mastodon.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
The original server operated by the Mastodon gGmbH non-profit

Administered by:

Server stats:

344K
active users

my thermonuclear take is that the issue here isn’t whether it matters that we call AI “AI”

but that Mastodon’s culture is poorly socialized and people here have a hard time doing constructive conversation

This is frequently lamented as “replyguyism,” but I think that’s imprecise. Something happened along the way to mislead people here that derailing the conversation makes friends

simonwillison.net/2024/Jan/7/c

simonwillison.netIt’s OK to call it Artificial IntelligenceUpdate 9th January 2024: This post was clumsily written and failed to make the point I wanted it to make. I’ve published a follow-up, What I should have said about …

The way to make friends on the internet is, in the words of Twitter viral dude visakanv, to support the OP

Which is not to say be a sycophant. Sometimes entertaining a bid means to take it seriously enough to disagree. But to do so in a way that PASSES THE VIBE CHECK is what matters.

something happened to the culture here that just lobotomized any ability to read vibes lmao

it’s funny because this is even baked into the product identities

Twitter at its best was a harmonious jam session of the mind. When it worked, it was jazz. To be sure, it was also among the most dramatic PvP platforms in internet history

Mastodon is much more people bleating a horn in your ear at max volume.

Going viral on Twitter inspired anxiety over context collapse.

Going viral here is the dread of “good lord am I going to be exposed to truly socially awkward shit for the next day”

@danilo personally I have had the opposite experience. Perhaps I was too late to Twitter to ever experience the jazz and instead got only the excruciating cacophony. But in both cases, there are definitely two types of people: repliers who regard the replies as *their* space, and those who regard it as *the OP’s space*. On mastodon if you encounter the posting HOA, they very much regard your posts as intruding on *their* experience; any reply they give is therefore fair game.

@danilo I do a lot of replying on here (I’m doing it now!), and sometimes I might be a *bad* guest, but I try to view myself *as* a guest in someone else’s replies, and I seem to have a lot of success with that. When someone very obviously sees replies to someone else’s post as their own space where the OP has already intruded, it leads to a lot of conflict and just generally incomprehensibly rude behavior.

@danilo Prior to social media, this was much clearer given the affordances of blogs. The comments on A’s blog are very clearly A’s space. If B is being an ass there, they get booted and nobody is confused about why. But on both Twitter and Mastodon, it’s annoyingly vague and there’s no social consensus. I think mastodon might benefit from giving this sort of control to posters, so that people can curate their own audience interactions without running their own patched single-user instance.

@glyph @danilo fundamentally we think the structural decision in the platforms that reinforces the current state of things is that everything is flat

like in terms of navigation features and data model, everything is basically just tweets/toots. replies to tweets and toots are themselves tweets and toots. all replies are of the same type (compare BobaBoard and Tumblr). thread navigation features try to make a little sense of it, but that requires extra attention to get anything from.

@glyph @danilo the result is that social norms which would be obvious if a stranger were walking into your home and shouting at you, or even interrupting you as you walked past them on a public sidewalk, are not immediately obvious to everyone and must be actively constructed.

@glyph @danilo this flattening isn't an accident; it had to be actively invented. older platforms didn't do it to the same degree. the likely core motivation for it is that it drives engagement. more than that, though, someone could even have invented the flattening thinking of it as a good thing for society; erasing distinctions is part and parcel with cyberneticism, which was an active driver in the structure of the early internet.

I want to start off by saying I agree completely – nay, ardently – with this observation that conversation on the Fediverse is largely pathological because there is no sense of people having spaces that are theirs. There's two important parts of this.

There's the "flatness", as you've put it, @irenes: the user experience here simply doesn't communicate that there are different spaces one is moving between. Or perhaps it is more accurate to say there *aren't* spaces.

🧵

@glyph @danilo

@irenes @glyph @danilo

In any event, users haven't got a prayer of having any sense that they are in somebody else's space – much less realized they should moderate their conduct accordingly – unless they come from some other online world, such as blogs, and carry over that paradigm.

Another way of putting this is that there are no visible *boundaries*. Good fences, the poet wrote, make good neighbors. Boundaries aren't just for keeping people in or out, they're lintels and doorways.

🧵

@irenes @glyph @danilo

Boundaries let us know when we're supposed to toggle our behavior from one mode into another. When you enter the library, you lower your voice. When you enter the house party, you go to greet the host. Visible boundaries let us know when we're free to treat a place like our own, and when we are to conduct ourselves as guests in somebody else's space.

We don't have those here.

🧵

@irenes @glyph @danilo

And the thing is, there are plenty of people out there in the face-to-face who even when they have all the signposts and fences in the world are still a little confused by the idea that different spaces, different contexts, actually have different norms and require different conduct.

So my theory as to how and why this happened here is a little different. I think it was something of an accident, but it was glommed onto by users who LIKE this.

🧵

@irenes @glyph @danilo

I mean, this problem here is directly inherited from Twitter. The people who built this thing built it to be like Twitter in this particular way.

I think that this flatness of Twitter and now of Mastodon is actually regarded as a feature by a certain population of users.

I think that out there are users who *like* the idea of not having to be a guest in other people's spaces to interact with them, who *like* the absence of boundaries.

🧵

Glyph

@siderea @irenes @danilo I want to be clear that the tone I am saying this with is reverence and enthusiastic agreement with everything you've said in this thread, but if I may "yes, and" a bit: there's a very fine line between the absence of boundaries (very bad) and the absence of arbitrary hierarchies of power (very good) and unfortunately a lot of design decisions that get you one also get you the other

@glyph @siderea @danilo absolutely agreed

so we've chewed a lot on this and we think that current platforms make access decisions in a kind of ongoing, forever-until-changed kind of way

which does not at all match social patterns that work well in physical reality. when you invite someone to a party you're hosting that is a one-time decision. you aren't making a binding ruling on who they can hang out with elsewhere or later, only on this one event.

@glyph @siderea @danilo we need to create online spaces whose metaphors facilitate these recurring, one-time opt-in things, rather than ongoing ones. make it an active decision who you spend time with.

hard block will always need to be there (for example, for the angry ex-partner threat model), but the goal should be to architect everything so that it's very rarely needed

@irenes @siderea @danilo 100%. Hard blocks are really the only *structural* option for moderating one's space here, and they are an all-or-nothing nuclear option. "Go home and think about what you've done, you're drunk" should not be the same signal as "all future communication should be through my lawyer, I never wish to hear from you again". People discover boundaries through minor sanctions, so if you have to take it from 0 to 100 in a single infraction nobody can ever learn anything.

@glyph @irenes @danilo

Ooh, if we're making shopping lists, I have some items to add.

Over on DW, which has probably the best boundaries model ever implemented (still missing some things I would like), I have the option of setting users to pre-moderation. I can set the default for my entire space to pre-moderation, I can set the replies to a single OP to pre-moderation, I can set certain groups of users to pre-moderation, & I can set individuals to pre-moderation.

@glyph @irenes @danilo

The thing about moderation tools is they're not just about moderating an individual's interaction with me, and this is an example of that.

My number one use of pre-moderation is to deal with commenters who are well meaning but hijack my threads.

I don't just do this because I don't want to interact with them. I do it bc I don't want them to turn my conversation to some other topic.

@siderea @glyph @danilo ooo, that sounds really neat

what's DW and where can we read more details?

@glyph @irenes @danilo

Like, there are people I know that if I ask a question "how do I do this in PHP" will start a flame war about PHP in the comments, thereby all but guaranteeing I will not get any answers to my question.

I can put those people on pre-moderation, and then they can make their comment about PHP being a fractal of badness – which it is! – and I can prevent it from consuming all the oxygen in my room.

@glyph @irenes @danilo

By these means I don't have to exclude these people from conversation. I just vet their comment first. And it means as they get the hang of doing this conversation thing, and I have to worry less and less about their conduct, I can take the pre-moderation off their account.

It allows them room to grow.

@glyph @irenes @danilo

But relatedly there's something else I want which I don't have over there.

One of the fundamental problems, oh, pretty much everywhere is a specific enduring confusion about boundaries and privacy: that THE way to have boundaries is to "make things private", i.e secret or invisible.

Like here on Mastodon (as on Twitter) you can restrict who follows you and then restrict the visibility of your toots to only your followers.

🧵

@siderea @glyph @danilo ah! yes! good angle! please continue

@glyph @irenes @danilo

Allow me to point out this confuses READ operations and WRITE operations!

If what one wants to do is regulate people commenting in one's space, one wants to regulate who can WRITE in one's space (and how they do it), not who can READ it.

What we really need are tools to allow Write control, as much as Read control. One shouldn't have to make one's content private to have some say in how people interact with you in your space.

🧵

@glyph @irenes @danilo

DW allows me to say "pre-moderate everyone I don't follow" (a W perm). DW allows me to say "only show this to logged in users on this arbitrary ACL list." (a R perm). DW DOESN'T allow me to say, "allow anyone to see this, but pre-moderate everyone except logged in users on this arbitrary ACL list." (a W perm).

So I can't set up a public conversation among known knowledgeable commenters.

@siderea @glyph @danilo makes sense, for sure

we choose to invest more time in those situations because gently guiding people is kinda an end in itself to us, and absolutely a thing we want to do almost as much as we want to talk about the original topic. we know this to be unusual and nobody should be forced into it if they don't choose it, it has far-reaching implications.

@siderea
> My number one use of pre-moderation is to deal with commenters who are well meaning but hijack my threads

You use Mute Conversation for this?

@glyph @irenes @danilo

@strypey No because it doesn't actually work for that. I was speaking of using DW's screening feature.
@glyph @irenes @danilo

@siderea
> it doesn't actually work for that

Of course it doesn't, sorry for the distraction.

I must stop posting when I'm tired. For me it's pretty much like posting drunk : P

@glyph @irenes @danilo

@glyph

Well I don't know that you and I will agree on this. I've noticed a lot of people who claim to be against hierarchy keep doing things which actually create the worst of them, and don't seem to notice. I would say that the very *best* hierarchies of power are the arbitrary ones. Especially if everyone gets their own.

@irenes @danilo

@glyph

Take Mastodon for instance. There's lots of supposed anti-hierarchy anarchists running around here claiming that the power relations on Mastodon are some sort of good thing.

They seem not to have noticed that they have reinvented the police.

@irenes @danilo

@glyph

I suspect a lot of people who claim to be anti-hierarchy are actually anti-other-people-having-defenses-against-them.

@irenes @danilo

@siderea @irenes @danilo There are some who I suspect are like this but those suspicions are based on those that I unfortunately know for an absolute fact are like this.

@siderea @glyph @irenes @danilo Maybe, but I think part of the problem is that hierarchy can be useful as a situational organizing principle (ie. a clear way to assign responsibility in a big undertaking at a given time), but it also can be toxic as a social status principle (ie. an assumption that some folks deserve to exercise power asymmetrically over others at all times). It's easy for folks following the first notion to slip into the second, and victims of the second to discount the first.

@JMarkOckerbloom @siderea @glyph @irenes @danilo as with so many things, one extreme *and its opposite* are both bad

in aviation we have this notion of crew resource management, and it includes the idea of an authority gradient (ie between captain and crew). too steep bad--junior people don't speak up when the captain is wrong in an emergency. too shallow bad--no one can make quick decisions. training people to be inclusive leaders+empowered followers=lifesaving.

@JMarkOckerbloom @siderea @glyph @irenes @danilo mind you as the stakes of the decisions shift and so do the speeds at which they need to be made, I think the range of best values for that authority gradient steepness changes too

@JMarkOckerbloom

Okay, so, I totally appreciate the thing you're trying to express here, and I think framing it in terms of "hierarchy" *can* work, but generally doesn't, and that's in part because the discourse that has coalesced around the word "hierarchy" was deliberately done so as to avoid using the word "power" for Bad Reasons, that are basically intellectual dishonesty & emotional evasiveness. So "hierarchy" tends to turn in the hands of anyone who wields it. :/

@glyph @irenes @danilo

@siderea @JMarkOckerbloom @glyph @danilo mm

see, to us, hierarchy is something more specific. it's not just the presence of power, it's the structure of it. if we lived in a world where everyone had the same power, there would still be power, just, it would not be hierarchical. we think that's a useful distinction, though we're happy to hear arguments to the contrary.

@siderea @JMarkOckerbloom @glyph @danilo the concept of power is present any time an effort to help people describes its goal as "empowerment". it doesn't go away, nor should it, we don't want to live in a world where nobody has any control over their own lives (or anything else).

@siderea @JMarkOckerbloom @glyph @danilo we're saying this without a lot of understanding of the history of the terminology here, so please feel free to give us background

@siderea @glyph @irenes @danilo IMO in order to be an effective opponent of the big harmful hierarchies one must understand deeply how hierarchies in general work, how they're strong and how they're weak, what their ripple effects look like, how to exist within them deliberately. as with "being political", refusal to engage with the concept is a paradox, you still end up political / in a hierarchy, but lacking important (self-)consciousness.

@siderea @irenes @danilo There might be some minor disagreement around the margins but (A) I've read "the tyrrany of structurelessness", so, yeah, and (B) clearly the current set of affordances are badly in need of reform in the direction of more boundaries & hierarchies. The confusion we are discussing is a clear signal that the user-affordance pendulum has swung *way* too far away from clear ownership and local moderation of online social spaces

@glyph Ahaha, I was just going to mention "Tyranny of Structurelessness".👍
@irenes @danilo

@siderea @glyph @danilo for every good idea that becomes popular, there's a zillion fake versions of that idea, pushed by people who stand to gain from confusion. it sucks.