Thanks for the pawpaw info!
1 Like
Speaking as someone who is relatively new to the site, the software is not intuitive and some sort of on boarding / mentoring process would be welcome. My first several observations were made were of trees I planted, because they were a known quantity and I wanted to understand how the AI works. The iNat app on on my phone displayed “wild/cultivated” with a check box. No clue whether checking the box meant wild or cultivated. I’m sure no matter what you do, you will get pictures of elephants in zoos and peoples’ cats marked wild, but some of us are trying and need more help.
11 Likes
Of course and, if relevant, I can confirm that many users thought iNat was just an alternative to the other automated plants identification apps. That means they were totally unaware of the potentiality to work as a global geodatabase to map living (wild) organisms and the existence of the community.
I was told to provide to the users the informations why their observations were to be flagged as captive/cultivated. As regards, let’s imagine how a newbie can feel when someone tells him that, somehow, his observation is not fitting certain “rules” he wasn’t aware of.
5 Likes
I’d like to see this implemented as a feature:
As developers have stated, for the (hopefully near) future it is planned to provide better onboarding for new users - but I am having a hard time imagining that without re-thinking the ‘casual’ category, much can be gained by this.
And as has been discussed here, users who want to get an ID for their plant will not mark them as cultivated although they know better, because they don’t want to make the observation ‘invisible’.
So, here is my feature request:
To motivate people to mark observations as ‘not wild’, provide a separate category for these, including a Research Grade-equivalent (nothing new so far) AND, when these observations are marked as ‘not wild’, have the CV making suggestions only for cultivated plants
The AI has already been trained on casual observations, and similarly to the ‘seen nearby’ feature, marking an observation as ‘not wild’ will toggle the AI to suggest only species from that pool.
Thus, for people who are interested in the ID and already know the plant is not wild, it is a motivation to first mark them ‘cultivated’ and by this, in the long run also the CV might be improved, because it will learn to recognize and differentiate these two categories.
Also, this will not effect the CV suggestions for areas, where those commonly cultivated plants are growing in the wild, because there, the ‘seen nearby’ feature would work as usual.
7 Likes
The checkbox says “captive/cultivated” (which is still confusing for many).
6 Likes
This is one of my pet peeves on iNaturalist as well. I could easily make it my full-time job to track down and flag obviously cultivated plants that are not marked as such. My specialty is Gesneriaceae, and if anybody can figure out how to discourage people from posting photos of potted, half-dead African violet hybrids (often without any flowers) let me know. I’m not sure what these users are expecting from posting such observations in the first place.
I think the problem is bad enough to seriously damage the credibility of iNaturalist when such observations make it to “research grade” (which they very frequently do). Potential solutions? First, I think having two identifications make an observation “research grade” is setting the bar far too low. I see many, many cases where the original observer simply agrees with a wildly wrong identification posted by another user, and then bingo, research grade. The ultimate result is that “research grade” means absolutely nothing because it’s applied far too often to inaccurate identifications, and such observations can stay this way for years until a more knowledgeable user comes across them (which becomes harder and harder as the number of observations in iNaturalist grows and grows). So set the requirement for “research grade” to at least 3 or preferably 4 identifications (barring any disagreeing identifications).
Many of these observations of cultivated plants come from new users who are unfamiliar with iNaturalist, so clearly they need a little more guidance and maybe monitoring when they’re new to the community. But there are some users who do this very persistently and post hundreds of such observations. Maybe such users could be restricted (if not outright banned) from posting new observations if they have been given a warning?
As far as plants growing in pots, which nearly always indicates cultivation, is there any way to tweak the iNaturalist algorithms so it will recognize these plants and automatically flag them as cultivated?
Or perhaps iNaturalist could be trained to recognize observations of species that are far outside their natural or expected range, and flag those as special cases (or questionable identifications) requiring special attention?
But of all my ideas, I think setting a far higher bar for “research grade” is the most important one.
1 Like
iNat AI alreasy recognises observations outside of current range, it is a new feature.
RG can’t be made harder to get as there’re many groups where you will never get 3rd id on iNat. As ider if you see a new user agreeing with your species id – mark this observations as “can it be improved > yes”, so it will stay at Needs id.
1 Like
A friend is posting some very interesting cultivated plants (and marking them as such!) in order to share the photos with others. I think that’s useful, especially because he posts odd things, not just the 12,000th African Violet. Most people who aren’t absolute beginners and post cultivated plants are looking for identifications.
7 Likes
Interestingly, in my enviro-ed days, I made an observation about what kids see. Little kids, early elementary: we would begin our walk in the woods, and within just a few feet, one of them will have picked up a fallen live-oak leaf with a tiny apple-gall on it, and asked what it is. Bigger kids, middle school: we get through our entire walk, and none have noticed the fallen leaves or the galls. I am not sure that observation is learned so much as too-often unlearned.
6 Likes
I’d say multiple things go in play here, young kids are still in exploration phase and they easier notice something on the lower level just because they’re lower, teens are in phase “we’re adults, not care about anything and just want to go home”, even when they seem engaged.
3 Likes
Hey excuse me! I’m in the middle school age he describes, and I have trouble going on walks with my mom now because I keep stopping to photograph plants! Insulting! (Just kidding, I’m not really insulted. But not all teenagers are like that!)
2 Likes
Medium teen is like that, nothing personal, everyone expriences at least part of it and everyone at some point thinks he’s already an adult when it’s not true, it’s normal.)
2 Likes
I think it’s likely that experienced ID-ers are scaring away non-experienced users with sometimes gruff insistence on marking things as cultivated. I wish more seasoned iNat users would keep in mind that people adding plant IDs want to know about that plant! That’s where their curiosity lies. If we immediately demote their observation, and then it subsequently never gets identified (because there is such a bias against identifying cultivated things), then of course we have discouraged that user. That’s my experience from sunny Los Angeles, where most of the plants (native or not) you see here are irrigated ones!
6 Likes
Is it still considered cultivated if its an invasive species thats growing but wasnt planted by a person and it just grew on its own?
I agree marking things as cultivated where obvious is really helpful. This is certainly an issue.
I’ve always been interested in this issue, because I like urban plants, both planted and “wild”. I think that Nathan Taylor and others make some good points that the system may discourage new users by treating cultivated observations as second class. I’d like the more experienced users to consider what it’s like to be a new users from an urban area (where most humans now live). When we interact with new users from this population we have to understand that 90% of the plants they see around them would be considered cultivated. This doesn’t really apply to any other group of organisms they see–most urban insects aren’t captive, most urban birds aren’t captive, urban mushrooms aren’t captive, urban squirrels aren’t captive (whether native or not) and so on. We should not blame the new users for this issue. There’s inherently a challenge in explaining why a decades old tree they see in an urban park, which has withstood the local climate and pests for years, and which is intimately connected with the other living things in that park should be treated different than a Rock Pigeon or House Sparrow or gull. Well the difference is that tree was planted fifty years ago by a park employee. But how are they supposed to know that? We know it’s a commonly planted non-native species (or not a native species) but how would they know that! That’s the whole point! And once we mark it as cultivated it likely will not generate ID’s from more experienced identifiers. Consider another scenario, two new users go to the park nearest their apartment, one takes a photo of the hummingbird, the other takes a photo of the flower the hummingbird drinks from. One is considered “part of nature”, the other is “captive”. It’s a bit of an abstract distinction, isn’t it? I’m not saying it can’t be made, but no wonder it’s confusing! Now I’m not saying we shouldn’t mark cultivated plants as cultivated–we absolutely should where appropriate, it provides critical ecological context. But I don’t think we should be surprised that this is an issue–and we have to meet new users where they are at, not at some level of understanding we wish they had.
6 Likes
Clownish, ouch! Where does this assumption come from that people only do research on non-cultivated plants? If you are studying any phytophagous insects you need to know about their host plant. If you are studying mycorrhizal fungi you need to know about the plants they colonize. Observations of cultivated plants have been incredibly important in our understanding of phenology and how climate change is affecting phenology (see Japan’s cherry trees, American lilacs etc.). The problem here as many have said is “research grade”. It’s just not a great way to classify observations. Not all research grade observations are good for research, not all casual observations are worthless. As others have said, the researcher (and peer reviewers) bear some (not all of course) of the responsibility for filtering and interpreting the data. This is not an unsolvable or hopeless problem–it may be hard to find a solution that everyone loves, but the iNat system can and should change as issues like this crop up.
5 Likes
hi @TheWoah this issue is covered by the iNat help page (posted in thread see quote).
To answer your question, no it’s not considered cultivated.
3 Likes
Oh ok ty I was just wondering since In my neighbour hood we have alot of I think its either agave or aloe that has grown from bits that were planted there and now they are growing on their own
I’m not convinced the current implementation is not the one a majority of users would prefer. There is a fundamental conflict between what established users would like on the site (which in my reading of this over years of being a user is to minimize the amount of cultivated stuff, and not have it ‘cluttering’ up the site and identify pages etc) versus getting new users onboard.
7 Likes
Honestly I don’t see this as much as a problem as simply a missed opportunity. Cultivated plants exist, and they have huge (and will continue to have) immense impacts on global ecology. The site has all of the technological tools to generate useful data about these plants but then stops short of encouraging users to help it do so.
As is, there are probably millions (if not now there will be in the future no doubt) of observations with crisp photos, accurate geolocation and date being hosted on servers (and being paid for) but for which the system is discouraging users to identify and engage with. Why wouldn’t iNat want accurate information on the taxonomic identity of those plants to go along with the photos and other data? The answer might be that “we don’t want to encourage more of them”. If that’s really the overriding concern there are probably more direct ways to deal with it. But I don’t think we need to. I don’t see why “cluttering” should be a concern. Having cultivated plants on this site shouldn’t detract from your experience, you can filter them out, or just skip them. Besides if there were an incentives to actually identify observations of cultivated plants, perhaps more people actually would (we’d probably attract more identifiers too!), in that way you wouldn’t have to look at so many of those pesky Pelargoniums with “needs ID”. Educating the public about what’s native, what’s endangered, what’s invasive, how plants impact insects and wildlife, those would all be priorities for me over teaching them first and foremost about the cultivated vs. not cultivated distinction.
Maybe you think that cultivated plants should be the mission of another app or project–which is valid but then one of the things that people love about the site is the opportunity to see plants, fungi, birds, insects, etc. all in one place. But like I said, I love the site as it is, I just don’t see why it limits itself in this way.
5 Likes