Hello! I made iNatGuessr as a bit of fun and I’m glad to see it appreciated by the community! There are plenty of ways it could be improved but adding features like leaderboards, multiplayer, duels could add a lot more complexity to the app. For me as the creator I like its simplicity, it doesn’t have a ‘back-end’ it just pulls data from the iNat API and displays it in a way that feels like a game!
This thread has prompted me to think again about some small improvements, like I think the UI could be a nicer and more mobile-friendly. I also think the scoring is rather harsh and could be a lot more generous. It can be quite disheartening get quite close but still be a long way off 5000 points!
Right now my biggest problem with it is that I kind of lose interest after a bit each time I play due to not really having anything to work towards. Leaderboards, multiplayer and duels would all be really nice additions.
Ah… but if iNatGuessr was too engaging then you wouldn’t come back to the site to add IDs to observations. ;-)
I tweaked the URL to just show me plants within Tigridieae and now I’m unreasonably proud of this accomplishment (which may have been achieved after several less-impressive rounds).
With the standard game play, if I’m reading the code right (a big caveat), it gets an initial random observation and then does a second search to find all observations within a radius of 10 km. Let’s say you stick with the standard 12-image game—the code then selects 11 additional observations to make up the set of 12. So, all of these could be within a few meters of each other, or some of them could be up to 20 km from others. (Also, bear in mind that lots of observations on iNat have imprecise/incorrect locations.)
If there are fewer than 11 observations within 10 km, you’ll see fewer photos, which is another possible location clue. If you use the “seeded” version of the game (which returns a predictable set of images for competitive play), then (I think) the images chosen will be from the most recent 12 observations with the radius.
We just discovered a very funny variation: since you can put any URL search, you can also play with just your own observations. And let me tell you from the first try, it’s not that easy if you are us :)
We set out to try to achieve a clean 25000 from our observations - and the game seems to do anything possible to prevent that :) The most insidious thing it did was to start with an observation that was obscured - in such case, it started at the random location where it shows on the map, but we did not have any other observations there, so it showed just the one, making us think “this must be somewhere where we were accidentally, because it shows only one observation”, which was completely wrong.
We eventually did the 25000, but after many tries!
I really love iNatGuessr but I just cannot get it to work with a custom URL. It loads for a bit and then the game starts but neither the images nor the map display :/ Does anyone else have this problem?
I wanted to do one with just my own observations but nope, doesn’t work :'(
The photos used already have a location attached on iNat, and many people travel and take pictures in places other than where they would live, if someone wanted to find out where an observer lived this game would not be helpful for that. Also the standard game modes are not limited to one person but one location and chooses observations from a radius around that location.
It is a good practice on iNat and the forum to assume good intentions, and I suspect that nearly everyone playing this game did not have that thought cross their minds.
Slightly off topic but:
I do understand your concern with publicizing your location online especially being one of the younger users of this platform, but that being said, when observing organisms that do not have a cosmopolitan range, even privatizing all of your observations as you do is not too helpful in obscuring your location. Just from a quick glance at your profile I can tell you are in the NE USA and probably in a 400mile radius around washington dc.
So if you obscured instead of privatizing in your case it would have the same effect.
Estimating from species ranges on species you have observed. I’m glad I’m not that accurate, but the species you have observed do place you in the NE quadrant of the US.
To feasibly play the off–the-shelf version of the game set to “only yours”, you need to have observations scattered around the world, because the code does not just select an observation at random. It first selects an area at random and then picks a random observation in that area. I guess this was done so that the game isn’t statistically dominated by a few super-observed regions.
This function does not get used if you give it a fixed area instead - so you can try using a custom URL set to the area where all you observations are. Alternatively, you can easily download the whole thing from the repository, save on disk and play by opening the HTML file in a browser - and then you can edit the JavaScript to change its behavior.
