Obscuring geolocation by date?

FWIW, one of my observations is a threatened species, but I made dozens of other observations on the same day at the same location. Those other observations have the geolocation obscured (my choice) but not to the point where unscrupulous people would not be able to look at my observation pattern and figure out where the threatened species is. If other people make similar observations of threatened organisms but don’t even obscure the location of other organisms they spotted around the same time, it might be even easier to figure out where the threatened organisms are.

In other words, is it possible/desirable to obscure (at the very least) the geolocation of all observations made on the same day that a threatened organism is observed?

1 Like

I moved this post from the topic about Canadian conservation statuses.

OK. Could you change the title to “by date” rather than “by day,” though? When I first read the title, I thought it referred to daytime, as opposed to nighttime, and that didn’t make sense to me!

–> Never mind, I found that I could change the title myself.

This is on the admin’s radar based on past conversations. Otherwise, it is exposing a vulnerability in iNat that may not otherwise be obvious, so maybe this thread should be removed.

OK. Thanks.

Ah, was just following the language used internally: https://github.com/inaturalist/inaturalist/issues/2037

i think this actually mirrors the whole to-obscure-or-not-to-obscure discussion a bit. is it better to educate people on a known vulnerability so that they can take action on their own, or is it better to keep people in the dark for fear that some minority of them will use the knowledge for evil? in this particular case, i think the secret is already out of the bag, and that the best path is to educate. (there’s no reason to wait for a technological implementation of obscure by date when one can already effectively do that manually.)

really, i think we should go even further in the education and really stress that any coordinates posted publicly on a site like this, even if obscured, are probably subject to be discovered one way or the other. so if people really need particular coordinates to be absolutely unavilable, the only reliable way to do that is to not post them in the first place.

It’s been proposed and discussed but it basically breaks the site for high volume users, bioblitzes, and place based collection projects so thankfully it hasn’t happened yet. You can batch edit and obscure all your observations of a day near an observation of concern though. When I have something I want well obscured but am still willing to post I remove or change the time/date and upload it first when I get home. That gets around some of the issues. But for some things it’s better just not to upload them to inat at all, as others have said.

An on site alternative would be just not showing date of upload or calendar for auto obscured stuff. It’s not perfect but it doesn’t break everything.

1 Like

Too easy of a workaround since you can (currently) use the observation ID to approximate the date/time of upload.

So yeah,

This is the better way around it, assuming your other observations from that day are relatively spread out on the map.

That’s why I upload it first when I get home. :). It just gets the first number of the day. It’s not perfect and if it’s something I’m super worried about like ginseng or a rattlesnake I just don’t add it to inat. To me that’s far preferable to ruining all the other observations of the day - I care a lot about precise location and don’t want to lose that on the range maps and such. In the longer term maybe the way this system works should be changed.


Yeah, I was saying that just hiding the date/time uploaded would be insufficient, and that manually futzing with the date/time uploaded (posting out of order from when you actually observed it, like you and I do), is necessary.


Oh, yeah. Sorry. :)

I think I need a clarification - I’m missing something.

I thought that the current functionality obscured all observations by that observer in the same day that a threatened taxon is observed. As @bouteloua this seems to have been done here (and my reading of the code seems to support that):

Was this not rolled out? Or am I misunderstanding the request?

It is not implemented, hence still being marked as open on github


from what i see here (https://github.com/inaturalist/inaturalist/pull/2196), it sounds like it is “Currently an opt-in test only available to staff”.


If it ever does get rolled out it should remain opt-in only. I’m probably not the only one living at the outer edge of an administrative area (NY state,) creating a situation where many plant species common here are ranked S3 or lower on a state level. I observe auto-obscured species most days I’m out iNatting with zero actual need to be obscured, much less obscure another couple hundred obvs from the rest of the day.


i’m the last one who’s going to defend the way obscuration works in the system even as it exists now, but it seems a bit pointless and super messy to me to roll out obscure by day to everyone while asking them to opt in. if you’re going to roll it out, you really need to just bite the bullet and roll it out.

i would prefer that they decided not to roll it out though. i don’t really care that obscure by date would make data less useful for some. rather, my take is that going down that path just makes it harder to get to the right path in the long run.

Not knowing what, if anything, is actually going to be implemented, my understanding is that there would be two pieces to this.

  • personal date geoprivacy at the choice of the user (opt-in)
  • automatic date geoprivacy when obscured taxon geoprivacy exists for any of the taxa observed by the user on that day

Speaking just as another iNatter, I 100% agree here, especially for the involuntary piece. I can maybe see the argument that it’s a necessary part of robust taxon geoprivacy for species super vulnerable to public knowledge of locations. Problem is, most of the species currently with obscured taxon geoprivacy do not rise to that level. Either that needs to be cleaned up first, or date geoprivacy needs to be a separate setting on conservation statuses that gets locked down to a much shorter list of critically vulnerable species – those for which ne’er-do-wells would go the extra mile to analyze daily observation patterns.

Otherwise, if a bunch of my non-vulnerable observations are going to be involuntarily obscured, just because I happened to also see a critically vulnerable taxon that day, I probably wouldn’t be alone in the temptation to either fudge the date of the vulnerable observation to prevent that from happening, or just not post the vulnerable observation at all.

I would much rather see this option implemented:


this other option, which i’ll call datetime obscuration, at least in the way that it was proposed in the other discussion, shares some of the same problems as location obscuration. so even if you apply datetime obscuration to all observations in a given day, in addition to location obscuration for a single observation that day, that does not really address the problem that obscure by date is trying to address. you could apply datetime obscuration to all observations that day and location obscuration for all observations that day, but then you’ve essentially come back to obscure by date.

interestingly, even though obscure by date is more effective than the proposed datetime obscuration in certain respects, the datetime obscuration can be a step down a path that leads to a more effective solution than obscure by date, i think.

You may not, but I’d hope the fact that many of us do care a lot about that would carry some weight. If I’m in a group deciding where to go to eat and a few people don’t really care where we go but one person really wants to go to a particular place, it usually makes sense to accomodate that, right? Especially if the preference is driven by food allergies, or something. My observation data is, in a sense, allergic to obscure-by-date.