robryk @robryk@qoto.org

@TechConnectify

I think I have a hypothesis what might be going on.

People's model of when overinference-from-omission happens might be that it happens ~only when the listener is maliciously trying to misinterpret[1]. Then they might (a) put it in a "troll behaviour" bucket (and assume that you meant nonmalicious listeners) (b) see less of it by virtue of avoiding interactions with people who do that.

I would be convinced my hypothesis is likely false if those people who think overinference-from-omission doesn't happen didn't find "malicious misinterpretation" to be a common troll tactic.

[1] A not unreasonable way of arriving at such a model is being presented with malicious people who choose that strategy of being obnoxious when faced with direct communicators.

@TechConnectify

I wonder whether part of this has something to do with what people who know you as "the guy who makes punny videos about random subjects" expect of you. I'm not sure how you could test that without a somewhat unreasonable amount of effort (with that amount of effort: start another fedi account and compare ratios of different kinds of responses between them).

@b0rk seems to often have a related problem, where people start unhelpful discussions under her posts (it's different insofar it's mostly about posts that ask for a response explicitly). See https://social.jvns.ca/@b0rk/112128313009699377 for her description of the problem and how she deals with it (tl;dr when asking for responses be very explicit about kinds of responses that are desired).

@mzedp

For people who speak English as a second language this might be strongly predicated by what their native tongue does. E.g. in Polish "ekskluzywny" means ~only the former, with something that's less of a loanword used to mean the latter ("wyłączny").

@rampagerslife

Watching Sendung mit der Maus is not really a way to learn German on its own, but it does help a lot and is imo interesting.

@kravietz huh, first time I consciously notice them. Do I get it right that when the whisker touches the coil the payload detonates?

@TechConnectify

That's surprising. I'd expect people who communicate directly to see that _more_ often, because it happens most often when the sender is a direct-communicator. I can totally understand that (a) they don't have a model that explains why that occurs or (b) they have a model that convinces them that it requires some malice for it to occur, but I'm surprised by people not observing it happening at all. (The way lack of good model manifested for me was by assuming that my model is less accurate, which probably would cause me to be much more wary of making general statements.)

Re accuracy above everything else: might it not be completeness as opposed to accuracy? Drive for accuracy alone shouldn't trigger inferring-too-much-from-omissions. Another hypothesis I have is that overinferring-from-omissions happens to a similar extent in different places too, but drive for correctness is what makes it visible for you. (I'm pretty sure this should be sort-of testable by comparing rates of different kinds of nitpicky responses, but it's too late at night for me to think this through.)

@TechConnectify

Ah, I see and that does unconfuse me.

@TechConnectify

In order to reduce the amount of typing, let me call these communication styles direct (only statements matter) and indirect (statement choice is intentionally used to convey information).

It seems to me that the original problem you describe is predicated on the listener assuming indirect communication is happening (otherwise they'd not make ~any inferences from omissions). This makes me think that groups that have a strong norm that communication is always direct should be mostly free of that problem (it might occur when someone joins who's unaware of that norm, or if someone finds it hard to assume that this norm is actually followed).

Do you want to say that the original problem (inferring too much about the speaker) occurs when the listener is a direct-communicator?

NB. I think I'm skewed quite far in the direction of direct communication, both due to personal preference and a different reason you might find interesting: it's way easier to ensure that everyone gets the same message in the direct style. In the indirect style it's hard to ensure that, and it's comparatively easy to ensure that people do understand the message differently. I think that nonadversarial communication is ~always improved by making more things https://en.wikipedia.org/wiki/Common_knowledge_(logic), so I think that making everything more direct is strictly good.

@TechConnectify

My experience is that nonneurotypical people tend to ignore information-via-choice-of-statement. Did you mean to say the exact opposite? If not, then I'm confused by what is the thing they struggle with (interpreting things that communicate via choice of statement? but shouldn't that make your original problem less severe by pushing the norms towards communicating only directly, lest you be not understood?)

BTW. The way I usually explain the statement vs choice of statement distinction to people is by showing/telling them about the "How to irritate people" by John Cleese.

@TechConnectify

A problem is that lots of communication works by having the listener infer not from the literal statement, but from the speaker's choice to utter this statement from all the true statements they could have. There are some groups that have norms against doing that, but IMO fedi is too large to even try.

@zkrisher @esther@strangeobject.space

Public transit ticket machines in my city have a mode for blind people where they read out choices that can be selected by touching one of the four corners of the screen. Is it something that you think is usable, or does it have some problem I'm not seeing?

@sophieschmieg or e.g. one's complement fixed point :)

@mattblaze @sophieschmieg

I expect the common understanding of the icon to outlive common familiarity with glass thermometers by at least a decade (just like it did for a floppy the save icon and the movie camera icon).

@_dm what incident are you referring to?

@Gordarnit @MJCarty

It's important what the company (potentially implicitly) expected of them. After all, you can say the same about every single one of your colleagues and some of them might be your friends.

@MJCarty I wonder whether this will end up being a thing that's recognized by labor law in some form (e.g. like unions).

@HeavenlyPossum

I think that many people associate anarchism with the things it disavows rather the things it avows. For instance, I don't intuitively associate those behaviours with anarchism, even though I associate anarchism with them.

My oversimplified mental model of how society shapes behaviour of people in it is that there are two significantly different ways it always does so: one applies to people who have reasonable amounts of empathy (even if they are prevented from abiding by it by e.g. terribly circumstances they are in) and the other applies to ones who don't. There are very few latter people, but societies have to somehow deal with them (v. serial harassers at e.g. conferences), and using the same approach for both either doesn't deal with latter well, or makes for an unpleasant society for everyone else.

If we take that assumption, an interesting part of the approach is how to decide which approach to use. This is where often some kind of presentability comes in, and this allows for exploits. Assuming I'm not speaking total bollocks, how would you describe this part in a large anarchist society (large in the meaning of low rate of repeat encounters of same people)?