As we’re preparing for the City Nature Challenge, we’ll (hopefully) have lots of new users making their first observations. I’m trying to figure out the most important things to tell new users who join our organized walks during CNC. I’m curious about what people think are the most concerning errors made by new users? You can choose up to 3.
Missing info (Date, location)
No ID
Relying on CV for a wrong ID
Photos of different organisms on one obs
Photos where you can’t tell what they’re looking at
Many different organisms in one photo
Not reading / responding to comments
Duplicate obs
Agreeing because ‘Well, they must know best’
Obs of cultivated plants
0voters
And, of course, please comment if you have others.
I agree that’s a problem, but I don’t think it’s one that can be dealt with by educating new users. Hopefully there are / will be ways of detecting deliberately false info by people just wanting to cause problems. For the purpose of the CNC, though, I think we have to assume that people who have chosen to come out on a walk or bioblitz are acting in good faith.
“Photos of different organisms on one obs” is at the top of the list as of right now. It happens all the time of course but in my experience new users get the message pretty quickly. They don’t do this for too long if they do stick around for a bit. The other issues on the list are so much more pervasive in my experience. People will be overconfident in the CV, or use the agree button as a thank you, or not read comments for a long time in comparison, and it can be hard to reach them.
Happy to report that the major issue - 60% - has an effective DQA - please split and then automagically Casual.
For me that pushes the second issue - 34% - to the top. Not responding to comments (for established iNatters remains an issue years later - because we lack ways to MANAGE notifications)
My third issue - missing date or location, which is automagically Casual - is a major issue, since the obs miss the taxon specialists roped in for the CNC deadline. Those obs are effectively disqualified from the friendly contest and may remain in limbo for years. I have only recently managed to clear the African backlog from CNC25.
This year anything Casual / Cultivated / Captive / broken and missing data will be excluded from CNC - so only the missing data needs to be caught in time.
PS while I am wading thru other backlogs, I come across iNatters who were here for CNC. For that one day. Please remind them iNat is like a puppy - Not just for Christmas - there will be comments, and discussion, and a learning curve. Please stick around. Please come back.
Honestly, it was hard for me to choose as well. I think that the one concern on the top of my list, though, is being available for communication. Checking notifications, reading and responding to comments. Participating, to be succinct.
I see it all the time with all sorts of bioblitzes; you have users who sign up for that one single thing, upload a few observations of wildly varying quality, and are never seen again. (I’m not just talking about duress classroom users, either, although they can be a special pain in the fundament.) I think if I were giving the orientation, I’d try and emphasize that this is a really cool and interesting pastime–but you have to give it half a chance.
I think sticking around, checking notifications, and asking questions if something isn’t clear would be the key message I would want to convey to new users as well.
Most of the issues that seem to come with new users are a result of either 1) not knowing how to make good observations and struggling with a complex interface or b) not realizing that iNat is a community and there are other people looking at one’s observations – that it isn’t just an app where you upload a photo and automatically get an ID. In both cases, there is a risk that new users never see feedback which would make their experience of iNat more positive for both them and others.
Beyond that, the specific problems that people encounter seem to vary quite a bit, depending on their style of interacting with iNat, how tech-savvy they are, and how much knowledge they have about biodiversity recording. Some of the mistakes are mostly problems for IDers (uncritical use of the CV, reflexively agreeing to any ID they get); while others are more likely to negatively affect the user’s initial experience of iNat (observations that are missing data or lack an ID are less likely to get seen and thus to get feedback and ID help).
Encouraging users to think about whether organisms they see are wild or cultivated would be the second point I think is important to convey to new users. Captive/cultivated observations negatively affect the experience of both observers and identifiers, albeit for different reasons (observers because they don’t get an ID if they observations are marked casual, IDers because the observations typically aren’t marked correctly). For someone who has never thought about whether plants are wild or put there by humans, it may not always be easy to determine whether something is a cultivated plant or not and quite possibly they will make mistakes at first, but asking them to think about this and why it might matter for biodiversity recording is an important first step.
“Relying on CV for a wrong ID” is closest to my main concern but somewhat misses the mark. Using the CV is encouraged (of course) and being wrong is certainly okay but using the CV to guess a species-level ID is counterproductive since it forces those who come after (the identifiers) to work harder (since it requires significantly more work to turn around an incorrect species-level ID). My advice is: Guess responsibly! :-)
Most of these would be at least partially solved if new users would just read and respond to comments, and be willing to rethink CV suggested IDs. I would put the main emphasis on that. “Identifiers may have follow up questions like which organism in a photo you would like identified.” and “It’s fine to tentatively use the CV suggestion for an initial ID, but if a real person suggests something else they usually have a reason. There is no shame in withdrawing your ID, even if you aren’t sure enough to agree with theirs.” (And explain how leaving an incorrect ID up wastes other people’s time, and may stop their observation from ever attaining research grade.)
I hate using that though, because new users generally don’t understand the DQA, so it stays casual even after they split it. I have to remember to check back in a few days and undo my vote if the problem has been corrected. And I’m not going to keep checking for months, so some that were belatedly corrected are probably still lost in casual.
Data Quality Assessment (see below) - evidence related to a single subject
I added the last line when someone (not so new) asked - what / where is the DQA.
And once split, I delete my comment, ID if I can, remove the now irrelevant DQA - and if the obs is interesting enough may call in reinforcements for the ID.
That is one of the things I check as I work thru problem obs / IDs. Why IS this obs Casual, what is broken ?
I think even more “experienced” users often are still unclear on what an observation is. I’ve had users with 1000+ observations push back when I put a species ID on something because “this observation is meant to be of all three species in the photo, not just that one”. Such a statement demonstrates a deep misunderstanding of what “observation” means in an iNat setting, and if experienced users are still confused, I imagine new users must be completely clueless.
Most of the top four choices require additional actions by the observer, whereas the others can be corrected by the community to a large extent. This makes sense, since the worst outcome for both the observer and the community is the delayed resolution of any issues that keep the observation in the needs-id pool (which might be forever in many cases).
However the simple rejoinder to that would be: then upload that same photo to three different observations and identify a different species in each observation. Hopefully that should clue them into what an observation actually is.
My main issue are blurry or otherwise unclear pictures not showing much details.
I would appreciate if beginners would learn how to make observations with more than one picture eg. showing a plant and some more or less randomly selected details of flowers and leafs.
I’ve been going through the “unknown” observations lately, and I think not adding IDs is a major problem. Going through and labeling “plant”, “spider”, “squirrel”, takes up volunteers’ time when they could be giving more specific IDs on the organisms they are familiar with. Sometimes new users mistake commenting for adding a ID.
I know sometimes it’s hard to give even a basic ID to scat, tracks, weird little invertebrates, eggs, etc, if you don’t know what you’re looking at, but some new users are uploading blocks of dozens of photos with no ID when it should be fairly easy to ID them at a basic level as birds, mammals, fish, or plants.
Another common issue is poor quality photos - zoomed in too close, dark, blurry, trees that are too far away, etc. Plants usually hold pretty still so it shouldn’t be too hard to get a decent photo. New users will have more luck getting IDs if they take the time to get good photos that include the whole plant and some closeups.
but the identifier in turn, can choose to keep up with their preferred taxon. Perhaps at a slightly broader level to retrieve more, and then the easy triage to move on to the right taxon what is interesting and possible for you. I throw unknowns to where I know taxon specialists wait, with their filters set. Insect, fungus, spider, bird …
Give me an honest Unknown every time, not a wrong ID, with support or Ancestor Disagreement - and iNat’s CID algorithm insists that FIVE identifiers must agree. That is a joke.