Well if it really has value, then it should be treated accordingly and not thrown in the “casual” bin together with all incomplete observations missing critical information.
3 Likes
If my understanding of English is correct, ‘casual’ is not strictly a synonym of ‘without value’?
Even incomplete, captive/cultivated, geographically imprecise (…) observations have value to anyone who considers them valuable. The fact that some other obs are “RG” (i.e. automatically exported to peculiar platforms) does not make such observations automatically “valuable”.
Casual is the word iNat uses for ‘Not Wild’
making life even more confusing for iNatters for whom English is a second or third language.
As words, both Research Grade and Casual are judgement calls.
3 Likes
I take it you mean ‘Captive/cultivated’ (as in ‘not roaming/spreading outside of human-intended area’)? (opposite: ‘Wild’).
Similarly, ‘Research-grade’ (as in ‘exportable to third-party databases of mostly academic use’) has ‘Casual’ for opposite (= not exportable, for a huge variety of reasons) - once vetted past ‘Needs ID’.
Too many words can become ambiguous and loaded… but if after a lot of explaining some people overinterpret things and keep equating it with ‘domestic’ vs ‘native’ and ‘valuable’ vs ‘worthless’ (respectively)… there’s little left to do.
Casual also means incomplete… an observation without picture, date or place will be casual, even when the subject is wild… hence the confusion for me.
2 Likes
Yes - this confusion has come up on the forum again and again in the 5 years I have been here. We work with, and around, what we have.
3 Likes
Further confusion is caused by the term “Research Grade” which implicitly suggests that other such observations have no research value. From my own experience this is demonstrably false, as I have collaborated with several researchers in my field who have used databases of cultivated plants in their research. We have discussed using iNaturalist in this type of research, but the barriers to community verification make it very messy, and in general because of how hidden this data is, most researchers in the field are not even aware of the huge amount of data on cultivated plants that is documented here. It could be of significant scientific benefit if it was reviewed properly by the community. While this could still happen in theory, the barriers are such that most interested and knowledgeable people never see these observations. In fact, I know at least one colleague (and master plant identifier) who quit the platform over frustration with this very issue. I would consider doing the same if I knew of another similar platform that dealt with this issue more effectively.
While sharing data with GBIF and similar organizations is an important function of iNaturalist, it is frankly not all that relevant for users. Making this distinction the primary way of categorizing observations does not make much sense practically. The primary purposes of these categories from a UI perspective are to help locate observations that need verification, and to provide a gamified reward for observers who upload good quality observations.
But what is the benefit of denying these features to observers of captive/cultivated nature? The only one I can think of is that a small number of identifiers don’t wish to see these observations in the identify tab. This is easily solved with already existing filters. In contrast, the costs are significant, to the research community, to IDers who may be unaware of a hidden set of observations they do not know how to access, and to observers whose observations never receive community feedback. This is particularly true for the many new users who start with iNaturalist because they are curious about plants they find in their urban neighborhood. I see many such users make a few observations which are largely ignored, then quit the platform–probably because (unbeknownst to them) they were accidentally exiled from the great community that exists here. It is also an issue of equity because not everyone has access to the spaces where wild nature can be observed. If you live in a city with no car, your opportunities to observe wild nature may be very limited–but there is an abundance of cultivated living things in almost every city.
Across this thread, I see several well-thought out and apparently popular suggestions to deal with this issue. I am left to wonder why the leadership at iNaturalist has chosen not to address this problem. Is it an issue of resources? Would these changes require a lot of time or money that simply isn’t practical for the organization currently? Since the community has been agitating for this for 5 years now, it does seem like it should have made its way onto the agenda by now, but if this is the problem, perhaps those of us passionate about this could do some fundraising work in support of this change.
Or is it a philosophical disagreement? While I could critique many hypothetical philosophical arguments here, I don’t think that would be productive, and ultimately, it does not matter. Even if the original purpose of iNaturalist was wild organisms, the tools created here have incredible usefulness outside of that philosophy, and it does no real harm to users focused on wild observations to remove the barriers blocking non-wild uses. While it may have been true that iNaturalist was intended only for wild organisms, this is no longer true of how the community uses the platform, and attempting to block this use is against both the democratic and scientific ideals I feel the platform aspires to. So I hope even those who don’t personally value these observations will still support this change which will be very meaningful for those of us who do.
Thank you for coming to my TED talk.
8 Likes
Using these definitions, it reports that around 44% of the world lived in cities, 43% in towns and suburbs, and 13% in rural areas in 2020. This makes the total urban share 87% in 2020 ([more than 6.8 billion people]
What people see, what they observe, is predominantly plants. And then mostly deliberately planted. Individually we each have our own focus, our own slice of iNat.
There is no point in telling observers to mark it Cultivated - then chucking their obs in the bin with the broken obs. They won’t - they want an ID please - that request is often spelt out for identifiers.
Identifiers will be better able to filter out Not Wild, if the Not Wild can still be Needs ID for us identifiers who filter it IN.
1 Like
My impression that that iNaturalist has been growing so fast that the developers are mainly focusing on trying to keep up with growth, with other projects being lower priority. Planned major changes I’m aware of like the app redesign and the Explore page redesign have been discussed or in progress for years (eBird changes are also very slow so I assume this is just how things go in development, I dunno this isn’t my field).
A complete reconfiguration of the basic philosophy of observation statuses would take a ton of thinking and discussion and planning to implement, so I can understand why there hasn’t been much attention to it.
See for example the share observations feature request which the staff labeled “very challenging” I think for similar reasons.
4 Likes
I had a similar idea and tried to post it on the forums the other day. It was rejected, but I was forwarded to this one. This was my post:
Captive / Cultivated - A “casual” observation rework
I think that the way casual observations have been implemented could be better. I understand most casual observations are not going to be very significant scientifically, but there are situations where they would be such as:
- Photographs of rare cultivated plants in botanical gardens.
- High quality photographs of endangered / rare animals in zoos.
- High quality photographs of domestic animals / exotic pets.
Although the observation itself is categorized as “captive / cultivated” and therefore marked as “casual”, the pictures can still be of value for not only training the iNaturalist AI, but also providing rare images of a species that may not have been observed yet on iNaturalist. Casual observations are lumped in a group together with other casual observations. The category of “casual” can even include observations where the user hasn’t included pictures of the species in question. When searching through casual observations of certain species there is no difference between how an observation containing photos a rare species in captivity is treated compared to an observation where someone listed an observation as a species with no proof or photos provided.
What I would like to suggest is a rework on the way Captive / Cultivated observations are treated on iNaturalist. Allow them to reach an equivalent of research grade as long as they have at least 3 species level IDs. They would still of course not show up on distribution maps as they are not wild organisms. I also thought that instead of a white “casual” marker icon, we could give them a blue icon instead, sort of like this:
2 Likes
@Jarronevsbaru Interesting. It is already possible to filter by captive/cultivated, so I think this boils down to giving those observations a blue label. I would support that, in that it creates a psychological distance between captive-casual and ‘got-a-data-quality-issue’-casual. It would require the filters on explore to work a little differently, so that if you select ‘captive’ it excludes ‘non-captive’ even if ‘verifiable’ is also selected. (that is not the case at the moment because the ‘verifiable’ category excludes captive).
Support for the OP’s original suggestion is built in, because you state that they should be left to reach an equivalent of research grade. I would only demur that the criterion for that status should be no different to normal research grade - two agreeing IDs at species level or >2/3s agree, etc…
I think this would amplify the effect of the OP’s suggestion and encourage people to mark captive observations properly from the outset, helping those who aren’t interested in such observations to filter them out, and vice versa, while also making it more likely that those observations get an ID.
3 Likes
This is an interesting topic - I’ll be honest I don’t want to search my local area and see pictures of houseplants because to me this has no value or interest to me, but I do see that throwing these in bin where they’ll probably be ignored forever is discouraging for people who genuinely want an ID to something.
The problem is for every captive plant people genuinely want an ID on, there’s someone who’s taken a picture of their cat, dog or (much worse) themselves, friends, kids or inanimate objects and uploaded it for reasons I never quite work out. These are never removed, but labelled ‘Captive’. This search shows that casual is littered with junk labelled ‘Homo sapiens’ Observations · iNaturalist, which includes selfies, photos of kids, toys, graffiti and a even bag of Co-op pasta.
Personally I’d like to see pictures of people and human junk outright banned, and reported and deleted rather than end up in Casual, because it gives the message that the ‘Casual’ section = any old garbage, rather than a potentially interesting section of captive organisms that may not have research importance, but naturalists may want to engage with for interest from time to time.
2 Likes
While I can get frustrated with Human observations, I don’t think that banning and deleting them is a good idea. They are the first observations of a lot of users as they are learning to use the platform. Banning/deleting would likely turn off a lot of new users who would otherwise become valuable members of the iNat community.
5 Likes
Maybe ‘Homo sapiens’ gets its own invisible bin then. It’s not so much the ‘is this thing on?’ original selfie that bothers me as such, it’s just that it’s hard for people to see their casual records as anything other as discarded, unwanted dross when it’s relegated to the same area as cuddly toys, pictures of discarded rubbish, gurning schoolkids etc.
1 Like
Identifying an observation as human already renders it casual with no need to use any other DQA option. They should not be marked captive/cultivated to put them into that “bin” though it still seems to happen quite a bit. There are also some older discussions about this issue. It’s frustrating, e.g. for someone who is trying to add IDs to cultivated plants, to have to wade through a bunch of humans marked captive/cultivated.
6 Likes
That ticks all my boxes. As it is, we have a mess. RG obs which are in fact garden or house plants etc.
2 Likes
I may even have been guilty of this in the past, as a well-intentioned act. Maybe “Human” should trigger a block on all other tags then, and after x amount of time with no argument they should be quietly auto-deleted. (I realise I’m grumpier than most on the use of iNat for some to dump any old rubbish off their camera reel!).
2 Likes
There appear to be a whopping 46,832 captive Human observations:
https://www.inaturalist.org/observations?captive=true&place_id=any&subview=table&taxon_id=43584&verifiable=any
assuming I’ve done my searching correctly.
So I think this combo does cause a problem. Some of the observations are ones where the downvote for Wild does make some sense (eg, a plastic flower in a flower pot initially IDed as a cultivated flower, but someone later pointed out it is fake and IDed as Human, etc.). However, these seem to be the minority.
I’m not sure how best to deal with it, and it’s a bit ancillary to the original feature request, but any solution to address the broader one proposed here could take this situation into account since there are a large number of observations. It could also be good to help to prevent these types of observations - perhaps greying out the Organism is Wild DQA if the observation is IDed as Human would be helpful.
6 Likes
It makes me wonder how much server space is taken up by the photos/data for those kinds of humans/pets/houseplants/objects/etc obs. I think one of the good parts of having those obs more hidden from searches/general view is that it doesn’t encourage folks to upload more of those things. Those observations add costs to iNaturalist to maintain without contributing to the intended purpose of iNat. That’s not to say that they aren’t valuable to someone, but just that it was not the intended use of the platform.
For those who struggle with feeling like they are turning users away by marking those observations as captive/cultivated, I wonder if maybe a compromise would be linking to other resources/platforms that might be a better fit. There are lots of local gardening groups on other social media platforms like facebook as well as some broader houseplant groups on platforms like reddit where folks could get help with those types of questions.
2 Likes
47 K obs needs a solution from iNat’s side.
Disable the Not Wild vote for a Human ID ?
2 Likes