well maybe we can’t use iNat to gather that type of data for special status moths that get auto obscured. Do any moths actually get auto obscured? Maybe we shouldn’t be auto obscuring as many things. But that’s another issue and i think what your ‘overzealous obscuring’ comment is getting at rather than this.
.
Bottom line is posting date and time of obscured observations just makes the obscuring nearly pointless in a lot of cases. If we are gonna obscure we might as well do it effectively otherwise it just messes up the range maps and irritates people without actually dissuading poachers or whatever.
Well, i find this absolutely harmful and absurd as well but i can tell you that a lot of these types were hammering away on the ‘un obscured date/time’ thing as one of the reasons that everything ever should be obscured. That is again, a whole other topic, but fixing some of the problems with obscuring takes away some of their ability to push their closed-data view.
Sorry, I’m not talking about auto-obscuring threatened taxa (ie. taxon geoprivacy). I’m strictly talking about user applied geoprivacy. I don’t know about moths, but a fair number of butterfly species in Ontario have taxon geoprivacy. IMHO, there are probably only 2 or 3 that actually should have taxon geoprivacy. I hope this gets fixed at some point, but it’s a separate matter, and it’s enough for us to contend with without observers willy-nilly deciding that they should obscure all their observations.
And for the examples I provided previously, neither species has or requires taxon geoprivacy. Users deciding to obscure these observations would only make it harder for us to learn things from these observations. Sure, in some cases there might be associated observations of other organisms that are threatened. In those cases, our project probably shouldn’t even know about the observations, since we can’t guarantee the same level of geoprivacy provided by iNat (well, we probably could, but it would require a lot of work, and the whole thing is getting way too complicated).
well, in my case, for my house, i either obscure the data, or don’t post at all. So if it weren’t obscured you wouldn’t see it to start with. I prefer the date obscuring too, i appreciate that extra privacy and it means i post more from places such as private land where i am doing field work. I wouldn’t post that stuff at all otherwise and yes the point is that you don’t know exactly when or where the observation was, that’s why i am obscuring it. Over the years i’ve used iNat i have started posting a lot more from my property and from other people’s private land due to increased confidence in obscuring. When i first started iNat a decade ago the obsuring didn’t work very well and i barely posted things from home at all. So i guess you can just assume the obscured observations are data you wouldn’t have at all otherwise. Fine to filter them out of course. I filter out obscured stuff for some things too.
I can live with that. I accept that there are folks who have a variety of perfectly valid reasons for obscuring their observations. For these folks, I have decided that our project probably should not use their data - we probably don’t want to assume the risk that we might inadvertently reveal dates/locations. And if we are not going to use the observations for our project, then I won’t waste my time reviewing these observations.
An alternative to iNat’s “randomization” method of obscuring is a “roll your own” method that has some advantages. Pick a nearby town, or major intersection, and use that as your location. It should be something close enough that the observation is still useful for things like constructing range maps, etc., yet far enough that nobody will know exactly which property the observation was made on. Even though dropping the pin on a town center or major intersection should be a tip off that this is not the true location, you should explicitly indicate that the precise location has been withheld, so as not to send some innocent observer on a wild goose chase. Then set your location accuracy to a value that encompasses the true location. In this way, you are in precise control of how far off the stated location is from the true one, and because iNat doesn’t have the true location, you never have to worry about it being inadvertently revealed. This is what I would do if I had a location I was worried about revealing. And yeah, if I get hit by a truck, then the secret dies with me. But more often than not, there will be at least one or two confidants with whom I will have shared it.
And then any particular research or citizen science project can look at your accuracy figure and decide for themselves whether that falls within the bounds of what they deem to be acceptable. For our own project, I think we’d accept anything within 5 km (I often report my own observations with around that level of accuracy). We might even be willing to accept 10km. But the ~27km that iNat puts on obscured locations is a bit much.
I know. But you see, the folks who decide these things don’t want input from someone like me. It’s all fine and well for me to do their work for them (in terms of reviewing and correcting observation data), but they absolutely do not want any input from me on important decisions like that. I’m not talking about iNat staff/volunteers here, but the conservation folks who are advocating a longer list of obscured taxa.
Thanks for the heads up. I don’t really have any issues with that list. I was under the impression that they might be adding back a lot more species. I can live with that. In fact, I might have been OK with them adding one or two more.
I would not presume to do something like that. Way beyond my area of expertise.
this is what i used to do. It;s worse than the iNat obscuring method as it doesn’t track the observations properly and you can’t yourself track the real locations. And it can’t be automated I don’t really understand why you’re proposing doing this, to be honest.
I wasn’t proposing it as something to be automated. You said you started to posting a lot more from your own property and from that of others because you gained confidence in iNat’s obscuration. I’m just saying that this is an alternative to relying on iNat to provide you with a ready made solution. Personally, as a user of observation data on iNat, I would prefer that observers use this method over the obscuration provided by iNat. If iNat obscures an observation, it isn’t just useless to me, it is a liability. If the observer obfuscates it, and only fudges the location by a few kilometers, the observation is still useful to me. And because I don’t know the true location, I don’t have to worry that I might somehow inadvertently reveal it. I can use the observation with a clear conscience.
I’m advocating that even if an observer gives my project permission to see the real location/date of their obscured observations, we should omit the observations because it is too risky to include them in our project.
i mean… what you are describing is literally what i used to do before they improved the obscuring algorithm. But… i am going to disengage from this part of this conversation as it doesn’t appear to be going anywhere useful.
that could possibly work for self-obscuring, though i think opt-in should be the default, if we can turn off the horrible auto sync we can turn that off. I don’t think it should be possible to turn it off for auto-obscure of rare species.
I think @raymie means manually obscured, it really seems as the thing that bothers people the most, though I don’t want my auto-obscured observations having no date too, most of them are common species with no particular possibility of harm from iNatters.
If you believe that auto obscuring is too excessive you should be advocating for changes to the list of taxa which are auto obscured, not calling for changes to the obscuring process itself which render the obscuring open to exploiting and defeating.
Who you answer to? Cause if me, I am trying to change obscuration list, but I don’t see much of a profit from it until local red lists will be used, having Lapwing obscured and Black Stork open is just funny, anyway we are asking for making current date obscuration optional not because something is wrong with lists we have, don’t remember a comment like that, and again most people are discussing manual obscuration.
It’s a general comment. Manually obscured observations should remain obscured, period, end stop. The user’s wish should be respected regardless of why they have chosen to do it. There should not be any loopholes or configurations they have to further enable (to me this is just hoping the user forgets or does not know about such a configuration so you can continue to exploit the loopholes).
If you believe that an auto obscuring on a particular taxa is ‘funny’ then seek to get that taxa changed, not to allow flaws in the obscuring to remain unaddressed.
As I have written before, it is no more difficult than this, obscured should mean obscured, not obscured until someone knows how to exploit it and defeat it.
Making it optional will provide user a way to actually express if they want it the way it is now, yes, maybe making it automatically on (instead of previous written suggestion to make it auto off) will be enough to not get it fully reversed.
Flags are not reviewed for months, so adding more is excessive, I let old ones be addressed first.
To me, this sounds like an accusation. For what it’s worth, I have already altered the URL of the bookmark I use to bring up observations for identification. It now includes “geoprivacy=open”, so I’m not even looking at obscured observations, much less waiting with bated breath for someone to slip up so that I can “exploit the loopholes”.