Proposal for a new filter option: the 'ID tips' button

General
1

I have in mind a new feature/filter option for the platform.
As this would be a rather substantial addition, before putting this as a feature request I wanted to gather some feedback and suggestions.

The issue:
I would like to see iNaturalist becoming a better place for finding information about how to ID taxa.
– ‘But this is something that has been proposed so many times before’ I hear you say.

Indeed!

This comes up every other day in a forum comment that (apart from better onboarding) we need some information, e.g. on the taxa pages, about important ID features, about similar species to be confused with, having ID keys, a warning sign that this taxon is tricky,… etc.

But the thing with iNaturalist is:
It covers any organism, and worldwide!

Other websites and platforms, focused on a certain region and/or taxon group, can much easier prepare and curate such ID help pages.

But on iNat, you cannot simply walk go to a taxon page and expect to find information for your local area or find a key covering only the species of your region. To provide this information would be a herculean task for a small amount of curators, shouldn’t it remain merely a patchwork of few taxa covered. It would also require regular reviews, to update information or remove non-working links, etc. And it would need to be applied on the various taxonomic levels, e.g. covering the genus and all or a subset of downstream species.

A lot of work, while also not easy to access for non-expert iNat users.

But the cool thing is: so much information is already there on the platform!

IDers leave a lot of valuable information as comments under observations or explain ID features in journal posts. Alas, it is not easy to find currently.

Say, I want to ID a tricky hoverfly genus, and I know there were some expert comments provided under a bunch of observations. I could either reach out to those experts or go through observations of this genus one by one until I find something helpful:

grafik
grafik678×552 85.1 KB

My suggestion:
Have the option to mark observations containing valuable ID information, by implementing a second ‘favorite’ button. Something like this:

grafik
grafik1080×368 78.3 KB

Thus, anyone can mark an observation as containing valuable information, and new filter options would be available (tick box and/or ‘sort by’ field):

grafik
grafik1464×638 87.3 KB

Using the standard filter options (such as taxon & location) can be used to narrow down the search. The more votes, the higher ranked the observation will be.

For example, I photographed a hoverfly in France and the CV suggests several species from the genus Chrysotoxum. I could then filter for observations from France for that genus and check if any observation shows up that contains an ‘ID-vote’. If not, I might widen my search to Europe, or just draw a larger rectangle on the map.

Advantages:

  • With the simplicity of this voting system, it would be a low threshold for the community to participate and I expect that tons of valuable tips and links can be quickly put together this way.

  • It might also incentivize IDers to write more comments or even to publish more journal posts.

  • It also would make it easier – as leaderboards can be misLEADing – to find ‘true’ experts for a certain taxon in an area

The process of finding ID help via this filter is of course a bit of a hit-or-miss at times, but on the other hand you might find information that you didn’t expect to find and also something that is normally not found in standard ID keys and more focused on IDing of living organisms.

To have a certain degree of curation (e.g. when links are non-functional or information is outdated), and to counteract misuse, I am imagining curators having the rights to remove votes from an observation.

Also, I was thinking about possibilities to tweak and refine the system even more, by deliberately having observations stuck at a higher taxon level (e.g. Syrphinae or Ortotrichales mosses) via opting out of community ID, thus providing links and general (regional) information for higher taxa.

Lastly, the ribbon at the bottom of the observation page might also show (nearby) observations of relatives with ID tips.

So, instead of

grafik
grafik1440×172 90.3 KB

We get this:

grafik
grafik1440×184 94 KB

While I am not sure this is something that can easily be implemented, I think this would be an elegant solution to an oh-so-often requested feature and would smoothly fit into how the platform is already been used.

Curious to hear your thoughts!

44 Likes
2

while I like this, this could be a tag or observation field, or a compiled in a journal post on your account

8 Likes
3

I think it’s ok to brainstorm potential ways to execute this, but fyi see this note in the about Feature Requests topic:

10 Likes
4

I do think this is the simplest solution to this problem that I have heard so far. Good brainstorming. Hopefully iNaturalist staff are already cooking up something similar, but if not, maybe this post will help give them some ideas.

16 Likes
5

6 years ago.

6 Likes
6

I like it! It seems like a fairly simple and elegant way to make it easier to find information that already exists, while avoiding the difficulties connected with trying to create systematic taxon guides.

I would argue that this does not fall under “ways to capture comments […] and including them on the taxon page and in identification tools”, since it would neither put this information on taxon pages nor create an identification tool, but rather is merely a new option for labelling and sorting observations.

Some of this already happens informally – many people use “favorites” for observations that include useful discussions and I believe there is at least one project for collecting observations with identification tips.

But – crucially – neither of these directly translate into a general way to find observations that the community as a whole has found useful. People use favoriting for a variety of reasons, not just for ID information, and adding observations to a project requires that one is aware of and has joined said project (tellingly, I was not able to find the projects I was thinking of in a quick search of the project pages, because the project search functionality is so crude).

8 Likes
7

I always tag my observations with helpful comments as “IDcomment”. Of course, I can only do it on my own observations. But for me it has been helpful and easy to use. Sometimes I also include some articles or other links that I found.
I like the idea that anybody can check this button and also vote upon.

6 Likes
8

I like this proposal very much, however, I think it can be generalised to fix another issue spiphany mentioned:

My lists of favourites is a jumbled mess as well. I favourite observations of cool organisms, with cool photos, with useful tips, of organisms I want to remember when I’m IDing, etc…
So I think being able to sort favourites into folders would be great!

It’s basically just your idea, but instead of limiting it to a second favourite button, the user gets to choose how to set the categories.
The drawback is of course, that a button may not be immediately available on the observation page.

5 Likes
9

I do that using bookmarks and folders on Chrome.

But this - being able to label and find again ID tips would be a wonderful improvement. I need a simple way to find this again, as we sort thru taxonomy of Nivenioideae

Only one species in Bainskloof

4 Likes
10

Tthis would allow users to better sort observations that they have favorited themselves, but I don’t think it would fulfill the purpose of pooling existing knowledge and helping users find observations that have been marked by other people as being useful.

If users could set their own favorites categories, it seems like all their votes would have to be fed into a single “favorite” pool, so it would be essentially the same as the current situation, where I can sort by number of favorites and the top observations are ones that users have engaged with, but they will not necessarily be those that have comments or discussions with ID information.

5 Likes
11

That is true. I hadn’t thought of that. Perhaps if the favourite-folders were to be implemented, “All favourites” and “ID tips” could be made standard folders that everyone has?
But that would make my suggestion basically a separate feature request on top of that by carnifex, and it probably shouldn’t be discussed here.

1 Like
12

If everybody used a tag IDcomment as I have said before, we could have something without any changes, of course not as good as the proposal because you can only tag your own observations.

2 Likes
13

Right, but my point was that it doesn’t need to be voted on by the community because they already think it’s a good idea and would like to do it at some point. FWIW they did confirm 2 years ago:

Not scrapped.

Why did you cut out the part of the quote about identification? How would a list of observations with ID tips not be a tool for ID?

One of the glaring problems being, say I use the same approximate ID tip on 100 observations (and also include some other text about the observation in my comment or ID, so they are not true duplicate comments), and then different people check the box saying it contains ID tips. It’s not very useful to have the same tip repeated over and over. Someone else’s tips might be much better but hidden among all my duplicates simply because I choose to add more comments or make more IDs. If I’m not copy/pasting the exact same text it would be hard to deduplicate automatically.

And should it be upvote/downvote? For cases where the ID tips were actually inaccurate or outdated (or duplicated?)? And purely community-driven or would curators be able to remove or nullify inappropriately marked ID tips?

4 Likes
14

Then we have at least one (hopefully useful) ID tip in contrast to none :wink: Which might already be enough in many cases.

And here, I am hopeful (naïvely optimistic?) that the power of the community will sort this out eventually. Similar to school projects with zillions of incorrect, CV-supported IDs or in cases of overconfident, unresponsive power-IDers that need to be corrected by the community, in the long run I believe the useful observations will be floating on top - just because everyone can contribute. It basically is the same system that has already been proven to be functional - because it is the core functionality of this platform.
Also, you can set your filters for a certain area, which might reduce the ‘copy comments’ and also, e.g. if only 2-3 species seem plausible, you can try checking each of them for concrete ID tips.

Rather not, too complicated (or almost non-functional → cf. downvoting annotations…)

Yes, as I mentioned in my original post

5 Likes
15

Because, as I read that sentence, it is referring to requests to compile information from observations and integrate that content elsewhere, in a guide for identification on taxon pages or in a separate identification tool. Merely determining that there is relevant information in comments is not a feature request because it doesn’t tell us what is to be done with this (how it is to be made available); the relevant bit is therefore the second half of the sentence.

Being able to sort observations for ones that have been voted on as containing certain types of information is neither a guide nor a tool that provides tips or criteria for identification – the user still has to do the work of looking through the results, reading the comments, determining which ones contain new information, whether the information is correct/applicable, etc. It is a research aid that helps people look for relevant discussions. It doesn’t synthesize the content of those discussions or help them apply it.

2 Likes
16

While it’s not as good as built-in solution, I made a project for observations with helpful ID tips. If enough people added observations to it, the project could become very useful when learning new taxa or trying to ID a tough observation.

https://www.inaturalist.org/projects/helpful-id-info

5 Likes
17

To me a research aid is a tool. So maybe just semantics here…

1 Like
18

I would expect an “identification tool” to actually help me with identification – to provide some structured guidance through the process, whether in the form of a list of relevant features or something like an interactive key/flow chart, much the way that I would expect a tool meant to assist me with cooking to do something more than direct me to the relevant aisle of the supermarket for the ingredients I need.

1 Like
19

Useful tools can be far simpler than that. Anyway, Tony has been tagged multiple times in this topic so if he thinks it’s important to distinguish or separate carnifex’s proposal from the already proposed “Ways to capture comments and ID remarks that are useful for making identifications” then he can make that decision.

21

When you wrote

  • do you mean in this thread? Because I do not see Tony being involved here yet
2 Likes