Label observation as an "iNat First"

I was thinking about the reasonable and serious users of iNat, not the gamers.

2 Likes

You can’t disagree with what is the best way to see what the first recognized record is. The first record on iNat that was recognized as being that species. Later, another observation might come around with an earlier observation date, but the first one identified will always be the first observation on iNat to be recognized as that species. Whether or not it’s useful to discern which observation was the first to be recognized as the species is a different matter.

1 Like

You are totally right. I will change then my words. I do not think that the first identified is the best way to look at what the first individual of one species on iNat should be. I would give more weight to the person that uploaded the first image of that species on iNat, because of the stated in my previous comment.

2 Likes

As a person who targets getting new plant species for his state, I can’t disagree with anything you’ve said.

I love this idea! I’d drop it down to first within a 500 mile radius rather than 1000 though.

https://forum.inaturalist.org/t/find-your-first-observations-for-some-place/10471?u=ribabo

One of the things I like about this idea is that it would encourage people to look for obscure species to record, or get into learning difficult taxa. The effect of that I think would be much more valuable than the detriment of a few gamifiers doing dodgy IDs (You’d probably have to get quite into iNat before you noticed these things anyway, and therefore probably not so irresponsible).

They might take up a bit of space unless they were done as little badges/rosettes of different colours next to the RG banner (hover over to get the explanation). I quite like the idea of having all of the above: ‘First uploaded observation’, ‘first identified observation’, and ‘oldest observation’. All of them should be applicable only to research grade obsevration, and should be redeployed if the truth changes.

In short, I like this idea. Whether the programmers like it… I will leave to them!

6 Likes

Out of votes, but supporting!

Surely the solution to confirmation bias with this idea is a simple one? For any observation flagged with the ‘iNat first’ label, an algorithm can be used to push such observations to the top of the ‘Needs ID’ pile with obs recurring at the top of this log indefinitely until such time as they have atleast 2 ID’s from 2 different accounts for the same species

How does an observation become flagged with the INat first label?

Probably it’ll be when this taxon will be first added (or if first one was withdrawn/deleted) to observation and there’s no disagreement with previous ids.

2 Likes

I should’ve said “Hypothetically speaking”, sorry. This feature is obviously not implemented already and judging on what other matters the iNat team has to attend to, may not be for quite some time still

2 Likes

I remember reading about this some time ago but stumbled across it again recently. I noticed that there are 1042 observations “iNatFirst” (no space) and 858 “iNat first” (with a space). :(

2 Likes

There’s another issue you didn’t mention. You mentioned first posted observation and first identified observation, but there’s a third possibility. If I take a picture with an observation date that comes before yours but post it after, which of these takes priority? I could easily go back through old slides and find an old picture from three decades ago, post it decades later and be the first observation…just not the first posting of an observation.

1 Like

Which also carries the risk of an unscrupulous person falsifying the date, since those archival slides have no metadata.

I woudl support a tag for “possibly outside known range” for observations far from other observation, this tag could warn of possible errors and new discoveries, without incentivizing bad IDs

1 Like

we already have data quality votes such as “Date is accurate” to turn the observations to casual

I was not talking about a tag that would render somthing casual, but one that would note out of range observations for the purpose of detecting range expansions

ah, my glass was half-empty it seems :)

Made a Github issue here for our team to iterate on. It will take some design work and other details need to be hammered out. I’m a little worried it would incentivize false IDs, but it might also draw more scrutiny to these observations, so perhaps it’ll be a wash.

4 Likes