"Cunningham's Law"; deliberately misidentifying things to get expert attention

I have only run into this problem once, and will not mention names, but in the group I work on a person deliberately misidentified something, knowing it would get the attention of an expert (in this case me). This is bad enough, but then it was made RG by someone who didn’t know better. When I corrected their ID and explained it, they admitted to using “Cunningham’s Law” on a regular basis here. Apparently this is a thing where the quickest way to get an answer on the internet is to post a wrong answer rather than ask a question. Have any of the experts (or anyone else) run into this practice? It’s bad enough as is, but since it was made RG it would have slipped through the cracks without me noticing until I re-curated the entire group had I not seen the observation alert. This is also sort of a curator question, as I am now questioning whether I should suspend the account for this behavior, especially given their rationalization for doing it. What do you make of this behavior on iNat. I find it unacceptable on all fronts and completely counter to its mission.

It seems something like this could exacerbate most other problems relating to bad ID’s being reinforced, trying to correct them, and false RG ID’s getting sprinkled throughout iNaturalist. They were doing this to further their own iNat Project which seems to make it even worse.

Mainly I’m just concerned how prevalent this practice is and whether other people have run into before, or if this is sort of an isolated incident?

12 Likes

Speaking solely on my own behalf, from a curating perspective, I don’t think it is suspension worthy. Suspension without warning should be reserved for the most blatant violations of site rules, or things clearly specified (such as posting commercial spam) as worthy of immediate suspension.

If they continue to do so after receiving a warning, then a case could be made in my mind.

10 Likes

Thanks, I’m inclined to agree. Have you ever run into this before?

1 Like

Intentional misidentifications, yes, but by trolls or students assigned to use the platform who dont care about their assignment etc. Personally, I can’t recall intentional misidentifications to get it in the face of experts, but I’m only seeing a very small fraction of observations.

2 Likes

An intentional misidentification is inappropriate and I’ve heard several active users, curators even, say that they employ this technique. I think a warning is warranted, especially if it’s done on other people’s observations and not just their own.

11 Likes

Was it an intentionally wrong identification, or an intentionally reckless guess? I’ve done the latter (with awareness of Cunningham’s Law!) and don’t think it’s necessarily bad practice. Pointing out that an intentionally wrong ID can be misleading should get them to stop.

6 Likes

I cant speak to this specific example, I did not interact with the observation or user.

I’m not sure how to even attempt to differentiate, unless the user admits to it between being intentionally reckless and just being wrong. I’ve certainly incorrectly ID’ed things that I thought I knew what they were but was mistaken. But they were not a case of me knowingly taking a chance I had it right, I was just wrong in my ID (did not know there were other species I should consider, overlooked a field mark etc)

2 Likes

I’ve seen messages about that on forum, imo if they know an expert it’s easier to tag you other than add wrong id, it’s an edge case of “don’t be afraid to make a mistake”.

3 Likes

They admitted to intentionally misidentifying it to get attention to the observation. Furthermore they said it was their standard practice to get ID’s quickly for their iNaturalist project.

1 Like

That was basically what I told them: mentioned the issues it causes, and said to tag me next time and not do that.

5 Likes

I would never knowingly use a wrong ID, but admit a couple of times I knowingly put the AI guess nonetheless, even if I wasn’t very sure it was right, because I know it is easier this way to get a proper ID than just sticking “Fungi” (apparently the most forgotten category on iNat) or another high-level taxon. But I only did so with the reasonable probability that it might be, indeed, the right answer. I still have a lot of “Fungi” or “Bryophyte” :)

8 Likes

I won’t deliberately add a wrong ID. But I will (thoughtfully) accept iNat thinks it’s this
That triggers comments and discussion. If the expert is kind and tells us why it is, or isn’t - then we non-experts can learn another link for the next round of unknowns.

10 Likes

I definitely wouldn’t suspend. But I would keep an eye on the account to see if there were other problematic behaviours.

1 Like

I often tell newer users to put in a rough ID, rather than leave an observation as ‘Unknown’. Using this idea, I stuck in an ID I did not know earlier this week, and after a fascinating discussion it was Identified to Genus level ( https://inaturalist.ca/observations/70948778). I gained nothing from the process except for some knowledge and satisfaction about the process. I have never come across an observation where I felt that someone was ‘gaming’ the system. If the people I generally work with need help, they ask for it.

1 Like

Also, tagging experts often doesn’t work. Many times I tag someone knowledgeable who did an ID on my observations, thanking them and asking nicely if they can check my other unsolved ones of the same type (e.g. mosses). Almost always I see no answer nor additional IDs. It’s not their fault, I know they’ve lots to do and I’m not the only observer here! I perfectly understand. Just to say that, however, there is a bit of a frustrating situation sometimes.

I’ve seen casual iNat users suggest doing this occasionally. I assume most people doing it will consciously do it as a reckless guess, or even just going with the CV suggestion when they know it could well be wrong, rather than intentionally incorrect. I think there may be kind of a sense on facebook bird identification groups that you’re more likely to get a correction/confirmation if you give an initial suggestion, perhaps even a faster reaction if your guess is wrong. I don’t think that logic works for iNaturalist because the process of identifying is so different, to the point that giving a species level initial suggestion if you aren’t confident is generally more harmful than beneficial. It is great for your species count though…

2 Likes

I find it sad that you even need to point that out!

This is about expectations, and many (particularly new) observers have an expectation that their observations will be identified quickly to a very fine level. They need to be reminded or made aware that it is an unrealistic expectation. They are not paying for the identifications, and there is no contract that says they will receive an ID within a given timeframe. We all dump our observations into a big pile and we all take out observations to identify as and when we wish to… it’s a fun hobby, not a JOB!

As for deliberate mis-ID’s… just leave them identified incorrectly, maybe bookmark them into a folder that you will come back to in 6 months and check to see if they still need correction, and then their efforts to “cheat the system” will be defeated :) A more confrontational approach might be to ID them even MORE incorrectly (make them all Homo sapiens as evidence that a human was involved in the incorrect ID!), still bookmark them to come back to in 6 months for re-assessment of your own ID, and then they might re-think their strategy. I know, two wrongs don’t make a right, but sometimes a little “push back” is needed to drive home a point.

I am “progressive” in terms of how I identify, and I do encourage others to be so as well, as it can stimulate a conversation that wouldn’t otherwise happen and create learning opportunities… but that is the critical point here. It’s a “could it be this” scenario, not a “if I label it wrong, someone will (more quickly) fix it” one. That means that if there is no response or agreement after a sufficient timeframe, you go back and pull back the ID to a more coarser and safe level. This active maintenance of ones own IDs is critical to the progressive approach.

It’s also about being respectful. Tagging is effectively asking someone to give you their time preferentially over everyone else… I do seriously have issues with a lot of the tagging that I receive, in that there appears to be no genuine reason for the “urgency” that tagging seems to demand. I can understand if someone posts 20 observations and one or two of them they are particularly curious to find out what they are because they have never seen anything like it before. Or they are doing a presentation and were hoping to use a particular observation as a talking point, and having the ID would be particularly advantageous, or if there was concern over a potentially invasive species or toxicity, or just any other genuine reason for the need for a preferential treatment as far as IDing is concerned… but too often it just seems to be “tag the top 10ers so you get it ID’d faster” with no actual advantage gained from the effort. I mean, if it’s ID’d today or in 6 months, and either way has no actual benefit… why not just let it happen when it happens?

When I tag someone, I generally give a little statement or question as to why. “@someone, do you know anything about the spines on these?”, and especially if I am requesting effort from them. The only time I will tag without any other text (eg “@someone”) it is usually because I have dealt with them a lot in the past, and I am simply alerting them to something they will likely find interesting.

Just read back through this for typos, and I can see I’m in a particularly grumpy mood today! Maybe tomorrow I would write differently…

5 Likes

I try to follow this also – if I’m asking a particular person to help out, it seems like the least I can do is ask politely, perhaps with a complete sentence. Unless, as you say, you “know” the person well enough to feel like that is unnecessary.

6 Likes

Perhaps it’s a bit different with moths, but the folks I routinely work with frequently ask for confirmation - a second pair of eyes, so to speak. I also do ask them if they have any insights into the ID.

2 Likes

Hah. Didn’t know this behavior had a name. I’ve done this some. Mostly it’s in popular groups where I know IDers are very active. Yes, a wrong wild-assed guess will get correction IDs faster than a correct ID will be confirmed. I think they get more satisfaction from correcting mistakes than just slogging through IDs.

3 Likes