I’m aware of the comments page, but I don’t consider it either easy or practical because the only way to access it is if one knows the url. Searching is very rudimentary (and again, one must know what to add to the url in order to do so). There is no way that I know of to e.g. search for comments of observations in a particular taxon, or region, or time period, and viewing search results is unwieldy.
1 Like
Nope. Quantity without regard for quality seems to be the default throughout most of the internet (especially now that SEO is all the rage). I will venture to say that even discouragement isn’t necessarily a bad thing, if it is experienced by someone who has overemphasized quantity to the point of neglecting quality.
3 Likes
I’d argue that discouragement is always a bad thing, regardless if it’s a newer user or a veteran being discouraged. Millions of people use iNaturalist for different things, it feels a little silly to judge what does and doesn’t have merit to be on here based on your personal and limited scope, especially when both quantity and quality are important in their own ways.
Personally I can’t imagine anyone enjoys the feeling of being discouraged, especially with the upvote/downvote feature proposed here. There are already plenty of options in the DQA to filter out ‘low quality’ observations that aren’t as harsh or crass, setting them to casual if they’re missing crucial information like the date/location/image, or checking “good as can be” if an observation realistically cannot be identified further.
7 Likes
Well, let me frame it in terms of the SEO business I mentioned. It is getting really hard to find useful information on some topics because of the proliferation of superficial fluff, or worse, outright erroneous information which floats to the top of my browser results because it is SEO. I have no qualms about saying that people need to be discouraged from posting such content – regardless of whether they enjoy the feeling.
Now as to whether iNaturalist content is in the same category or should be subjected to the same standards, well, that’s something we can discuss.
2 Likes
I think the biggest problem with discouraging the posters of poor-quality photos is that many of these people are beginners and some of them will go on to post high quality and useful observations later.
11 Likes
I don’t see how comments on the decline of the internet and society in general move the discussion forward in terms of providing insights into ways to reduce or weed out low quality iNat observations.
Since you think it would be worthwhile to incentivize quality over quantity even at the price of discouraging a few users – what would this look like concretely? How is quality to be defined? What is the threshold for “good enough”? What determines whether an observation is useless or valuable? At what point are a user’s observations considered to be “too many” and superfluous? I’m not asking this to be obstructive: Complaining is easy, but unless these questions are answered it’s hard to figure out any kind of constructive solution.
Certainly I’ve had my doubts about whether events like the CNC result in people uploading observations merely for the sake of uploading observations (competing to finish with the most observations) and whether a different format of competition could be designed that would encourage a somewhat more thoughtful approach.
I’m not sure exactly what this would look like. I believe some of the local CNC organizers have emphasized aiming for a high species count rather than a high total observation count. I don’t know whether or not this has been effective, but it does at least ask people to think about what they are observing instead of just uploading as many photos as possible.
2 Likes
Yes, so do I, rarely, but it does happen. But I’m thinking here of obs where I really don’t know enough to understand which experts might be able to help. A practical example… I quite often come across photos (often quite good) of marine/shore organisms that are, sadly, way out of my comfort zone. Sometimes I’m ashamed to say I can’t even decide if they’re animals or plants or something else besides, but I’m quite sure that there’s someone else out there with a different expertise who would be able to take them through perhaps to species in a second. An “interesting” flag could really be a great help in this… or is there some other way I haven’t come across yet?
1 Like
I have witten a chrome extension to help downvote the bad observations which have IDs which cannot be improved and remove them from the needs ID pool: https://forum.inaturalist.org/t/new-chrome-extension-to-select-that-the-community-taxon-cannot-be-improved/47239/13
I’m actually pretty much with @jasonhernandez74 on this. There’s active discouragement and there’s passive discouragement and I’d say that placing the accent predominantly on quantity (as it feels at the moment) is actually a passive discouragement to quality. If the identifier/observer ratio was more balanced, perhaps it wouldn’t matter so much, but as it is I can imagine particular a new observer genuinely interested and willing to invest time and energy in their observations could easily get discouraged when they fail to get any feedback for months, maybe even years.
I’m not saying that “poor” observations should necessarily be downvoted or given a “label of shame”, but that a way of giving priority to “good” observations where the observer really is making an investment in iNat seems a pretty sound way of helping people truly interested to “engage with nature”, while perhaps at the same time raising the usefulness of iNat at a scientific level.
I often feel that perhaps the majority of new users posting “poor” observations are using iNaturalist as if it were Seek and actually don’t expect any interaction. These could easily be “discouraged” at source (registration) by pointing them towards the sister platform that could perhaps be more suited to their need for an immediate no-strings ID. But that’s a matter for another topic.
2 Likes
…maybe? I’ve got over 100 followers and I have no idea who any of them are apart from a few friends and a handful of people I recognize as users who live in my region. None of them ever interact with my observations, and I suspect most of them aren’t even active users. I am quite active on the social side of iNat but that only merits me a little extra attention from actual friends, not really anything from parasocial followers I’ve spoken to once or never.
which is great! An issue I’ve found though is that when I’ve done lower-level IDs and the observations reach species, I’m counted as having helped on that species when I have zero knowledge.
I get tagged for Euphorbia plenty, and I can help some there, but for example, I am the top observer in the world for mushrooms in at least one country just because I added “Agaricomycetes” IDs to a bunch of stuff…
so, please, it doesn’t hurt to look at the user’s page to see if there’s any indication of what they usually ID.
I concur, the DQA is enough to filter out the truly unusable observations. I want to encourage new users to gain more skill and learn what makes an observation more helpful, without making it seem personal. Granted, I’m usually looking at observations which are years old, but if it’s been less than a month I tend to leave an explanatory comment when I downvote in the DQA.
there’s some people who look at “State of Matter: Life” specifically, and others who specialize in marine bio like @hfb who don’t mind a tag. Or, for anything that looks plausibly shell-like, @pierrenoel …
1 Like
Well, it seems to me that this:
is indeed condoning active discouragement of certain behaviors (i.e. downvoting certain observations or similar), or there would be no reason to be concerned that people might not enjoy the feeling.
By contrast, passively discouraging certain behaviors by encouraging and rewarding different ones is unlikely to have the effect of making users feel discouraged (i.e. it would not make them feel that their observations are not valued or their participation is not desired).
If there is some feasible mechanism to make it easier to reward high-quality observations, I would be all for it. But so far I haven’t seen any suggestions that would avoid the issues I mentioned above.
2 Likes
I don’t think many do that. The sadder ‘poor quality’ is … I am here on iNat for the weekend of the challenge. And now I’m gone. Leaving wrong IDs. Lots of cultivated obs.
If we can encourage those new CNC and GSB iNatters to come back, to stick around, to respond to notifications.
The quality of iNat rests heavily on - 130 users provide 25% of the IDs. We need to make iNat IDs sustainable. They are not automagical - CV suggests - but an actual human must ID.
@lynkos Marine life is a gap on my list. I won’t tag in a taxon specialist for … no idea where to start beyond marine life. Another reason why I miss @joe_fish
iNat’s stats focus on
How many observers
How many obs
Identifiers and IDs not so much.
1 Like
I don’t quite understand that this tool does.
What I would like to be able to do is have a filter in the identify modal that could be set so that it would show me species with e.g. <X (could be 10, 20 etc., great it if could be adjustable) observations in a particular place. I would want it to catch the older observations of these uncommonly observed taxa as well as the newer ones.
For me, this would show “interesting” observations. I think it would tend to show the truly uncommon stuff that has been found in a place and it would also show the incorrectly IDed observations (possibly CV facilitated).
This would be a more efficient way of catching CV-facilitated errors than what I have been doing, which is quite idiosyncratic. I just happen to notice a species I haven’t heard of and then I check to see how widespread it is on iNat. Sometimes it’s a species that is common outside of my focus area and for the most part is correctly IDed. Sometimes it doesn’t occur in NA at all. A recent example is artemisia indica, which as far as I can tell has not been documented in NA. It looks similar to the very common artemisia vulgaris. There were something like 600 a. indica observations in NA (only a small proportion were RG, I think about 40-50). I fixed a bunch, and other IDers helped me and we got it cleaned up, but I can’t help thinking that there is a better way to deal with situations like this than me just stumbling on them.
Morus nigra is another problem - not yet cleaned up
https://www.inaturalist.org/journal/bouteloua/67407-mulberries-in-north-america
1 Like
Try the projects Beach Blobs and “Beach Finds & Washashore.”
3 Likes
I think there’s a way to get more-or-less what you want on the “Explore” page. If you filter for a place and then click on the “species” tab, this essentially gives you a list of all species seen in that place, sorted by descending number of observations. If you scroll down far enough, eventually you will find the species for which there are only a few observations. I use this periodically to check for CV-inspired wildly out-of-range bee observations. Note that I recommend filtering for a general taxon (say, a particular insect order or family) as well as for a place, so that the species list is not too huge.
2 Likes
The problem with using projects is that there doesn’t seem to be a way to mark observations reviewed in a batch like I can with the “Identify” tab. It gets really time consuming to open each observation that I know I will not be able to ID just to mark it reviewed. And without doing that, can’t tell which observations I saw before.
You can select a specific project in the identify tab actually ! I underlined in yellow the field you can use (sorry for the French UI)
4 Likes
If you open the observation, or if you’re seeing it as a thumbnail in “Identify” mode, you have the option to simply mark it Reviewed without applying any name to it.
That’s the problem: before @petitcrabe explained it, I didn’t know that a project could be viewed in “Identify” mode. And opening the observations one by one was what I said was too time consuming.
1 Like
it’s not perfect, but it’s supposed to help point out when there’s an observation that has been recently identified as something that hasn’t been seen before (in the area), or at least hasn’t been seen in a while.
for example, the circled record below is the latest species-level observation that fits this criteria in my area. there are strong signs it’s probably an inaccurate ID because when i look at the “ID via comp vision” column, i can see that the ID was CV-assisted, and i can also see in the next column that the ID was done by the observer:
then if i go to the observation and look at the map for nearby observations, i see that there aren’t any unless i travel more than 500 miles in any direction:
3 Likes