Looks like a good boy
A bur marigoldā¦ Iām saying that because Iām smart and not because it happened already and I saw it.
3 Likes
I have zero drawing skills, about all I can do is press a shutter button. I suspect itās a @kueda original but Iām not sure.
2 Likes
Yeah I think search mole and fail whale have been on inat maybe as long as I have
1 Like
Not to discount other observations but the target observation looks well presented, worthy of its lofty position.
4 Likes
Curious about whether there are any up-to-date projections when the āregularā 9-digit barrier will be surpassed.
@tiwane @kueda Any nice graphs somewhere? 
Unchecking āverifiableā, the explore page currently shows me 95788005 total observations, so 4211995 to go to 100m. Last year, November and December combined averaged 0.683 times as many observations added per day as October of last year (3329294 in 61 days vs 2478242 in 31). Assuming that trend holds, and it shows 3111676 observations added in October of this year, then we can expect the average upload rate for November and December to be about 68500 obs/day. At that rate, we expect 61.5 days to reach 100m, which would be January 4, 2022 at 5:10 AM EST.
To get an estimate of the statistical uncertainty, I can assume all the quantities I used have errors of about the square root of the number of observations and then propagate the errors. Because the numbers Iām averaging over are so large, that ends up only giving me an me an estimated standard deviation on the rate/day of only about 498 observations. That propagates to a standard deviation of about 9.5 hours.
That gives me a 95% confidence interval of roughly 10:40 AM EST on January 3, 2022 to 11:35 PM EST on January 4, 2022.
Of course, not all uncertainties are statistical. Thereās no guarantee the ratio of daily observations in Nov+Dec vs Oct will be the same as it was last year, although the rate of uploads seem to generally follow a fairly predictable seasonality. Also, some observations will of course be deleted between now and January. Things like day-of-week effects in the upload timing and the holidays will also add uncertainty.
Based on observation number-number of observations reported, about 4.4 million observations have been deleted. I have no idea what the distribution of when things are deleted looks like, but if it looks about the same as a % of when they are added, then we can estimate that about 143000 of those were deleted in October of this year, or about 4610 per day. We would then predict 3150 deletions/day in Nov to Dec, giving us a new net addition of observations of 65400/day.
With that, I get a revised prediction of 64.3 days, or January 7, 2022 at 3:50 AM EST. Arbitrarily assuming day of week effects double the uncertainty, I then get a revised confidence interval of January 5, 2022 at 11:00 AM EST to January 8, 2022 8:30 PM EST for when inaturalist will hit 100m total observations.
6 Likes
No, we havenāt made any predictions on that as yet.
I tried to make a similar prediction for hitting 100m verifiable observations, and it is much harder to get a credible range. The difficulty is partly because of the longer timescale, and also because the 2022 city nature challenge is a big wild card that accounts for weeks of normal uncertainty. It appear entirely within the realm of possibility, perhaps even the most likely scenario, that 100m verifiable observations will be hit during the 2022 city nature challenge, between April 29 and May 8.
How many is left to this point?
I see 85182220 verifiable observations on the explore page right now, so about 14.8 million to go.
1 Like
Looks like the site hit 100m observations including non-verifiable sometime in the morning of Saturday, January 8! Right now I see 134,298 observations from the 8th, 9th and 10th combined and 100,128,441 observations, so it happened at whatever time on the 8th 5857 out of the 64704 observations that day had been uploaded, hence some time in the morning. My best estimate from two months ago was within ~24 hours, and within my estimated confidence interval!
5 Likes
The predictions were hilarious! But Iām surprised no one predicted that it would be a cultivated plant in a schoolyard, if you know what I mean.
One month later, we are just about 10.5 million verifiable observations away from the 9-digits event. I just noticed, however, that another important milestone already happened some time back: 2 million observers! I wonder when this happenedā¦
4 Likes
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.