Don't expect to make an enormous amount of sense out of this.
Back. I didn't stop writing, I just stopped posting it. I don't think most of it is postable though.
(Not without reflexive shame.)
Common problem in healthcare: you need X data, you don't know who has it, you have to send HTTP requests to literally 100,000 organizations in order to find out.
Kind of wondering why we don't syndicate the requests on a feed (each with a return address in the form of a webhook, and an expiry time), then say "Hey, anyone who answered in 5 minutes, _that's_ who has the data"
This doesn't save on sending 100,000 instances of "please send me the data," but it does save on receiving 100,000 instances of "no." It maybe saves a little bit because replicating a log is a very solved problem.
The log of requests doesn't need to be kept eternally -- you are assumed to only be listening for a set amount of time, and your request can go away after that time has elapsed.
This design has downsides --
For one, it can't distinguish a hospital outage from a missing patient. This prevents optimizations like "only fetching data from care organizations that previously reported knowledge of the patient." However, note that this optimization is already impossible -- hospitals have to assume you could have been hospitalized in a new place at any time.
For two, every care organization knows what requests you made to other care organizations. But for bulk requests, this was already true.
For three, this data structure cosmetically resembles a blockchain. It's a much older data structure but that's what people will think of.
-- and of course there are probably other downsides I haven't thought of.
Anyway, that's my idea! For all I know someone has built this already.
PS: I made a YouTube video! I'm not totally pleased with it -- it was pretty spur of the moment -- but you might enjoy it.
Benny Blanco compared his chemistry with Selena Gomez to "heroin and xanax combined."
If they ever have sex, they both die instantly. Fortunately, neither of those drugs will let you hold an erection.
There's a TikTok creator I remember who described herself as schizophrenic, and said that what that means is that she sees things that aren't there.
She described an experience of going back to her old house and seeing her deceased dad on the lawn, where they played football a while -- and she really did go there, she says, but of course later she found out he wasn't real and the football wasn't real either. It had all been an illusion.
I guess my thought is that if someone has the level of schizophrenia that presents with immersive visual hallucinations that present to the person as real, I would not expect them to be able to explain that so clearly, or with such a neurotypical affect.
Like, I had kind of thought the most universal symptoms were disorganization and like, a general incapacity to do things volitionally.
(I'm kind of using Elyn Saks' description of her own history of schizophrenia as a mental template here -- she had fairly severe delusions, middling hallucinations, and very disorganized speech production mixed with uh, just huge piles of ego-dystonic behavior and compulsive negative thoughts about it.)
I don't know if she actually is schizophrenic, but I don't think this is how schizophrenia works.
I think some people are into the idea of having a disability -- and some of those people actually have one.
I think that even if you're an X, you can behave in ways that advance prejudice against X people.
I think that if you're autistic and you know allistic people are looking at you, you often have pretty strong social incentives to make other autistic people look bad.
I think that the processes that fetishize people often look really close to the processes that demonize and build prejudice against them -- because usually fetishization is very othering.
But I guess I also think people experiencing a certain alienation from their own body is a thing they get horny about --
It's complicated -- I want to clarify that this is one of those cases where I say a thing is "complicated" and don't respond by attempting to force myself jetpack-style into a distanced state of aporia.
Because that's like every ethics longpost on the internet -- people try to convince you the issue is complicated and then they perform that there's no resolution and then --
It makes me ask myself "why the fuck did you start typing?"
So I'm going to describe the way that I personally litigate this -- in highly specific terms -- and I'm doing this as an explicit recommendation that you do the same.
And I recommend that you only take my recommendation if you are capable of feeling the same moral reasons that I feel --
So I think there's a few causes of action for "this person is stirring up prejudice." There's (1) this person is promoting stereotypes about the thing; (2) this person is promoting the significance of a binary about the thing; (3) this person is fetishizing the thing.
I am not saying these are the only ways to be bad -- these are just three common ones.
My observation is that for the three specific things I just picked out, there's really strong reasons to participate. That's true for people in both the marginalizing group and the marginalized group.
The incentives are not the same, but they align --
They align way more when you're in a setting where you are turning yourself into pleasing packaged content to be consumed, like Twitter --
Can I be concrete? I think that for some people it feels good to be a token, it feels good to be a sex object --
These are entrypoints into kink -- you get stuff like raceplay --
Stuff I'm totally fine with it, because when this stuff is framed as kink it's easy to distinguish it from a real-life attitude. Communication is usally multilayered enough that I can tell when someone is promoting their kink concept basically as fiction.
Even if like me, they're a person for whom framing their activity as a droppable mask is painful, so you'll never see an overt nod --
(Please be assured that if you're doing your thing as a sex thing I'm probably picking up on it and not judging you. As is my moral responsibility as an adult in public.)
The actual thing I'm complaining about is this in a context where the presentation of the prejudice content is about personal empowerment, education --
I mean, like, PSA-style posts that start "HERE'S WHAT YOU NEED TO KNOW ABOUT AUTISTIC PEOPLE."
Stuff that is characterized by the deployment of an "it is morally wrong to ignore this" device --
Like, didn't we have that whole era of trans empowerment content that was just overtly about the inferiority of men? That's Cause of Action 2.
But like, OK, we also have the second order of this content, the people who are affected by it and repeat it --
If someone with very neurotypical presentation said to me "I think I might be autistic" and presented me a list of stereotypical autistic traits, I would think they were trying to come off as autistic, but also that they might actually be autistic.
My assessment of their ability to do harm to autistic people would not be affected by their personal self-identification, because -- well, for one, they are clearly making that self-identification from a position of uncertainty. They are identifying themselves from a context where their incentives are statedly the exact same regardless of whether they're Actually Autistic.
They may even be in doubt about the metaphysical existence of an Actual Autism to be connected to.
They're -- when it comes to me, at least -- not pursuing anything intrinsically threatening to me. Because the position of being "friends with Nyeogmi, an autistic person" is not rival --
Oh wait, you don't know the word "rival?" Or maybe you do and you think it's a little rival --
Backing up --
OK, like, the idea of being a Token Autistic Person? There's only one of that. I'm just saying that like, if this same person were auditioning for Token Autistic Person on my social media feed, I would recognize that their ability to do this directly threatens to silence other people who are worse at jockeying into the position of Token Autistic Person. That maybe my ideal Token Autistic Person is someone who experiences significant impairment, not someone who can readily explain it in a benign-sounding way to non-autistic people --
I am saying that when this person comes specifically to me none of that stuff actually plays heavily into my thought process because there's really no additional harm that can be done by them if I simply allow them to befriend me.
(Even that is an oversimplification because like, I cannot simply allow a person to befriend me -- we have to hit it off or no friendship occurs. What I really mean is that I can let them try.)
This is literally the specific run of thoughts that goes through my head when someone claims to be in a marginalized category and then self-describes in ways that do not match my image of that category.
There's also a lot of sub-verbalized thoughts which I would say amount to "I am trying to figure out why this person is making this post, and if there's some defensible good faith reason that they would do it."
I think it can kind of be tl;dred as these steps.
- Does one of Nyeo's Prejudice Causes of Action appear in the post?
- Can I plausibly excuse this as kink activity?
- What do I think the person's reason for doing this is?
- Are they jockeying for a rival social position? (That is, if they "win" what they want, does some other marginalized person lose?)
- Is this a big enough deal to even give a shit about?
I had a really strong callout tendency for a long time -- right now I mostly don't, or maybe it hasn't been tested.
I still think about this a lot -- I suppose my first reaction to a person is often deep distaste that really percolates for a while and then my brain produces a verdict, but it's not like I feel good at any stage during the process leading to the production to the verdict or after the verdict actually occurs.
I guess the one other piece of advice I would offer -- and it's not actually epistemological advice this time -- is that like, if you ever find yourself taking a significant amount of satisfaction in this kind of thing --
Well, I don't know if you should stop. Sometimes you might be onto a guy. But once you're doing a thing for two reasons, you have to accept that you may not always know what the reason you're doing it is.
If you are seeking vindication, I kind of recommend finding a way to engage with the task that makes you feel less vindicated -- and I kind of recommend finding a second party that has less of a vindication motive.
If you're like me there's something deeply libidinous about calling people out for being shit, and that's OK, but it does make you a certain kind of predator and someone should keep you from nibbling on people you shouldn't.
And yes, that's a thing among social justice people -- particularly the people who are the most loudly insistent about their own goodness.
Oh, by the way, when I made posts on Twitter where I was euphoric about doing X to You --
If X was "drinking your blood" or "controlling your mind" I was probably sincere.
Otherwise, I was probably exaggerating for attention. Sorry about that.
While I can't say I support the NRA, I increasingly adore their mascot, Eddie Eagle.
He exists to teach kids gun safety so he's like, a bird whose hobby is not owning a gun. And if he does own one, not touching it.
He could also be the mascot for abstinence education.
Reddit thinks Phil Hellmuth, 17x bracelet winner at the World Series of Poker, is total trash.
The account is something like this: "he plays like garbage in cash games with tiny stakes. He has no idea what he's doing."
The pattern they've pegged him to is that he'll show up at some random event -- anything, especially if it's on stream -- buy in for the smallest possible amount, lose it stupidly, and then throw a tantrum.
Here I am thinking -- "this man has calculated how much he needs to spend on playing like garbage in amateur games in order to convince people that he really is that bad."
I'm starting with one point seven five pounds of flank steak. I don't mind my beef chewy -- I cut it very thin. I'm looking for flavor.
Two virtues of a very good dry rub: being tasty, and coming in a very large container. I use 365 Daily Value's Black Pepper and Garlic, but if I were making my own I would probably do it very simply with cumin and paprika. Those are available cheap and very tasty at my local safeway.
I coat both sides. Then I lay my steak on a cookie tin which I've covered in parchment paper. I tend to reuse this tin for several cooks without cleaning it, which might be unsanitary but it saves me a lot of time.
I don't know what temperature to make the oven. Sometimes I'll do a 90 minute cook at around 180 degrees F and other times I'll do a 30-60 minute cook at 300. My observation is that the higher temperature makes the toughness a little worse, but it saves time.
I flip it into a pan with some sunflower oil after and fry it until both sides are between red, brown, and black.
Steak is a dish where there are very wide tolerances. It's very easy to cook.
Today I watched something I didn't really understand about a guy who, apparently, learned Japanese with a whole lot of effort and then started going around claiming that he learned it entirely from watching anime.
Nobody in this argument seems like a reliable narrator -- this is all YouTube drama stuff.
Anyway, what I did get is that he's selling courses. Because I saw his fans in the comments.
And they're saying stuff like "studying Japanese will not get you anywhere. We tried that. It doesn't work."
And that seems like a load of shit, right? Because this guy is selling them a course that they obviously think is fine. And how is anime going to teach you what mistakes you're making?
Oh, and I got into an argument with myself (no one really disagreed with me) about the utility of slurs --
I am a person who deliberately uses a lot of language that is historically used to stigmatize people similar to me -- I have called neurotypical people "crazy" really often.
It occurred to me that I don't have much of a background having slurs used against me.
I'm a non-practicing Jew. Something I've noticed is that Jews seem to like nounifying Jews and antisemites seem to like nounifying Jews, but I feel like people who are neither will very often write "Jewish."
In general, being called a Jew doesn't distress me.
I have had the experience of having very strong internalized self-negative attitudes, which is something a lot of people associate with slur use. But as a kid the slurs for my problems hadn't been invented yet for AMAB people.
We were talking about the GIMP, which is -- you know, this big art program lots of people use. And of course it has a name that's been interpreted by a lot of people as a slur against disabled people.
And what I've noticed, after being in endless arguments where people defend the name, is that it's way less common for people to say "Well, I'm not offended" -- in its defense, that is. It's more common for them to say "I don't know why you're offended."
This is self-evidently absurd -- they're being told.
And I remember the 2010s where I saw all these endless idiotic arguments started by white people with the basic premise that, like, there must be some socially acceptable context where a white person can use the N-word.
And they'd gleefully use the N-word in their description of what they thought that context should be.
So -- I think that slurs are indeed offensive, and some of that comes from their intent to offend. But I don't think this is the only reason people use them.
If you treat this use of language as a symbolic gesture then a lot of the discourse falls out.
Some people don't get it, and other people pretend not to get it.
There's a general agreement among white antiracists that other white people should not use the N-word.
So -- if I want to prove to you that the other white people around you are not antiracists, I can use the word.
If I'm wrong, I get kicked out, banned, or fired. If I'm right, I can expect their lack of reaction to that to make you feel threatened and maybe also betrayed.
And they know I'm being honest, because they know I know that the stakes are that they would kick me out if they're not racist.
So you complain -- and everyone acts irrational. "I don't get it." "You mean I can't say [****] around you?" (Only they use the word.) We are dogwhistling at you -- we are doing this deniable gesture of pretending not to understand because they want to underscore to you that you don't have the social power to make them stop.
We have created a coalition of people who are signaling to you a willingness to be racist without doing anything that makes it completely undeniable that we are racists.
I think there's a huge amplifier on the effect of using slurs like this.
A lot of people who post the Nazi bar thing (that stale old thing) will act as if the primary effect of the Nazis is that they drive the non-Nazis out.
Everything I just wrote assumed that "whether someone is a racist" is a static thing, which is kind of the general version of that assumption.
Actually, an important part of their effect is that they move the center. When people are behaving as a coalition, other people join the coalition.
And I think people's willingness to be racist (or transphobic, or so on...) is kind of a sidelong thing of other feelings they have. There's a fear of drama and a fear of social exclusion --
Like -- have you ever talked with someone who's a trans ally but not serious about it? They'll be great for a while and they'll say the right things. Someone will then say the opposite --
You'll realize that they were more motivated by the idea of harmony with you. That is, they thought the expected value of disagreeing with you was negative, so they agreed with you.
I want you to consider that for basically any group activity, the social EV of that activity can be manipulated by a highly motivated coalition.
I also want you to consider that the social EV calculations can be very, very spiky for people who are the target of the racist comments. Or, well, for an example closeto home, people who believe that they're the only queer person in the room -- because they have not sent a costly social signal or built a coalition -- they don't usually stand up for themselves.
To be clear, I think that's also a problem for Black entertainers -- but whenever I litigate this at length (and beyond mere mention) I'm going to stick to communities I'm in.
I also want you to think of this like a hypnotist and realize that in a big room where multiple people laugh at a racist joke, where you're not the butt of the joke, you're likely to believe the joke is funny.
I'm not saying "you will laugh just to be polite." I am saying that by tacitly accepting there can be objective facts about things which are only measurable in the form of social outcomes, you're turning the literal will of the group into a certain kind of epistemological decider in your mind.
What do you think about the people you know who are trying the hardest to be funny?
For that matter, what do you think about the people who are trying the hardest to be liked?
Coffeehouse Crime's new video is about a therian schizophrenic man who decided, with no real prior warning, to eat someone else's face. The person was alive at the time but he had a knife, so that was a tricky situation for both of them if you think about it.
I don't think I'm doing the story justice, or that there's really any way to do it justice. It's a pretty inexplicable series of events.
I really love it that every time he makes a video about someone who happened to have a YouTube channel, he includes a segment about how their videos were bad.
I have to admit that if bad videos were a sign of insanity, most of YouTube would have to be committed.
I think that most board games are based on a permission-granting structure.
This is often a flowchart, but usually, it's like four flowcharts. For simplicitly, I'll just call it "the flowchart."
There's facts about the game that are related to the flowchart -- whether I moved my pawn or my knight, for instance -- and there's facts that are related to the way that flowchart state is realized. Like if my king is standing in the middle of the square or on the edge of it.
I call the state of the game according to the flowchart the "structural" state of the game, and the remaining state the "physical state."
(Note that in the flowchart of chess, the exact position of my king inside the square (its physical state) has no effect on what moves the king can make. (its structural state) Chess kind of constructively makes that impossible, because it has no physical-physical transitions -- every action in chess is a structural-structural transition that the player then realizes as a physical move. The physical move is only regarded as valid if it is a correct interpretation of the structural idea.)
In video games, the design is similar, but the physical realization of the flowchart often takes over. There are a lot of physical-physical transitions in a first person shooter -- moving an inch in any direction cannot be approximated by moving the player to a canonical position, calculating the goal as a new canonical position, then snapping the player to that. You can't quantize Doom.
I think that when we describe a "buggy" game, we're often describing a game with unintentional branching caused by physical-physical transitions.
We can look at the game and see how it was flowcharted. Imagine some kind of buggy Roblox dungeon. I'm supposed to move from the key room to the dungeon to the throne room -- and then we start to notice -- if I'm looking at the torch just right in the key room, I can skip directly to the throne room because I clipped through the wall.
In a game like that, effectively, a system that is meant to be understood as "pure physics" has grafted itself onto my flowchart and added branches.
A lot of games that are good and which make sense with their intended flowcharts are suddenly bad because of extra branches that weren't intended to be possible.
I think a thing that I like about the phrase "permission-granting structure" is that it captures some information that is missing from the word "flowchart."
A flowchart tells you what states move to what state -- it does not tell you what flowchart to use or which decision to make at a choice point.
There's a third entity (between the player and the flowchart) which turns the player's creative input into choice point decisions and performs actions for agents other than the player.
This third entity looks at a koopa that can fire a shell and make the koopa fire one. Note that "the koopa can fire a shell" is a structural fact -- "the koopa does fire a shell" is underdetermined by the structural rules of the game, so by my formulation, it is a physical fact.
... Of course this second entity may be driven by flowcharts of its own. We are simply observing that following nondeterministic rules requires some external system that handles the things that are underdetermined.
We could argue that a game is a tower of systems that each underdetermine the realized state -- "Slay the Spire" says that there's a dungeon and here's what your cards do, "Menuing" says here's how your cards move when you put the mouse over them, "AI" says what the agents do --
This implies that my concept of structural/physical state leaves a lot up to interpretation.
I would say that, typically, the most useful way to look at a system will involve strategically reducing the amount of branching.
A system is "resilient" if there's kind of a straightforward, continuing (or even looping) happy path down its flowchart, and all branches guarantee a return to this happy path.
For instance, navigation in a game is "resilient" if a player's accidental fall into a pit results in the game transitioning to a state where the player is in a location that is part of the level's planned flowchart.
On the other hand, a system that can completely foul its own state is not "resilient" -- soft-locks are an example of a resilience failure, as are cases where the player gets stuck in walls.
Resilience is kind of about accounting for everything with your flowchart, and making sure that (1) whatever's supposed to happen in a straight line still happens in a straight line (2) whatever's supposed to form a loop still forms a loop.
This is usually something designers intentionally build into a game. It's a very algebraic kind of thing -- the exact states don't matter, the relationships matter a lot, and we try to set those up on the flowchart level.
For one thing, we try to atomize events. Suppose the player can fall into a pit but also be mauled by bears. To avoid having a flowchart node labeled "the player is in the pit and also being mauled by bears" we have an extremely limited set of relationships between pit-states and bear-states. We probably even represent pit-related hazards and bear-related hazards on totally separate flowcharts.
For another thing -- we try to create closed sets of states.
If the states of your system form a closed set (which is a nice way to say that any move from one valid state results in another valid state) you have nothing to fear. For instance, if your flowchart kind of requires every state to be reachable from any state the player happens to be in, you might build your game with mostly reversible movement.
If the only thing you care about is the end-state being reachable, then you flowchart your game with one specific "impassable wall" style boundary and then the only thing you care about is making sure that you can pathfind backwards from the end-goal to every possible starting point without crossing one.
I have spent the last few days complaining about physics sandbox-type engines like Unity and -- well, I mostly still do not like them.
The gist of why I continue not to like them is that, in my experience, enacting a flowchart on a game-world usually requires a permission granting structure where specific agents have a very large amount of power over other agents. That is only the default way of working in some engines. It is rarely the default way of working in Unity-category engines.
If Mario stomps on a goomba, in any hand-coded game where Mario and goombas are a thing, the default way this happens is that something in Mario's code detects the goomba and from there Mario's flowchart has basically unilateral goomba control.
This is highly simplifying for a game designer because Mario's flowchart isn't about goomba, it's about Mario. Mario's goomba stomp never has any effect on Mario other than that which is specifically flowcharted in a Mario document.
A designer can ensure that Mario never gets enough momentum from the goomba to execute a skip. Or the designer can specifically allow for the idea that extra momentum exists in this situation. (Look at the widely exploited, clearly-intentional "midair Yoshi jump" trick from Super Mario World, for an example.)
When you're confronted with an existing physics engine and asked to bludgeon it into shape, you're given a dungeon master whose initial priorities are, basically, not about your character.
If Mario is a cube and a goomba is a cube and they are governed by the rules of interacting cubes, then it's possible and even likely that the engine will prioritize goomba-specific information in resolving Mario. A big and heavy goomba may be disruptive to Mario's course of action and it may have the accidental function of behaving as a platform in a puzzle where this was not intended.
I see this primarily as a way that the physics state of a game leaks into the structural state. I'm not saying that that can never be allowed to happen --
Specifically, I think that simulators (which I've written about) can be characterized as games where objects start out in a known structural state, get turned into physics for a while, and then a herding system converts them from physics back into structural information.
Take Roller Coaster Tycoon -- guests spill like agitated water into your park layout, ride a random selection of rides, and produce a randomized amount of money that will still tend to fit a distribution that is affected by your park.
No flowchart could possibly encapsulate the aggregate behavior of hundreds of guests, although each guest certainly has their own flowchart.
A flowchart could accurately capture the game's main loop and the progression structure that the game uses to offer the player short and long-term rewards.
This kind of interaction is popularly called "emergent gameplay" and many devs see it as a desirable property. I sort of feel it -- but like, for level design, it's maddening, right?
At this point it's not even that I'm strictly opposed to it. I just think that relying really hard on physics systems that involuntarily change the structural features of your game means that you abdicate a lot of control of your game. You kind of end up making art that is about unmasking what Unity is, instead of maybe making art that is about the expression of your vision.
Mild, boring, technical addendum --
A certain kind of annoying person will likely respond to all this by pointing out that programming languages themselves, by my description, are structural systems where the realization of any state of any particular program is physical.
What I'll say is that programming languages happen to be permission-granting structures and they kind of disappear, conceptually, from art once they're used to implement a much more specific permission-granting structure.
That is, the fact that I can take Java and implement my flowchart exactly on top of it without any Java ever leaking through is a testament to the fact that Java is a very good tool for building resilient systems.
I'll point out that this is not true to all programming languages. While I was writing this, an acquaintance pointed out that a LuaJIT bug had broken her game's physics for years.
I think it would be safe to say that -- while her program is clearly physics from the point of view of the structural system that is LuaJIT -- from the point of view of her program, these odd behaviors of LuaJIT look more like physics. Why? Because they are underdetermined by her flowchart.
Hotline Miami is as old now as Majora's Mask was when Hotline Miami came out. I thought I'd take a moment to bring you back with me to that era.
Since then, game design has advanced and homogenized a bit. Storylets and deckbuilding wouldn't become mainstream for years after this, but now almost every game has them. The (weirdly large) presence of sardonic British narrators has decreased.
Conducts and ascension systems have also been totally normalized -- but they've been retooled as mobile-style progression tracks which totally escaped into the single-player world as a tool to add playtime to small-budget indie titles.
I like to think Hades (2019) and Slay the Spire (also 2019) were the games that made... all of those breakthroughs, actually. This is an oversimplification as a lot of things were happening in different places at the same time.
While I was making this list a friend asked me if I liked Dark Souls and I had to admit I didn't get far in it. I got to the big church and I remember the big bridge with the dragon on it.
That's not the only game from this list that I didn't play. (I didn't play Braid, The Walking Dead, or Gone Home either. I utterly bounced off The Last of Us.)
With Dark Souls -- well, let me divert and talk about branching, actually.
This is my understanding -- if you plot a game like it's a flowchart, you're going to have a lot of places on that flowchart where multiple things can happen. The player either falls into the pit or they don't.
There are also places on the flowchart where multiple things could have happened. The player was in the pit -- now they are not. In a sense it no longer matters if the player fell into the pit.
There is an unspoken general rule that games should be played at close to the maximum difficulty that you-the-player can tolerate. Not everyone thinks this, but a lot of the people who do think this will just assume it. And what I will observe is that following this rule means you are going to observe a lot of branching.
The thing that impressed me about Dark Souls is that the level design rewarded me for taking branches.
For some reason RPGs rarely account for this, so I have to choose between "piss easy" and "game feels like it's stuttering in response to my unplanned mistakes."
It's not perfect -- bosses are bad -- but outside that, someone at FromSoft knew what they were making and came up with a design that would sort of work.
Now let me divert again from Dark Souls to talk about how --
I think that a lot of games, even games that aren't simulators, have a kind of simulationist design.
Imagine Wizard Bastard -- a game where you start with five hit points and five mana points -- and getting punched reduces your HP by one, and casting a spell reduces your MP by one.
And when your HP is zero, you die.
So all the states with 5HP are related to all the states with 4HP, and the same is true of all the states where you have 5MP.
And every gameplay action you take kind of pushes you towards those states where you have 0HP or 0MP. Because, like, a game lasts a maximum of 10 moves until you die. The flowchart for this game has a lot of join points -- because I can get to 4HP+4MP either by casting a spell from 4+5, or from being punched at 5+4.
So that structure in math is called a "lattice" -- and it's really common. And lattices are really good for representing systems that are super independent until they're not.
Wizard Bastard is super, super abstract -- because you can do a lot of other things in a fight aside from casting spells and getting punched, like, for instance, winning the fights. It's not a true simulation, but it lets you play in the kind of space created when systems have the illusion of being independent.
Why do I say "the illusion?" Because none of the states where you have 0HP are actually distinguishable from each other -- and because in any state with 0MP, you stop having choices with the direct effect of being forced into interaction with the other system. There's a point where your decisions stop branching and start being scored.
That is, I think, the place where simulations usually exist. You're allowed to feel like you can make free decisions -- before eventually finding that these decisions are accounted for. I think this distinguishes them from games that encourage purely active play by imposing no resistance; it also distinguishes them from games that encourage purely reactive play by imposing total resistance.
Anyway -- I don't think Dark Souls is a good simulation of anything but itself, but it gave me a lot of space to play around in in the brief time I got to enjoy it.
I can use this space to re-recommend brilliant games that you genuinely should check out, like Deus Ex 1.
By the way, you might wonder what a game looks like that offers independent, simulationist-feeling systems but isn't a lattice. A lot of pure sandbox games keep the immersive sim mechanics but throw out the resource management -- that gets you stuff like Breath of the Wild.
You also have a subcategory of games that relish in presenting you some kind of abstracted version of a multiple-independent-systems category sim. Civilization 6 is, I think, a really good example.
Civilization 6 is a super object-oriented game. It has, like a lot of simulators, a bunch of systems where the player is tasked with making a number go up and the player is globally rewarded if the number is high and globally punished if the number is low.
I think Civilization 6 is interesting because it basically does not push the player towards a merging of its systems -- they are dependent in the sense that they are mutually reinforcing, and in the sense that sustainedly engaging with a single system increases the opportunity cost of doing it again.
But Civilization 6 lets you push very, very hard into the direction of its various expressive possibilities -- it does not really force the player into the "final exam" scenario where the numbers associated with every system suddenly have to be converted into a single currency if the player wants to escape the game alive.
I think some people would point out that the game does have symbolic currencies associated with victory but like, my observation is that there are about seven of them.
A game I love but didn't find an excuse to talk about in this post is Quake.
Quake is really bad to play but it revels in a lot of stuff that only FPS games can do.
There's a hidden axis you can judge games on which is, basically, how consistently the player is regarded as an object rather than as a subject.
A subject player has unilateral control of the world and is generally the source of permission for other actions to occur. An object player is heavily affected by physics and is playing reactively towards other objects. Quake tilts very far towards "subject."
Compare another game I like -- Saints Row 3. In Saints Row 3, you're in a camera that's consistently projected out somewhere behind your own head. A lot of player actions are associated with a randomly chosen animation that you're meant to find delightful, and your player character has quips that are meant to amuse you.
There's this repeated gimmick used where the camera will blink out, the player character will reappear in some new situation, and you'll get dialogue to the effect of asking (with some fury and confusion) "Why the fuck am I doing this?"
The joke is that you the player are not living in you the player character's head.
There's games that stretch your player identification in different ways. When you're playing Super Mario Brothers 3, you can see things that Mario would not be able to because they're obstructed by a wall. You are not encouraged to identify Mario's pain as your own pain.
For a third example -- Dwarf Fortress is a game with a suite of mechanics specifically for modeling bodies. While in Adventure Mode, your camera exhibits FOV designed to model the limitations of your character's actual view, your exposure to your character is identical to your exposure to every NPC.
(That's not totally fair -- there's some subjectivity when your player character's emotional state is reported to you between actions)
Back to Quake --
It's a first person shooter game, so you don't literally see yourself.
It's a game where you never see or hear your voice and where (in the single-player) it feels as if you're playing by an entirely different set of physical laws from the environment.
It's a game where your viewport is much more important than the overall shape of the levels and its objects -- with a lot of pinhole-sized windows, as well as gameplay-relevant objects that you can barely see.
Oh wait, let me talk about that!!
First person shooter games have to engage with this fascinating tension between the laws of perspective (which say that size diminishes with distance) and the laws of narrative (which say that the most important object should swell to fill your attention)
The usual way that level designers deal with this is by accounting for it some way -- giving vistas that preview the most important objects -- but in 1996, that hadn't been invented yet.
(There's another subjectivity that I only see a handful of games engage with. The square cube law doesn't make intuitive sense to most people -- a six-meter cube looks twice the size of a three-meter cube, shows four times as many faces, and has eight times the volume. Minecraft can be interpreted as a slot machine whose near miss mechanic is based on bad human intuitions about size.)
Sorry, tangent about first person shooter subjectivity over.
Anyway -- just know this -- I mean this!!! -- I think that games which force the player to engage as subjects -- basically as souls that are too big and too disorderly for the simulated world -- are fascinating and desirable.
(Oh, before I get off of Quake -- the levels designed specifically by American McGee are really good in totally conventional ways. John Romero's levels are the perceptually trippiest ones but they suck to play.)
ACK!!! OK -- uh, stealth games!!
That strategic separator will separate me from my disorganized ranting.
I do not like stealth games as much in practice as I do in theory.
They're kind of a central example, for me, of a genre that should produce flow. I kind of want to feel like a weapon, not a person.
For whatever reason, this is not a central design goal of stealth games -- usually they're designed in such a way that the world stays fucked up and possibly even impassible after the first time you screw up, so you never really get a chance to learn.
I do not get it.
In the cases where I've enjoyed stealth, it's often been a secondary commitment in a game that isn't centrally about it --
World of Warcraft's aggro mechanics are an abstracted stealth system based on site alarms (getting one guy mad gets his mob-pack or his entire neighborhood mad at you) -- resets are rendered as a transition to a specific world state.
Skyrim has a quicksave system that makes the world highly resettable -- I think that's huge and I think it's good, even though people make fun of it --
I think we have to reframe the goals of a stealth game in a more general way. It's a game where the player's own objecthood is the enemy --
That is, you're on the happy path as long as you see people and they don't see you.
You're on the sad path when someone decides to fire a rifle at you and subject you to the laws of physics.
You go very quickly from feeling like a soul flitting around outside a body (the player's actual relation to game-world characters) to feeling tragically embodied. Because -- don't forget -- in game worlds, Cartesian dualism is true.
After that, well -- the game resets --
Because being embodied is the enemy. It's like, extremely dysphoric, and then you die.
(I think my early instances of gender expression involved murdering people in Vampire the Masquerade: Bloodlines.)
Anyway, someone should make an extremely good furry FPS. I might give it a go, or I'll just hypnotize Bhijn and make them do it.
Today's music was Basement Jaxx, then Fugees' The Score.
Basement Jaxx is the kind of music they would listen to in The Bad Place. Less facetiously, it's really cool and hooky, and a lot of the songs verge on being bitonal.
Specifically, their melodies are often written in a key about a fifth above the key that the bassline is in. You could also just analyze this as saying that they think a major 9th is a fairly consonant interval.
As for The Score -- well, I'm pleased to hear a rap album with dynamics.
I don't like Godot or Unity. I have not used them enough to have full basis for an opinion, but I'm an experienced programmer and I have a nose for bullshit.
My current opinion is that they totally de-emphasize things like systems design and narrative, which are usually where the soul of a game is.
All the primitive tools GameMaker gave you to get there -- not good tools -- are missing or badly and artificially crippled.
What are you given? A shitty physics sandbox that doesn't work right even after aggressive tuning. An environment that totally resists designing an actual game because you can't build resilient systems on top of it.
I think Unity's shitty design created the asset flip problem and its success started a massive race to the bottom.
Today I spoke to someone about designing games and ranted unsolicitedly in their direction at great length.
This is a version of that rant --
Suppose you're new to game design. Do you start by coding or do you do some writing?
My recommendations change a lot based on whether you've done zero projects or like, maybe two projects.
If you've done zero projects, the first thing I would personally recommend is getting acquainted with the tools. You're likely going to take a lot of cues from those tools because they were designed to privilege certain kinds of games over others.
(For instance -- good luck making something in Godot that's not somehow connected to their shonky physics sandbox. But I digress!)
I have done significantly more than two projects. I personally do design without tools in front of me because I don't want my tools to decide in an initial sense what game I will make.
I learned game design in GameMaker, which is an engine I still have a high opinion of but I haven't used it lately. I'll opine on it sometime soon in a separate post.
I personally start game projects by making a design doc.
It's not formalistic and not for others' eyes -- it's just a tool for me to manage chaos.
I focus on the general idea first -- maybe a paragraph or two on that. Then I write a description of the game that has the specificity of a loose flowchart, all in words --
If what I'm writing implies a system, I try to skeletonize the system. ("The player fights the witch?" OK, I'd better write down "2D combat, button mashing" or perhaps "Like Slay the Spire, turn-based where you can see their action.") Usually this is a tagline and one or two bullet points.
At this point, I usually start to recognize what the problem areas of my design idea are.
This possibly makes zero sense to you -- if you're new, you might not have an intuition for what will create chaos in your life, or what you will regret not designing.
Basically, this is like a sketch -- you want to get an idea of what the whole thing looks like, and then you want to really zero in on problem areas, where one thing you're doing plays badly with some other thing.
(If you're _really_ new, ignore everything I said and just try to get some practice in!)
Something I think a lot about is the local structure and the progression structure of my game --
In chess, you move a piece and I move a piece, and we continue until neither of us is allowed to move -- then we score the game.
That's the local structure.
In a chess tournament, I earn a slate of wins, one at a time, and then receive a prize based on my overall score.
That's my progression structure. It has no connection to the local structure -- a checkers tournament would be the same -- except in specific ways the local structure intentionally leaks out.
These loops occur at different rates. In three chess tournaments I play twenty games of chess and maybe move a thousand pieces -- you can sort of think of these nested schedules as the hours, minutes, and seconds of the game of being a chess player.
These loops are often interlocking and unsynchronized -- while I'm playing the game of being a chess player, I'm also playing the game of being a vampire, meaning I'm eating, sleeping, and draining people of blood according to a schedule that follows the sun more than it follows the particular structures of the tournaments I run.
I personally try to be very intentional about my progression structure.
If I want the player to feel a sense of mastery, I might give the player a super hard boss monster every 10 fights, followed by four easy fights with monsters that used to be hard.
If I want to explain to the player what the meaning of their actions was, I might give them some expository text or a cutscene after the tenth fight.
If I'm trying to create compulsion in the player, I might do what World of Warcraft does and give the player about six progression meters that fill at different rates as I kill monsters. That way, the promise of an intermittent reward always gives them a reason to keep playing.
The practical use of this concept as a designer is that you can use it to make more compelling games.
Against the Storm is a very good example. They took the palette of actions from Frostpunk -- specifically, its two short gameplay loops of placing buildings and watching the seasonal cycle -- but grafted it onto a progression system resembling the one in Hades.
(It's the thing where when you interact with an object that you're given access to only every so often, you get to pick between three cards that each grant a different progression perk.)
To look a little more at the round structure. The game is clearly intended to be hard to lose, so it has an unusual number of catch-up-when-you're-behind mechanics that are mostly original.
It has a "different objects offer divergent amounts of utility to different villagers" mechanic that I actually don't recognize from a prior source -- I'm sure someone did it, maybe Tropico? -- that I feel ties in super well with the game's topic of fantasy racism.
There are also progression-oriented design decisions made above the round-level structure of the game -- basically, between rounds your character gets better. These are also clearly influenced by Hades and other roguelikes.
From there, it's clear they didn't want the player to reach a point where the game was trivial or to reach a point where the player feels like there is no further improvement they can make to their settlement. The game therefore offers you an exit screen once you hit the point where future success is inevitable.
From there, they added mechanics where you can take extra risks to get progression sooner -- encouraging the player to kill villagers and feel like it is their fault, which matches the theme of the game.
They introduced a Nethack-style conduct system at scoring, where the players can receive greater metaprogression rewards for avoiding certain actions that would trivialize the game. This means that if the player picked a too-easy difficulty, they still have challenges there.
If the seconds of the game are spent instantiating and moving buildings around, I think that most of these can be understood as changes to the minutes, hours, and days of gameplay.
There's actually very little to speak to the game's quality below that level of abstraction -- it's all in the broader systems!
Someone in a community I'm in is ideating about things that scan to me as schizophrenia symptoms.
It kind of feels like it would be disrespectful to repeat exactly what they are ideating about, but they're talking about receiving auditory messages from distant sources and they're giving examples of government activity designed specifically to screw with them.
What I've noticed is that I am OK at responding to "behavior that looks delusional" in a two-person setting, but incredibly bad at dealing with it in a three-person setting.
Because there's a moderator here now who's responding to their text by pointing out that it is disorganized but not disputing any of its content. And I guess my thought is -- OK, most people who are having thoughts like this are never going to be particularly organized.
I don't know if they know this and are pointedly not mentioning it, or if they don't know this and are going to keep going in circles with someone who is experiencing (at minimum) severe executive dysfunction.
(My guess is that this mod knows a lot, but one of the things they know is the value of kayfabe. I feel a sense that I may have disrupted that.)
Separately in the same place, I kind of encouraged someone into baiting me into doing a conceptual analysis of Twitter teasing --
That is, the thing I posted about yesterday -- the thing where doms write very generalized posts intended to recruit submissives and you read it intending to self-insert.
I think that to give people the feeling that it is about them, it is stereotyped that the submissive reader:
1. clearly wants to be a servant of some kind
2. has a limited ability to resist and no genuine desire to resist
3. is interested in sexually servicing whoever is posting the Tease
4. is looking to be regarded as a member of a community connected to the person who did the Tease
An influencer comes out of nowhere and says something like "I'm putting you in a maid outfit [like all the others]" which, at first blush, seems totally unrelated. (Like, that's cross-dressing -- not everyone is into that -- is that even a hypno post?)
But it can totally be interpreted as an encoding of 1, 2, and 4.
I think people who want to submit to this kind of thing will often translate it in their heads to its underlying meaning, even if the specific kink content doesn't exactly work for them.
Someone very young joined a writing server I'm on a few days ago.
She's come in every three days with a question like "I'm writing about murder; is that bad?" "I'm writing about cannibalism; is that bad?"
Everyone reassures her -- "that's fine, you're not a bad person for writing that kind of thing."
So now she comes in with a new question:
"People are shipping my characters! How do I make them stop?"
I think my answer is that you go to the store and buy yourself a cake to commemorate that your fic has enough readers that they are making fanworks about your content.
I think I understand why this is different for her, though. Based on her comments, this is likely a self-insert fic. If people are shipping her self-insert character with other characters -- well, I could see why that would make her feel gross!
Unrelated -- in another chatroom I'm on. Someone dug up a Twitter post where someone claimed that Bluesky just isn't as debauched as Twitter is. They wanted to get the goblin-stays-up-at-night-jerking-off energy from reading everyone's posts -- from what I can tell, they mean the teases that dominants put on main as a way of making overtures to submissives.
I have a version of this rant saved in my brain, but it comes out slightly different every time. This rant is about me and to the extent it applies to anyone else, it's still projective --
When I went on Twitter and represented that I was horny, I was usually lonely.
I think when you go into a room online and people represent it as a massive party where everyone is having fun constantly, that is not true, because such things are not sustainable.
I think that typically in a room full of people who are loudly insisting that they are having fun, some people actually are having fun. But there is also a population of people who are not -- and those people are seeking to emulate the expressions of those who are. And the expressions they seek to emulate are usually the more extreme ones.
This is not even close to being unique to Twitter. This is my suspicion when ostensibly happy couples post detailed text about how well their relationship is going (it's often correlated with a vacation -- the iconic example of an unsustainable activity) -- or when people on LinkedIn post about how much they love their jobs.
I don't know what I can recommend to other people as a tool for having fun. I like fun, but there's a neurotransmitter tank in my brain that rapidly depletes to empty when I have too much of it.
I think the reckless pursuit of fun has worsened my depressive tendencies.
"So, without further ado, welcome to South Korea, folks, and let me tell you, there is a heck of a lot to talk about when it comes to this nation. Now, unfortunately, it's pretty famous for having one of the most heavily militarized borders of the world -- which, of course, it shares with its northern counterpart. North Korea.
"But there is so much more positivity to refer to when it comes to the nation. For example, South Korea seems to manage to blend ultra-modern cities with jaw-dropping geography quite perfectly. Plantations sprawl across the rolling countryside. Striking cliff faces rip through the coastline, and temples and rivers are scattered throughout the land."
"But the real fascination at least for me comes when you step into the city of Seoul. This place is absolutely unbelievable, to say the least. With a population of 10 million residents, this city still maintains its position as one of the best cities to live in, with the highest quality of life. And this means that Seoul is extremely popular, all while remaining tremendously fruitful."
"Its economy primarily focuses on advanced manufacturing technology, finance, and commerce. And as you can likely imagine, all of these industries continuously exceed expectations and worldwide standards. When it comes to modern practice, with an average salary of forty thousand dollars per capita, living in the big city does come at a premium."
Coffeehouse Crime is a true crime YouTuber.
Those four paragraphs are taken from his video on the Nonhyeon-dong Massacre, in which arsonist and stabber Jeong Sang-jin killed six people and injured seven more.
You can basically see the places where he splices phrases like "to me" and "quite perfectly" into the humdrum musical language of the source text in order to personalize it to himself.
I absolutely adore him and his incredibly gormless energy.
Anyway, lately he managed to make me pretty mad.
I'll skip the obligatory travelogue and get to the good stuff -- Terryon Thomas is a content creator who is accused of murdering therapist William Abraham in September 2024.
Terryon hasn't gone to court yet.
The reporting on this crime is inconsistent and full of holes, but that's not surprising: basically any detail that can be linked to the case is a detail issued by a cop, and none of this has gone to trial.
What police assert:
- William Abraham and Terryon both lived in Baton Rouge.
- William Abraham's body was found in a tarp by Highway 51 near Tangipahoa, about seventy miles away.
- The day after the murder, officers stopped someone in eastern Baton Rouge who was driving William Abraham's car. That person interacted face to face with the officer and vandalized the officer's patrol car. Despite that, they evaded arrest in an unspecified way and crashed the car elsewhere.
- The officer involved in the traffic stop later picked out Terryon Thomas in a lineup. There is also body camera and in-car camera footage supporting this.
- There is surveillance footage as well as eyewitness accounts supporting the idea that Terryon Thomas physically visited William Abraham's house.
Additionally, online jail records prove:
- Terryon Thomas was arrested in Dallas, Texas about a week later.
So -- Coffeehouse Crime knows a bit more, but basically all he knows is details connected to Terryon Thomas' online persona. The gist of his account: Terryon Thomas is a very edgy TikTok guy who stole a lot of video ideas from other creators and was widely hated. Terryon Thomas claimed to have bipolar disorder and was in a very low mental health period at the time of the murder, and had periodically joked about killing someone.
Coffeehouse Crime goes on to infer that:
- Maybe the jokes were earnest?
- Maybe William Abraham was Terryon Thomas' therapist?
- He probably went crazy and killed his therapist.
I am not saying this is literally impossible.
However, I think it's incredibly irresponsible to make a video accusing Terryon Thomas of this when there's no evidence clearly connecting him to the crime.
I think the police in this case have behaved in a way that is consistent with corruption. My reason for thinking this is that they have shown incredibly wild incompetence when it comes to evidence that would have the tendency to exonerate Terryon Thomas.
For one thing -- if there actually was a face to face traffic stop with someone driving William Abraham's car, it's insane that the person wasn't apprehended and was instead able to get back into the car, drive it away and crash it elsewhere.
For another thing -- they're claiming that the stopped person's face was on video. If that were true, why did they bother with a lineup? They had an image of his face -- he could have been positively identified from that. Why pick the lossier, less accurate method when there was a guaranteed-reliable method easily available?
For another thing -- so, we know a little bit from local news about how the police found evidence of Terryon's presence. Basically, they printed off a large number of fliers featuring images of Terryon in a convenience store, asking if anyone had seen him.
This particular campaign infuriates me because it's stereotypically the kind of thing likely to create false memories -- if you show everyone in the neighborhood a picture of Terryon Thomas's face, they're vastly more likely to indicate that they've seen this man before when asked about it in a police interview.
It infuriates me for a second reason because the choice of these particular photos implies that they didn't have any better ones. If they had distributed zero photos, we could have assumed they were somehow injoined from distributing evidence -- but instead they have only distributed evidence of a very, very low grade.
By the way, Terryon doesn't look like someone who has recently been in a car accident.
Of course -- I don't know, I guess Coffeehouse Crime has his favored theory, and it happens to match what the cops are asserting without real proof.
Something I've held back -- Terryon Thomas was publicly LGBTQ and William Abraham was a public LGBTQ ally.
I think this pattern is consistent with the police having a pretty good idea of who did it, but wanting to arrest an unrelated black man instead. I think it is consistent with having a lack of real interest in finding justice in this case.
Generally I watch this guy because it's kind of fun to watch an incurious and inarticulate person stumble badly through an AI script, and he's kind of hot. But it's genuinely frustrating to see true crime YouTubers uncritically adopting the opinions of the police state.
I think my request of people every day is always the same: I am not asking for kindness and niceness, I am not asking for anyone to exist in any particular way around me. I am saying please do not rip the throat out of anyone who has not consented to it.
This is the third thing I wrote. (I got rid of the other two things, so you won't see them)
A friend wants to hang out and I'm thirsty, so I'm going to leave you hanging on this one.