Well, with my older photos from scanned slides, data accuracy will often be within a mile (1600 meters). That’s very good, actually, especially with photos of native plants that have probably since been extirpated.
I probably won’t post the ones that would need accuracy circles of 10 miles (16000 meters) but even those would be within the county (which are usually squares about 24 miles on a side, in that area) and for some purposes that adequate.
Accuracy circles like that are comparable to the accuracy one can figure out from many herbarium specimens, and researchers have been mapping from those for decades. Centuries in fact. That kind of accuracy won’t work for every project, but for some they’re just fine.
The big question is obviously: What is this position being used for? Does it matter if a plant is “seen in XYZ Park” or within 10m of the actual position?
I agree 100%. This is how I report my own observations (directly into our organizations database - I don’t have time to put my photos onto iNat). I try to keep my accuracy numbers to a reasonable size, so if I’m doing a very long linear hike, I try to break it up into sections (particularly if I am crossing the boundaries of birding Atlas squares).
I prepared a youtube video illustrating how to do this on google maps, for folks who are submitting observations to our Atlas via spreadsheet. IE. it illustrates how to look at the satellite view and roughly pick out a central lat/long for the area surveyed, and then estimate an accuracy figure for the observations made at the location in question. I had intended to create a similar one for iNat users (who participate in our project), but as so often happens, I got side tracked and it never happened. I really should get on that.
If everyone did this (at a minimum), all locations would at least be accurate, though some might not be very precise. As long as the precision is within a few kilometers, I’m fine with it myself (YMMV). If somebody really wants to get the precision down to a few meters, that’s fine, but as with so many things, we shouldn’t make the best the enemy of the good.
What drives me bonkers are the observations where I know for a fact that the observation location is incorrect, yet the accuracy circle is down to a few 10’s of meters.