Identification Etiquette on iNaturalist - Wiki

Yes, I think it’s more like people are in their own bubble, sort of feeling like they can talk to each other and no one else will be able to see their conversation. Sort of like the new people who don’t think anyone will see their observations.

2 Likes

I’ve been guilty of that myself if it’s unresponsive observer etc, but I was thinking more about the times when the observer is actually trying to be a part of the conversation, but being ignored. It got me to thinking about the convos I’ve had, and could they be interpreted that way!

9 Likes

Hi all,
Great topic and discussion, thank you.

This may be a tad off topic, but since we are talking ID’s, I think it might be useful to use different font and or text color during the initial algorithm ID phase when an observation is being input. That is, when the algorithm states “we’re pretty sure” then maybe that could be in bold. Similarly, if it states “we’re not confident”, maybe that could be in red or with a stop sign or something. Keeping along these lines, is there a way to prevent the algorithm from offering a species suggestion that doesn’t occur on the same continent or within some large number of miles?

Just a few thoughts on initial ID’s. Feel free to move this to another thread if I’m too far off topic here.
And, I again offer my many thanks for all the fun and learning I’ve had at the hands of iNaturalist!!

5 Likes

I think this would be a great feature request if one of the moderators could move it to that section of the forum!

1 Like

That could add a huge processing burden and slow down the list being displayed. I am not saying it needs to be fast, but there is potentially some adverse effects to consider with these suggestions.

See these affiliated feature requests and topics:
https://forum.inaturalist.org/t/computer-vision-should-tell-us-how-sure-it-is-of-its-suggestions/1230
https://forum.inaturalist.org/t/add-a-flag-for-frequent-computer-vision-identification-errors/1905
https://forum.inaturalist.org/t/better-use-of-location-in-computer-vision-suggestions/915
https://forum.inaturalist.org/t/species-suggestions-for-the-wrong-continent/789/

3 Likes

I think it is a good idea to utilize any field guides you have as well.

5 Likes

I have been at fault here, listening to someone who I thought knew what they were talking about with regards to an ID, which was blatantly wrong. The repliers her on iNat kept at it, and I am thankful they were so persistent. I ended up researching it myself and learning that my well meaning but very wrong friend was incorrect. I usually just accept an ID which is different from what I have listed because I figure there are many people out there more versed in the sciences than I. *My advise for stubborn peeps, be patient try and steer them towards validating resources and keep at it, try not to take anything personally.

5 Likes

This comes up over and over and over in the State of matter Life observations. I have been taking this literally, but as others point out (and I agree) it appears that great numbers of app-users are just choosing something from the drop-down list of the Computer Vision suggestions and not really even looking at it or thinking about it. Lots of everyone’s time is involved when some people want to go with the observer’s intention (such as figure out the plant if any plant ID was entered and there is a plant in the photo) and others want to identify something that they believe was really the intended subject. Should these app Computer Vision IDs be treated differently than as stated in paragraph 11? If yes, should the observer still be asked first? There doesn’t seem to be any consensus in how to treat this particular group of observations, so I’m just hoping we can get one here so we’re not working at cross purposes.

3 Likes

My thoughts would be yes to both questions. If it’s reasonably clear that the Computer Vision ID was accepted blindly, and the photo(s) are ambiguous as to the intended subject, treat it as if the user has not (yet) expressed a preference for which organism gets identified. In that case, you can either query the user first and try to get a preference from them, or just go ahead and identify one organism and leave an appropriate comment. Mention which one you are identifying, and let them know that if they prefer a different one, you will be happy to come back and modify your ID accordingly. Then see where it goes (if anywhere at all…)

BTW, the wonderful “elephant or daisy” example was actually provided by @tony_wills as an edit to the Wiki. But the way this forum works, any and all quotes from the Wiki post get attributed to the original poster (me in this case). This is probably because it would be too complicated to figure out which editor(s) contributed the material quoted from a Wiki.

5 Likes

I like this.

3 Likes

Once I noticed that placeholders exist and where, I started saving them as a comment. Since I was mostly moving things out of a year or more of Unknown Jail, I don’t think much harm done. In several cases it alerted the OP of the issue and they re-IDed.

They should put “(placeholder:whatever)” with a different colored background to make it more obvious.

8 Likes

While far from an expert, I would assume the OP is following their own observation if they have any interest in it. I will generally tag them if a long outstanding problem exists and they still seem to be active, such as naming the correct genus but from the wrong kingdom from 2 years ago which some have posted correct IDs a while back, but still needs a push to get it out of “life” jail.

3 Likes

Be aware, too, that the placeholder is lost once you make an ID… with no apparent way to get it back or find out what it was, so copying in the placeholder to a comment is a good practise!

5 Likes

Actually you can find out what the placeholder was by looking in the .json for the obs. It should be species_guess I think. But looking this up isn’t something most people are going to do since it is inconvenient and hidden.

3 Likes

Actually… no you can’t… try looking at the .json after you have added an ID…

I tried that, and it did replace species_guess with the formal ID. But I think that was because species_guess shows the last ID entered by the observer. If you enter a placeholder ID to species and someone comes along and IDs it to kingdom, I think you can still find the placeholder ID you gave it as long as you don’t add another ID to it. I made this observation with placeholder ID “sillyate” and I think if someone else adds an ID species_guess will still be sillyate.

Check the .JSON now that I have added an id

https://www.inaturalist.org/observations/26540879.json

species_guess = sillyate

2 Likes

Cool! My tests didn’t involve others putting IDs, so I wonder if that is the difference

1 Like