I agree with @charlie . Automatic 2 km obscuring of all observations would effectively break my usage of iNat. I would stop using iNat.
Thank you for the link. There are several quite good ideas in that thread. Anyway, I do hope inat observations will not make much harm to threatened species. Poachers for smaller things (fishes, plants) will hardly use it, poachers for larger things probably have better devices, general public will mostly not recognize the species even if it hit their face. The rest - irresponsible photographers/observers are few and will hardly make significant impact. Land developers and similar - they will hardly use inaturalist, too.
i don’t want to rehash the same discussion i’ve had a bunch of times (except i just did, heh) but i use iNat for applied conservation work: mapping of wetlands, creating species lists of small towns in Vermont, fine scale spatial ecology. I need my accurate data points for that and as said earlier, if I couldn’t do that I wouldn’t be here. I understand that for mobile animals the issue matters less because they don’t stay in the same place. But plants are the most observed organism on iNat.
As for a desire for the ‘search’ for a species… that desire shouldn’t supercede applied conservation of said species! Especially when there’s an easy solution to those who want to retain the mystery: Just don’t use inaturalist.
I don’t want to dredge too deeply into internet drama i’ve already been involved in far too much, but the one herpmapper guy has been carrying on about this issue for ages, he doesn’t seem to have a lot of evidence that iNat actually makes the problem worse, and he’s got alterior motives, either consciously or unconsciously (he runs a ‘competing’ site). All I’d ask is you take his opinion with a grain of salt and know not everyone agrees.
The bottom line is that iNat is a new and very important branch of conservation - it’s free and available to most people (though you need the ability to access the internet), it is available not just to policymakers and non profits (which each have their own issues in many cases) but to all people and small organizations, it allows for collaborative sharing of data, and it bypasses many (but not all) of the barriers that keep disadvantaged groups away from conservation work. I understand ‘traditional’ conservation takes the view that biodiversity data is an important secret that only should be revealed to the in group (though note that poachers are already in all of those in groups). But here’s the thing about traditional conservation.
it Does. Not. Work. We are experiencing an extinction crisis that literally is as bad as a large meteor hitting the planet. We have lost 80-90% of many habitat types. This coincides with a bunch of other corresponding social and environmental ills as well. I flat out don’t believe that iNat increases poaching rate, and no one has ever demonstrated that but even if it does, the bast increase of local knowledge and empowerment as well as multiple discoveries of new species, range expansions, and observed new introductions of endangered species, phenology research documenting climate change effects, etc etc etc… i have no doubt at all that iNat in its current form does far more good than harm. I think the concern is higher in herps than other species, but there’s also a secretive and sometimes toxic culture in the herping community… poaching seems to emanate almost entirely from in-groupers, and there seems to be a lack of broader context here (ie conserving habitats and empowering local citizens is more important than hiding data). And yes those who don’t use or like iNat aren’t represented in the conversation. But why would they be?
So yeah, if the herper silo ever succeeds in shutting iNat down (unlikely) someone else will just recreate it. Heck, that person might even be me if i could figure out how. This is in my mind one of the most important aspects of applied conservation right now, and i’ve seen real results in ways other programs and projects just don’t get. I think obscuring species with poaching risk is fine, and also encouraging awareness (just don’t post the highest risk species) and reminding people that they can choose to obscure a whole day’s observations or all their observations if they choose. Have some situational awareness. But letting local communities see what is in their forest or desert reserve or backyard is far superior to a small in group of ‘conservation minded’ people sitting on data silos and fiddling while 95% or more of said ecosystem is paved.
And lest anyone ask, these views are not official views of iNat or the forum, i am not posting this as a moderator, just an iNat user and a person who’s done applied conservation and ecology work my whole life and is so, so tired of constant loss. And I am gonna try really hard not to get dragged into a debate on this again.
14 Likes
If all observations were obscured it would deal with that issue to some degree.
Can you share what your actual use of iNat is that would be broken by that light obscuring, and why you couldn’t just do a formal data request for that information instead?
I’m not sure who the one Herpmapper guy you refer to is? I quoted a lot of people and none of them have any affiliation with Herpmapper so far as I can tell, at least they’re not on the Board of Directors nor any of its committees and I’ve never heard their names in association with Herpmapper before other than that a couple of them enter records. The person who started the conversation isn’t even part of the broader herping community to any significant degree as far as I can tell - at least out of the 250 herpers who I have as Facebook friends, he is only friends with one of them and before this I’d never heard his name.
Most databases which don’t release public data do all the things you suggest through formal data requests. And those data requests are open to anyone, so long as they are clear in their explanation for why they need the exact data. I’ve seen data requests approved for everything from a personal website to a school project.
I feel like my suggestions helped create some defensiveness that probably has a lot to do with other conversations and agendas that go beyond what I’m trying to do here. I don’t want to shut iNaturalist down or harm it in any way. I think iNaturalist is a GOOD thing. I currently prefer it among the databases due to the amount of community interaction, which I think is very very important. I have only just begun entering my data but over time it will result in a very impressive set especially among herps that will include numerous rarely-recorded or even never-recorded species for the database. And I’m still going to keep running my projects and recommending that people create iNaturalist accounts, as I’ve successfully done several times already.
But I just want it to work better. And right now I know a lot of people who don’t want to participate (including many who could fix many of the mistaken IDs on the site) because they are worried about this one issue. And I’m afraid to label species as sensitive because I’m afraid of breaking people’s projects, but I’m also afraid of damage-sensitive species being targeted for the same reason. And I’m seriously considering obscuring all of my observations going forward because of this worry.
I just think that the 25km obscuring is a sledgehammer approach to what is not a sledgehammer problem, and am looking for a compromise that serves everyone’s purposes better and helps build us towards a better site.
p.s. - this is also the 3rd time this week that I’ve seen a negative statement about the herping community on this forum. Really really not helpful.
3 Likes
Well, for sure there are compromise ideas that could work. But please do believe me that your proposal would break the site for a lot of people (and there are a ton of problems with “formal data requests”). So something else would be better, if indeed there really is a problem beyond just people being bothered by open data.
No way you are gonna make everyone happy. No matter what there will be people who don’t want to use inat. That isn’t a problem. It’s not for everyone. And it’s not a good place for super secret data. (Nor is any other place on the Internet). If you have non public data on sensitive species in North America the NatureServe program Biotics is the way to go.
If you want to obscure your observations please do. That’s fine.
FYI all I am muting this thread so I don’t get pulled in. If you really want to know how I feel in detail you can search through all my old posts on the subject, there are lots. 
1 Like
So your proposal to replace the “sledgehammer approach” of 25 km obscuring is to…load a 2 km uncertainty onto > 27.5 million observations? Totally rational and not a sledgehammer approach.
People are “getting defensive” because you’ve proposed a change in the way that the website will behave that will create massive side effects, mostly on taxa that have nothing to do with the problem you perceive. When people object, you’re attempting to convince them that they are a weird niche case that could totally be dealt with by having them fill out forms instead. This comes across as arrogant: both the assumption that you understand how most people use the site and that any people who have problems with your solution are weird outliers, and the casual imposition of transaction costs on people trying to examine biodiversity data.
I think there is room for a discussion here of how to tweak the obscuring function to work in a better and more protective way, but you are not doing a very good job of selling your proposal.
8 Likes
I demonstrably am not doing a good job selling my proposal. Beyond that I think you were pretty harsh. “Irrational” is a suggestion that does not apply to my proposal, nor does the sarcasm help either of us.
If I did want to make assumptions about how most people use the site, I would assume that “most people” don’t ever use other people’s exact coordinates and primarily use the site either to enter their own records or to casually view, ID, and enjoy interacting over other people’s records without actually needing to go there? Do you think that assumption would be incorrect?
Of course, there are many things I don’t know about how people use the data. And I’ve asked for clarification several times and haven’t always gotten it. In the clarifications I have gotten - stuff like “mapping of wetlands, creating species lists of small towns in Vermont, fine scale spatial ecology” = they seem like uses that are already well served by project creation and/or formal data requests. One of the big advantages of both of those is that they are limited in scope and require one to attach one’s identity and purpose onto something - which poachers of course would be reluctant to do.
It’s the ability to anonymously lurk and steal data that could facilitate poaching, without ever having to state intentions or even be a part of the site, that is scary to a lot of us who have worked with groups that fight such poaching and have seen how such individuals work and the effect they can have on populations.
4 Likes
Honestly I’m not sure how to explain the way that 2km obscured coordinates would ruin my use of the site. I don’t currently use the data for scientific studies, there’s not even any concrete or practical thing about my use that would change with obscured coordinates. But I really enjoy being able to look at a map and see observations pinned with a few metres of accuracy. If that became meaningless, I would enjoy using inat less, my usage of it would probably decrease and become less enthusiastic, and eventually I might leave altogether.
I’m also still not understanding why your suggestion is focused on obscuring all observations. The problem for you seems to be that the current obscuring system introduces too much inaccuracy and so you don’t want to just add more species to the obscured list. If that’s the issue, why not suggest multiple levels of obscuration, to be used depending on factors like how threatened the species is, the kingdom it belongs to, etc? That would solve your problem without introducing a new downside for all other observations, wouldn’t it?
4 Likes
I would be happy with multiple levels of obscuring. It would take far more work but could certainly be a good solution. I think it’s likely that other people who have an issue with iNaturalist’s current practice could also be brought on board with the right application of such a tiered system.
3 Likes
Thanks @jonhakim for your suggestions and thoughtfulness on this topic, and sharing perspectives from outside the iNat community. We know there are strong feelings in all directions on how locations should be handled.
I’d like to chime in with some clarifications as a staff member.
- We aren’t at this time considering adding any additional scales of obscuration. We are aware that for some circumstances the obscuration cells are too large, and for others they are too small. For the time being, 0.2 x 0.2 degrees is what we have to work with. We aren’t ready to introduce new levels until we can better clarify the current system.
- We are actively working on how to better protect obscured data that really needs to be obscured, without breaking the normal functions of the site.
- We are also thinking about other ways to improve the dialogue around how to decide which species should be obscured where. This has been greatly informed by the exhausting process we’ve gone through in Canada, which is not scalable.
- We can’t include species in place lists and still protect obscuration. We’ve made exceptions for Standard Places. Most projects use community created places, which is why obscured data don’t end up on their lists.
- It’s important to flag taxa if they may need to be obscured. Please do this for any taxa you are concerned about.
- Traditional projects and user-to-user trust are the primary mechanisms for getting access to hidden coordinates. That is our most scalable “data access request”. The only other way in which we currently share hidden location data is in the context of the iNaturalist Network. We are working on another process, but it is currently in review with the legal team.
14 Likes
I am being somewhat harsh, because I think pursuing the idea that “everything should just be obscured to 2 km accuracy, and if you don’t like that, you can fill out a formal request” is a non-starter, and if you continue to insist that it can, too work for everyone, you will have people just straight-up shouting at you.
It sounds like the essence of the problem here is:
- We need a better framework for determining which taxa are “poachable” or otherwise in need of obscuration, because it doesn’t always map well to conservation rankings. This is a problem the community’s aware of, and for which we still don’t have a good solution: working with the Canadian conservation agencies has, as Carrie said, not scaled, and it’s led to a pretty rancorous forum thread.
- Obscuring to 25 km more or less has to break many projects in order for the obscuration function to work and not be gamed by people of ill will, so people are reluctant to mark taxa as obscured in the face of threats.
I think tiered obscuration would be the best approach to the second problem, although it would somewhat exacerbate the first problem (now we’re deciding not just how to obscure, but how much to obscure), but alas, it sounds like it’s not in the cards for now.
It sounds from the first bullet of Carrie’s response that obscuring everything to 2 km may not even be technically feasible in the iNaturalist system at present, but I think the idea is socially infeasible for the following reasons:
- It basically inverts iNaturalist from an “open stack” to a “closed stack” model. At present, we treat data as open unless there’s good reason to make it obscure. Given that the vast majority of observations in iNat, as far as I can see, are of taxa that are not at risk, I think this is the model that makes the most sense. As Jay said, it’s hard to describe the exact harm that’s being done, but I think my library analogy may throw some light on it: the feeling of browsing in open stacks in a collection is very different from filing out a request for a particular item or set of items that can be brought to you, and the latter definitely limits creativity and serendipity. To throw out a quick example, the other week I took some friends to a nature preserve in the mountains. By using iNaturalist, I was able to compile a partial plant list that we could check off and expand as they went. If all the existing observations there had been thrown around by 2 km, that would have been much more difficult to compile and I probably wouldn’t have tried.
- It’s not clear how the infrastructure for a “formal data request” would work. Is it being made to individual observers (who may or may not still be around to respond to the request)? Landowners? iNaturalist site functionaries? Basically, it sounds like you’d like iNat to be sort of a crowdsourced equivalent of traditional government natural heritage databases. As Charlie said, it isn’t clear that model is actually working very well for conservation; it would generate a huge amount of clerical work for someone; and even that relatively minor barrier of entry looks pretty significant when I have to do it every time I want to see if anyone’s found goldenrod at the creek down the street.
- From my experience sifting through the stream of observations to be identified, most of the data on iNaturalist is not high-risk stuff in fragile natural areas: it’s relatively common species, or at least species that are not going to be poached, taken in people’s backyards, along streets, or on trails. Yes, the number of observers and users of iNaturalist keeps growing, but the number of observations is growing much faster. I’m not (yet) convinced that single iNat observations are driving inappropriate traffic to individual sites as opposed to things like press coverage of “superblooms”.
I do appreciate that you’re coming at this from a position of good faith, and I would definitely like to see herps protected and herpetological experts feel that they can participate without facilitating poaching. But I think a proposal to do that is going to need to have a way more narrow scope than “everyone’s observations, ever”.
8 Likes
I would add to this, the issue is both what needs protecting / obscuring and where does it need protecting ?
This at least partially comes down to have an agreed reason for obscuring. Is it to prevent poaching, is it to limit and minimize traffic, is it something else ?
For example, the province I live in is over 1 million square km (to put in perspective for Americans, it is 60% larger than Texas). Just one example, provincially, I can think of a dragonfly species which is really quite common - so long as you go 200km north of the population centres. Provincially it is not threatened in any way. There are hundreds of thousands of square km where this species is relatively common and unthreatened.
However, in the populated areas, there are perhaps 3 or fewer known breeding locations. To protect those, do you obscure at the whole province level ? Do you obscure in just those counties ? Is there any point when the locations are well discussed and revealed on the forums and message boards dedicated to odonata in the province ?
6 Likes
Jim mentioned briefly but like him, I have personally use iNat to follow up sightings. For instance, when I visited the Bay Area recently, I was using iNat observations to hook me up with recent sightings of certain species. In most cases, these are non-native species I am trying to resolve for Jepson, but it can also be native species. This relies on a location that is as exact as possible as even a colony can be hard to find when the location is inaccurate by even 100m.
And I can’t lie, I’ve used iNat points just to look for species of interest too. In some cases it has been the only way to track down a location of sighting. What this experience has taught me is that it is pretty easy for anyone to “poach” a lot of sightings, but the ease doesn’t necessarily equate to species of concern.
2 Likes
precise locations are critical for katipo conservation, as they have such a restricted habitat. Here in Poverty Bay Gisborne NZ, we have a colony that is maybe 200m by 20m in area, and a proposed walkway through the dunes would have wiped it out. It is a fairly significant colony in that it is at the southern extent of a cladal variation, the furthest know southern population of black only katipo on the East Coast. This has significance for potential future climate change studies as well as ecological indicator values.
This colony would not have been known about if it weren’t for an iNatter (me) who used iNat observations by someone else to identify habitat that was likely to have them, which would not have been possible if I didn’t have access to the precise pin locations of that other observer. Since that initial search, the obscuration of katipo has come into effect, and the environmental consultants who undertook the survey of the area for the proposed walkway did look at iNat, but didn’t “see” the katipo pin locations I had made, as they are now obscured and showing as scattered throughout the Poverty Bay area. It is such a small colony that when they undertook their physical survey they didn’t pick up on it’s presence (it has a look-alike that it is easily confused for). My sharing of the location of that colony has created a small community here in Gisborne that now frequent that area of the dunes, and we are removing weeds that are encroaching toward the dunes, as well as being a presence that deters the motorcyclists that are decimating the dunes.
Get more people into nature, valuing it and learning how to interact in a low impact way, and you will increase the “police force” that guard it, and reduce the market for any destructive poaching…
13 Likes
I definitely agree with your point on ethics and other potential problems it brings up. I also agree that cases which are sensitive, such as nesting sites, should be obscured. I try to have as little effect on the natural environment as possible, and I would like to think that most users of iNaturalist do the same thing, as people using this site should be much more likely to respect nature. I 100% agree that the search is a major part of the experience and fun, but I still want to be in the best place to make my search. I am trying to narrow down a spot where I am most likely to encounter a species, not an individual animal. I would certainly not go to find an individual if it could have a detrimental effect on it or the environment it calls home. If I am looking for a Cottonmouth, for example, I could know that I need to look in the SE US, but that wouldn’t really give me a good chance of finding one. But if I knew that they are commonly seen on this one trail in the Everglades from iNaturalist observations, I could go there and probably find one. Again, I really don’t think this is a problem for a majority of species or observations, but sensitive species or individuals do need to be obscured. Perhaps curators should be able to obscure observations on a case-by-case basis.
2 Likes
First off, thanks for bringing this up and for suggesting a middle ground, whether or not it is implemented.
I’m curious as to how these folks are contacting iNat. I’ve answered pretty much every email to help@inaturalist.org over the past few years and I can’t remember seeing many addressing this.
Is this specifically related to iNaturalist? As far as I can tell, all of those species (I’m also including Terrapene carolina ssp. carolina) are auto obscured on iNat in Canada at the very least, (the skink is not obscured in the US) I can’t find any public true locations for them.
It would be great to have specific examples of this, like URLs of the actual observations. They can write to help@inaturalist.org with those. A lot of people think the publicly available coordinates of an obscured observation are the true coordinates. This was exacerbated by a bug in the iOS app that didn’t differentiate the two on the map, but they now have different icons.
In general I think the biggest conversation is deciding which species should be auto-obscured, although obviously that to can get contentious. It would be really great to have land managers who actually deal with poaching to chime in on what is and isn’t at risk by having its public coordinates available on iNat.
8 Likes
I agree strongly with your statement that we need to hear more from land managers who deal with poaching. The people I know who deal with poaching issues are here in south/southeast Asia where iNaturalist isn’t common enough to have any effect yet, but there are likely places in North America, Europe, and possibly Africa where someone could have formed an opinion on it already.
On those three quotes it should be clear that I didn’t make any of those statements myself, they came from three different individuals, none of whom I know. Just want to be clear.
Another person in that same thread did give a specific example later:
“There used to be an amazing spot in the San Gabriels for Arboreal Salamanders. Then, somebody found one there and uploaded it to iNat, public record. Ever since then, on the first rain, or really any rain, if I went there, there would be all kinds of people there including Troops of Boy Scouts, turning over everything they possibly could. Each person I asked said they were looking for salamanders. Now, it’s really difficult to find any in that area.”
Now, Arboreal Salamanders are by no means a “sensitive species” or “poacher friendly”, though they are not easy to find in Southern California and are rather habitat restricted. People take them home as pets but probably don’t sell them to pet stores in any meaningful numbers. As a conservation issue it’s probably not a big deal. But when people who are not careful (and to a lesser degree even people who are careful) search for a species like that, they do inflict habitat damage, and when lots of people search in the same spot the habitat damage is cumulative. It’s akin to the vegetation-free areas around big trees.
So in terms of those “common but locally vulnerable” populations like that, do we consider it an issue that we might move attention from being dispersed across an area to being localized on a single spot which may then be damaged as a result? With 10 million people here in Los Angeles, it only takes a very small % using iNat in this way (and then telling their friends) to bring a ton of pressure on a specific locale in a way that really wouldn’t have happened in the past by word of mouth alone.
I think one reason herpers tend to have a unique perspective on this issue is because we can see the aftereffects more clearly than most others. The habitat damage and localized population effects that occur from pressure being placed on a herp population are more clearly visible (due to low mobility on the part of herps and their tendency to be found under or inside things in many instances) than they are for at least some of the other taxa. That doesn’t mean those other taxa aren’t negatively affected as well, just that it might not be quite so visible to the naked eye.
7 Likes
Hm. What about a system sort of like the checkboxes in the DQA, where someone (maybe just curators?) could flip an observation to “obscured”? The original observer could then have the option of de-obscuring it, but it would allow for interventions where someone has failed to obscure a record they really should have done. (I think if you just allow observations to be obscured without recourse for the original observer to deliberately veto it–maybe they were keeping it open for a project or study and are willing to take the calculated risk of it being seen–it will make people upset, especially if someone gets overzealous hiding these things.)
(Technically, there is a way that anyone can emergency-obscure any observation, but it breaks things and would make people upset if it were being done regularly.)
2 Likes