even that without understanding that it’s always been there, and that this is the best-case scenario for using it in a positive manner. Even if it had been communicated better, I doubt the people who were already prepared to be outraged by something like that would’ve changed their minds. And the solution certainly isn’t yet another app that divides the user base and gives quicker (read: less precise) info for “casual” users.
As far as I’m concerned, iNat should never pretend it’s better than, or not intended for, some people. It is not purely a scientific resource, nor is it purely a casual engagement outlet. Scientists who use it know the shortfalls of using the dataset; it’s the same as using any citizen-collected dataset, and will curate accordingly. All in all, that blog post read like one person’s personal differences with the wider leadership, which is always unfortunate, but these things happen. Most of this would have passed totally unknown to the majority of users (I only found out right now when I came to check something else).
One last blog post I’ll share directly here for now since I think it’ll be helpful in this discussion — we just published an overview of our January–June product goals. Please do continue keeping an eye on your iNat dashboard for future blog post updates, we’ll be sharing more as the year goes on!
Thanks for this, it’s great to see.
We’re also working to substantially reduce traffic from search/scraper bots to our website and Network Sites. While some automated traffic is inevitable on today’s internet, finding ways to cut down bots will help us reduce costs and let us focus on other site improvements.
I don’t know enough about the behind the scenes of the internet to know what other mechanisms are possible for this (I know a couple other citizen science websites stall me on some kind of bot-checking page for a couple seconds before letting me in), but requiring logging in to access more of the platform seems like a reasonable response to this that would also have some privacy benefits for users. I admire the openness of iNat and the fact that increasing amount of the internet are account-locked is kind of sad, but there are enough costs to it that it seems worth it to me.
Because frequently, casual users are sloppy about entering the correct date/location for their observations, and don’t respond to queries for clarification where that data is obviously incorrect.
As an active identifier and data user, this wastes a lot of my time.
I’m not a scientist, but I make use of observation data from iNat (and other sources). I try to at least partially curate the data on iNat prior to downloading it ( to avoid huge discrepancies between the data in my database and the source observations on iNat ). I do the same with specimens in institutional collections - if I find a mistake with a specimen data, I offer the correction to the curators. If doing that curation on iNat becomes too onerous - either because there’s too many low quality observations, or the site lacks the tools I need to manage the deluge - I may be forced to do all my curation “off-site”, within an ecosystem I’ve created for myself. Then iNat will no longer benefit from my curation. You’re making an excellent argument for why I should stop wasting my time doing ID work on iNat.
I constantly read folks lamenting that there aren’t more “experts” willing to perform identification/curation work on iNat. At the same time, there are those who appear to do everything they can to drive those experts away in the name of egalitarianism. You can’t have it both ways. If you want to revel in iNat being a big egalitarian mess, that’s fine. But don’t be surprised if the experts opt to spend their time elsewhere.
I keep seeing people saying that there’s no need to clean things up on iNat because “scientists” should know the limitations of iNat data and should curate their data offsite. I’m starting to be convinced that this is in fact the best approach. After all, it’s no skin off my nose if IDs are incorrect on iNat, as long as they’re correct in my own database. If I have to build my own tools for curating the data at my end, why should I waste my time correcting IDs on iNat? Why shouldn’t I just go the extra mile and do the ID correction at my end as well? It would be a lot simpler. No need to explain ID corrections to users who probably won’t even read those explanations (and couldn’t care less), or worse, engage in pointless quibbling with listers who only want to pad their life/year lists. No need to wrangle other experts to help upvote corrected IDs, etc. etc. etc.
I could just periodically download observations from iNat in whatever state they’re in, filter/cull them to get rid of the garbage, and then correct IDs as needed on what’s left before the data goes into my database. Of course, if I ever get questioned about discrepancies between my data and what’s on iNat, I will have to explain that observation data on iNat cannot be taken at face value because there’s nobody minding the store. ie. that all data from iNat (and by extension, much of the data from GBIF) is riddled with errors, and requires extensive filtering/curation/correction before it can be used. By extension, other data users will have to repeat the same curation work for themselves. Or they can save themselves the headaches and get the curated data from me (which happens frequently).
Up until now, I’ve defended iNat against criticism from taxon experts who argue that it’s a huge mess. I say that this isn’t the case for my taxon/region because I make sure that it isn’t, and if their taxon/region is a mess, perhaps they should roll up their sleeves and do something about it.
Arguments can be made that pursuit of excellence can create its own virtuous circles.
that feels uncomfortably like using the data, from the observers and the active identifiers - but giving nothing back. Could you mute the black sheep or is their data still useful to you ?
Does muting a user prevent their observations from appearing in an identification search? I just tried it and it doesn’t appear to work.
I already have a “blacklist” of users whose data I’ve decided to never use (for various reasons). I can and do filter some of these problem users from my identification search using the URL filtering, but the URL rapidly becomes unwieldy (my blacklist is getting long). But that’s a different set of “problem users”. They are semi-regular contributors who have shown themselves to be “unreliable”, have opted out of community ID, etc.
The “casual user” is in a different class. It’s impossible to filter the casual observers because they are legion, and may only ever contribute a single observation (in my taxon/region of interest). There may be absolutely nothing wrong with their observations, but it’s difficult to know because the observations are often one-offs. Unlike the problem users I’ve described above, there’s no pattern of behavior. But that wasn’t really my point.
In the context of the question of why there should be two versions of iNat - one for casual users and one for more “serious” users - I see it as giving two different sets of users what they need. In other words, I agree with what the iNat co–founder said in their blog. If somebody just needs a quick ID and doesn’t want to be bothered with data integrity, does it make sense to lump their observations in with those made by meticulous observers? Does it make sense to force identifiers (and data users) to wade through a lot of dross to find good/useful observations?
Nobody tries to sell one-size-fits-all shoes, and AFAIK, this isn’t regarded as elitist or discriminatory. I don’t see why there can’t be a simplified version of iNat for the casual user and a more sophisticated version for users who want and need that functionality.
But it’s like there’s a group of people who insist that iNat can only work for EVERYBODY if it works for NOBODY.
Hi All, the topic of this thread is
but the conversation here has diverged pretty significantly from that topic. Please keep any future posts here on the original topic. There are plenty of threads about the usability of iNat data and how it can/should be used where these types of discussions can occur. Thanks!
I suppose that data consumers (unhappy of not being paid “enough attention”) could also be considered part of the ‘user base’, even if they are neither observers nor identifiers (…or no longer, after failed/vain attempts).
I suppose that the paths taken and pursued (focus dev work on apps first, jump on the genAI bandwagon, promote data-harvesting events) may also relate to their unhappiness? and ultimately their snubbing iNat data?
To clarify the different app is Seek. Seek does not directly connect to or upload to iNat UNLESS the user expressly chooses to do so. It is more useful as an ID app, which is how I used it for many years. For people who mostly just want to know what they are looking at, it is a much better option than using iNaturalist.
Seek also used to regularly have monthly “challenges“, i.e. collect X many observations of something. I used Seek often enough that it used to recommend I join iNat. I chose not to because it was overwhelming at the time and I didn’t understand that my observations could have scientific value.
The concern raised by some, which I agree with, is that binning Seek in favor of smushing all users into one iNat app that attempts to cater to both is unworkable or generally extremely difficult. It also introduces more casual users onto iNat where they may not be interested in participating, and we end up with more observations that are hard to ID.
If the comments this refers to are off topic, despite the overall message seemingly advocating for two apps which is not the path the iNaturalist directorship is taking. I think you should split the comments and move them to another thread, rather than what I feel like is a signal to people not to comment and engage with what that person wrote, which stiffles conversation and can hurt when people obviously put time and effort into writing something.
This is largely true that much of the reasons a taxon isn’t a mess is because there are users working to manage it. It’s one large reason it’s very important for iNaturalist leadership to try and retain certain users as without them, certain taxa risk falling into disrepair. Currently there are large amounts of taxa that are a mess because no users regularly maintain them. Some select taxa have basically no identifiers.
Really what you describe sounds like a feedback loop. People see messes, say Dark Winged Fungus Gnats. It looks like too much work to deal with, so users knowledgeable about that taxon refrain from joining and helping, the issue doesn’t improve because it requires work, that work doesn’t happen because of a lack of knowledgeable / willing to help users, repeat.
It makes sense to me as some of the taxa that are messes on iNaturalist need multiple 10s of hours to fix and can be exhausting. Honestly the more I think, the more I don’t really think iNaturalist greatly values or respects people’s time.
If a user has uploaded 300 copyrighted images. There is no moderation tool to flag in bulk.
You have to individually flag everything 1 by 1.
If there are 10k misidentifications to fix from a single taxon.
1 by 1 it is.
If you need to mark 1000s of plants as cultivated.
1 by 1 it is.
If you want to add or remove many observations from a traditional Project.
1 by 1 it is.
If you want to mark as good as can be on 1000s of observations.
1 by 1 it is.
If you want to use any of the annotations in large amounts like life stage, sex, etc.
1 by 1 it is.
iNaturalist just lacks options to do many things in bulk that can become very time consuming. When doing repeatitive tasks in high volumes, time can really add up.
Though there is some irony that the place which could probably benefit the most doing things 1 by 1 is going through notifications. Which right now you have to check all at once when using the drop down menu.
Do note there are valid risks to consider when enabling the ability to do bulk changes. It’s not as simple as iNaturalist should enable bulk editing for everything, everyone.
The overall message of this comment. iNaturalist directorship should really look into allowing some form of bulk editing to certain things that do not have any available bulk editing tools. Especially as every year there are more observations, more forms of data avaliable, more users, more things to deal with, etc.
That example could be mentored away - should not be using a taxon specialist for grunt work. We need a better mentoring interface.