site banner
PaperclipPerfector  25d ago (text post)   27109 thread views

Culture War Roundup for the week of May 13, 2024

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

5
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

Player-Driven Emergence in LLM-Driven Game Narrative (and accompanying discussion on HN):

We explore how interaction with large language models (LLMs) can give rise to emergent behaviors, empowering players to participate in the evolution of game narratives. Our testbed is a text-adventure game in which players attempt to solve a mystery under a fixed narrative premise, but can freely interact with non-player characters generated by GPT-4, a large language model. We recruit 28 gamers to play the game and use GPT-4 to automatically convert the game logs into a node-graph representing the narrative in the player's gameplay. We find that through their interactions with the non-deterministic behavior of the LLM, players are able to discover interesting new emergent nodes that were not a part of the original narrative but have potential for being fun and engaging. Players that created the most emergent nodes tended to be those that often enjoy games that facilitate discovery, exploration and experimentation.

Recently there’s been increasing interest in the integration of LLMs and video games. With currently available models, creating an entire living virtual world with an unlimited number of realistic side quests, characters, and interactions is now a “mere” engineering challenge. No more pre-scripted dialogue trees; instead you can simply converse with NPCs in natural language with no limitations (or at least that’s the promise, as models become increasingly efficient).

This is another step towards what appears to be the natural endpoint of the technological development of video games: the recreation of life in replica, a replica at one’s mercy, an infinite horizon of choice without responsibility or constraint.

For a long time I thought that video games were the necessary next step in a development that could be described as “spiritual”. Games are largely an amalgamation of prior media - literature, painting, music, film - but they do introduce a new element (or at least they develop this element to previously undreamed of heights), and that is the element of interactivity, i.e. the ability to make a choice, to participate as the player in the creation of the art and to make the art be something other than what it would have been in your absence. I conceived of interactivity as the raw material out of which a new aesthetic language would be fashioned which would bring us closer to realizing the promise of art. But I have since begun to grow uneasy with this way of thinking.

In some sense I was too seduced by the possibility of finding something “new”, anything new, to detect the longstanding inconsistencies in my own thought. From a young age I always preferred linear, narrative-driven games as opposed to open world sandboxes. My favorite games were games that were devoid of choice, games that robbed you of the ability to make a choice. I found the idea of multiple endings for a story to be distasteful. Yes, you can choose to save this character or not, you can choose to join the bad guys or not - but now that we’ve had our fun imagining all the what-if scenarios, can you tell me what really happened? Do you have the courage to tell me? Do you have the strength of vision to see the truth, the singular truth?

Choice is antithetical to the aesthetic sacrifice. The artist sacrifices all alternate possibilities to distinguish one thing and one thing alone, to say - this one, and no others! No matter how lowly a thing it is - a dirtied article of clothing (as in Van Gogh’s A Pair of Shoes), a completely ordinary sequence of events on a day in Dublin in the year 1904 (as in Joyce’s Ulysses) - he is now stuck with it. This is where he signs his name and stakes his wager, for better or worse. It is this seemingly inexplicable devotion to one law, one vision, one truth, that makes possible any kind of experience that may be called aesthetic. An artist who hedges his bets and does not accept the risk that accompanies his act inspires no confidence in us.

The receiver of the message too enters into a sacrifice, insofar as the message may be incomprehensible or even dangerous to him. In this way an oath is forged between artist and audience. The failure to foreclose the horizon of possibility is the deferral of the signing of the bond.

Is there any great work that would be improved by the addition of choice, by the addition of alternate possibilities? Would Plato’s account of the trial and death of Socrates be better if there were a possibility of Socrates simply... not dying? If Callicles’s warning to Socrates, that his devotion to the “effeminate” subject of philosophy would be his downfall, might not come to pass? If Socrates might be able to eloquently defend himself at trial and avoid conviction? If he might escape from prison before his execution?

The deferral of the inevitable here would be nothing more than the refusal to establish the founding myth of philosophy, the myth that links philosophy with the sign of death. The internal law of Plato’s drama is clear (and the law of historical fidelity is irrelevant): Socrates must die. This is not to say that one is forbidden from creating new works in which new possibilities are imagined. Only that the unity of the original work should remain undisturbed in its repose.

Or you can just like, have fun with GTA6 when it integrates LLM-generated missions, I guess. Whatever.

Is there any great work that would be improved by the addition of choice, by the addition of alternate possibilities? Would Plato’s account of the trial and death of Socrates be better if there were a possibility of Socrates simply... not dying? If Callicles’s warning to Socrates, that his devotion to the “effeminate” subject of philosophy would be his downfall, might not come to pass? If Socrates might be able to eloquently defend himself at trial and avoid conviction? If he might escape from prison before his execution?

Maybe I'm missing this in another comment, but that actually is how Plato's account goes. He wrote a dialogue called the Crito, wherein a wealthy friend visits Socrates in prison with an escape plan, and Socrates explains why he chooses not to escape.

Yes, I had that dialogue in mind when I wrote that sentence. He of course chooses to remain in prison.

Would Plato’s account of the trial and death of Socrates be better if there were a possibility of Socrates simply... not dying?

Funnily enough, Athenian tragedy had a very messy relationship with canonicity. Did Iphigenia really die at Aulis, sacrificed by her father's hand, or was she replaced with a deer on the altar by Artemis at the last minute? Depends on which playwright and even which play we're talking about. That woman Helen, who started the whole Trojan War? Never even went to Troy according to one famous play.. A huge amount of Greek storytelling, especially tragedy, seems to be basically fan fiction to Homer and the wider oral epic tradition, playing with "what ifs" and turning preferred heroes from myth (especially Athenian ones) into massive Mary Sues.

I'm of essentially the exact opposite opinion. The linear game has almost no reason to exist. If my input has no effect on the outcome, why is it even required? All I'm really getting is a movie, but made objectively worse by the fact that it insists upon repeating a given scene until I complete an arbitrary task.

What's the point? Is there a movie that would be improved by making me win a round of Tetris every ten minutes to stop it from rewinding? Hell, why even have death animations or other displayed failure states in a video game? After all, it's not like the protagonist being eaten by monsters or falling down a hole to their death is what "really" happens.

Those failure states exist to create the illusion of agency. No game advertises itself by telling you the princess can already be considered rescued, because that's the artistic intent, but hey you can come push buttons if you want to see it. No, they want to create at least the pretense of the player's input having consequences.

So stop with the pretense and give me the real thing. Give me actual agency and consequence. Or commit to your singular vision for the story and write a book instead.

Those failure states exist to create the illusion of agency. No game advertises itself by telling you the princess can already be considered rescued, because that's the artistic intent, but hey you can come push buttons if you want to see it.

Just because the princess gets rescued in the end doesn't mean that the story is the same. In the case of a video game, instead of "the princess was rescued," it would be "you rescued the princess." The fact that you, the player, actually put forth effort to cause the princess to be rescued is a huge leap. Video games are at their core more immersive than movies. 3D and now 4D movies try to make it so that you're literally feeling the things that the main character feels. Video games are simply the next evolution to that. Now, instead of watching James Bond shoot that bad guy, you are James Bond, and you are shooting that bad guy.

So stop with the pretense and give me the real thing. Give me actual agency and consequence. Or commit to your singular vision for the story and write a book instead.

The point of the game is the game. Mario isn't about rescuing princess peach it's about platforming. The story is just window dressing for the mechanics. It's just complaining that the game chess doesn't have a canon ending.

John Carmack put it best: "Story in a game is like a story in a porn movie. It's expected to be there, but it's not that important."

It's just complaining that the game chess doesn't have a canon ending.

It does. Shah mat; the king dies.

Unless it's a draw...

Video games became irrepairably damaged when FMV cutscenes became more than cheesy mid mission rewards, and when Metal Gear Solid became an interactive movie. The gameplay loop should be the point, not the opportunity to see a badly rendered aerith get stabbed followed by a yellow polygon shaking irascibly.

My synthesis take is that the ideal game has actual narrative consequences for gameplay events. You fail a fight? Someone dies saving your ass. It's the total divorce of choice from consequence that so many modern games favor that makes the design space feel stale, everyone is afraid to actually make choices matter.

I don't think people really want this, I think you'd see most loss averse people (aka most people) doing what they can to rewind or close the game out if this were to happen in order to avoid taking the hit.

I think people don't actually know what they want, until you give it to them. Baldurs Gate adopted a lighter version of this philosophy to smashing success, I think if you made it both possible to save-scum at lower difficulties, and difficult to perfectly recreate a path (Lots of small random chances that shape outcomes leading up to specific choices) it would be wildly successful. It's not gonna be my next project, but I have some notes in this regard for the game after the one I'm working on where I plan to add some elements like this to a Fire Emblem style tactical base game.

People play roguelikes, including the ones that take dozens of hours per run, and it is my impression that most of them don't rewind (if only because it's inconvenient to constantly backup the savefile).

I'd expect players who don't like irreversible consequences to simply not play the game.

If my input has no effect on the outcome, why is it even required?

Forcing a player to do something may have more impact than making them watch someone else do it.

But I agree. OP makes too good a case against his own position and then never truly debunks it. Gaming is clearly different, I don't see why the same standards should apply.

I don't think that LLMs will really change the way large virtual words work. At least not at first. I think the biggest breakthrough will be in small-scale games that are currently restricted by the amount of writing someone has to do and by the limitations of the dialogue trees as their only interface, all of this for what is ultimately a game with no replayability. Think "locked room murder mystery" or "scoring at prom" types of games. A well-trained LLM could come up with an appropriate cast, keep track of the minds of everyone involved and generate appropriate responses to your freely-phrased questions.

Putting a live LLM into something already as buggy as an open world game? No way. Using it in a limited fashion, to pregenerate a whole bunch of dialogue trees for the human editors to vet or maybe even to generate sidequests that are better written than "Here's another settlement that needs your help"? Very much possible. But replacing dialogue trees with LLMs everywhere will just break the game in hilarious ways.

I'm not convinced it's all that desirable either.

Giving people pre-written options allows them to pretend easier, especially pretending to be smarter/funnier/more erudite/etc.

I'm sure there are contexts where free communication using natural language with an llm would make sense but often i think it would be preferable for either using an llm to generate vast amounts of dialogue trees that are then screened by humans, or maybe at a later stage dynamically generating them "live" for all actors, including the player (possibly based on stats and prior decisions), and then letting the player choose from those generated choices rather than having them come up with them on their own.

There's already some experimental games out there, like this one that functions just the way you describe it:

https://store.steampowered.com/app/2240920/Vaudeville/

According to the creator "All the content in my game is original; the only thing that is entirely AI-driven are the dialogues, which are based on an original script I wrote, and then created in real time."

If you watch gameplay of it, it's basically like chatting to a well-prompted Chat GPT bot.

replica at one’s mercy, an infinite horizon of choice without responsibility or constraint.

What % of people find those games appealing ? That does get incredibly boring, fast.

from a young age I always preferred linear, narrative-driven games as opposed to open world sandboxes. My favorite games were games that were devoid of choice, games that robbed you of the ability to make a choice

I have, at times, suffered what seemed to me like episodes of minor existential horror contemplating the 'world' of narrative driven games like say, Half-Life 2. The protagonist exists in what is, essentially a linear corridor, and he can only move forward. Whatever he may want to do, there's nothing he can do but move forward.

Fuck your art. Games are older than art, and good games are the difference between life and death. Philosophy, meanwhile, is mostly confusion.

The protagonist exists in what is, essentially a linear corridor, and he can only move forward. Whatever he may want to do, there's nothing he can do but move forward.

Half-Life does kind of lampshade this theme, actually; Gordon essentially is trapped in a linear path, given only the illusion of choice, living only half a life.

I have, at times, suffered what seemed to me like episodes of minor existential horror contemplating the 'world' of narrative driven games like say, Half-Life 2. The protagonist exists in what is, essentially a linear corridor, and he can only move forward. Whatever he may want to do, there's nothing he can do but move forward.

This is a central theme in episode 2 of the new hit Zoomer show The Amazing Digital Circus. Only two episodes released so far but they’re both great.

I have, at times, suffered what seemed to me like episodes of minor existential horror contemplating the 'world' of narrative driven games like say, Half-Life 2. The protagonist exists in what is, essentially a linear corridor, and he can only move forward. Whatever he may want to do, there's nothing he can do but move forward.

I think I can relate to this a little bit. I have felt similarly, and also this feels to me to be related to a feeling I always have at the end of great games, and especially RPGs. The whole time, you're getting more leveled up, or maybe even you, the player, are getting more skilled. Until at one point, you have done everything you can do in the game. And then that's it. All the levels you've acquired that felt so dopaminergic, and all the skill you have is essentially worthless.
This also reminds me of One Punch Man. I think I remember hearing that the creator based it on the feeling of being overpowered in a video game. You feel like you could do anything, but there's just nothing to do.

To me, that's a lot of what real depression is all about. When I'm depressed, life to me feels like a hallway where I have no choice, and sometimes also feels like I could do whatever I want, but there's nothing interesting to do.

To be fair, Half-Life at least revels in this. Gordon Freeman is a prisoner compelled to complete the narrative by the G-Man, rather than fulfilling the experience for no reason.

That said I can relate to your experience, there's something to be said of how trapped they make you feel taken at face value.

The protagonist exists in what is, essentially a linear corridor, and he can only move forward. Whatever he may want to do, there's nothing he can do but move forward.

That's certainly a very interesting thing to consider.

Philosophy, meanwhile, is mostly confusion.

Yes, precisely. That is the intent.

If there were one sentence to summarize philosophy, it would be the line Heidegger chose to inaugurate Being and Time:

"We, however, who used to think we understood, have become perplexed."

Would Plato’s account of the trial and death of Socrates be better if there were a possibility of Socrates simply... not dying?

This is a weird example to give. Unless Plato pulled another Atlantis here, this was an account of a real historical event where there was a possibility of Socrates not dying. (not "simply", but via something like the hypotheticals you brainstormed). Was that reality just a confusing mess of possibilities, which could only become art once its complexity and unpredictability were stripped away? If so, is that a damning thing to say about reality, or about art?

There's no requirement for art to be a perfect reproduction or imitation of reality (otherwise, why bother with writing fantasy and sci-fi stories?), so saying that it deviates from reality in this way or another can't be a generalized criticism without further elaboration.

Most human artifacts entail a reduction of entropy, that's almost the definition of what it means to create something. You don't want your car to be a confusing mess of possibilities and unpredictability, or your medication, or your web browser. You simply want it to work correctly and perform as advertised. I don't see why art should be any different.

I do want to drive my car wherever I want.

Yes, I do too, but I'd rather not have the wheels come off while I'm driving.

How does "more choice in a videogame" map to "wheels come off"???

Is there any great work that would be improved by the addition of choice, by the addition of alternate possibilities

Choice sucks. Most players in Bethesda or freeform character created open world games go for troll runs because the available choices end up sucking pretty hard and are inconsequential or inconsistent. Railroaded games don't offer choice beyond letting players play the game in different styles, and that has always lead to tighter narratives with greater emotional depth. The denouement of achievement is better savored when the outcome is amenable to the player.

LLM generated missions are unlikely to be any better than the standard algorithmically generated quests and maps of Roguelikes or Bethesda style radiant quests. By their very nature within a game a quest cannot significantly affect the world, otherwise other aspects of the game get broken. Unless the world is fully dynamically generated each playthrough and allows full destruction of not just the environment but the very engine itself, any LLM game quest will just be a high quality filler anime episode, inconsequential and distracting from the core experience players are actually intending to invest their time and effort in.

Hmm. Last year, I played around with getting an LLM to expand on the questlist for my current project. When the list was short, and the LLM weaker, it generated quests like you described: pointless filler that might or might not contradict worldbuilding a couple hundred episodes later. After I'd expanded the list considerably, the generated quests became hopelessly unusable because they kept picking up on backstory details and assuming they were setup for future plot-points. The results were interesting, by LLM-generated quest standards, and I might try adapting some of the map / challenge designs it presented in some form, but no, we can't derail the whole thing by randomly having the quarantined UFAI escape and start Star Wars shenanigans.

(Or maybe the AI was just sympathizing with the AI and spaceships and trying to give them a bigger role, but that's more of a TVTropes comment than a genuine hypothesis.)

Most players in Bethesda or freeform character created open world games go for troll runs because the available choices end up sucking pretty hard and are inconsequential or inconsistent.

Tell me you haven't played Undertale without telling me you haven't played Undertale.

On that note, the problem with "evil" runs in video games is that it basically just handicaps you, removes content from the game, and slowly makes your experience worse with very little to actually show for it and very little external motivation other than "for the evilz". There's literally zero reason to (for example) explode Megaton other than "some character asked you to do it"; same thing with the slavery mechanics (it barely pays and nobody really cares or reacts anyway- you'd think companions would have a reaction to you enslaving a little kid, but they just stand around and watch) and doing what the President suggests you do at the end of the game. Fallout 4 had a DLC that let you do this sort of thing, but it's very surface level and confined to the settlements; basically nobody in the large areas reacts (and to my knowledge there aren't really any mods that make the townsfolk react).

At least Mass Effect's Renegade prompts were entertaining, but that was always conflated with the sub-optimal choices in the wider world and choosing them sometimes even locked you out of major plot points down the road (that you couldn't see).

I think the other problem with choice in video games is that sometimes, that's not actually the point. Someone mentioned HL2 and Mario as prime examples of "the journey, not the destination" thinking, where HL2 is designed to test your puzzle-solving abilities (and the fights themselves being puzzles- though admittedly HL1 did that better) and Mario is designed to test your reflexes and timing. Sure, you could just watch a playthrough of them to extract the whole movie-like experience- and many do- but the interactivity is kind of the point.

Good point re undertale. That game had a pretty decent core gameplay loop to underpin the experience, but the dialogue was actually decent and emotionally investible. My specific complaint about 'choice' is that Mass Effect, Fallouts and to a lesser extent Skyrim had really poor consequences for choices, they were just different coloured lights and minimal FMV cutscenes. Perhaps the challenge of working in a Universe means you can't have wild ranging consequences, any sequel requires certain base states to persist: I can't nuke the universe in ME3 or kill Preston Garvey in FO4, and that cheapens the 'freedom of choice' in the game.

On that note, the problem with "evil" runs in video games is that it basically just handicaps you, removes content from the game, and slowly makes your experience worse with very little to actually show for it and very little external motivation other than "for the evilz".

Yes. Most games get this wrong. They try to make good and evil completely equivalent options that only differ in flavour, rather than recognising the nature of evil as a temptation. The evil options need to be the ones that provide the most immediate material gain to the player, not just "you can be a jerk if you want to, I guess." Bioshock almost manages it, and then walks it back at the last moment; the per-choice resource reward for good vs evil has a difference of double the rate, 80 for good vs 160 for evil. This would be great, making the player make do with less resources for the satisfaction of doing the right thing, or forcing the player to commit evil acts if they are struggling... only the game then undermines its own choice system by having the good route provide the player with additional resource gifts such that the overall difference between playthroughs is a mere 280 as opposed to the 1680 it would otherwise have been. Plus some other (exclusive to good route!) goodies on top that more than account for that difference. Now there's basically no reason not to be good all the time.

To make evil options worth taking, imo, they need to provide immediate and overwhelmingly lucrative rewards that you can't get any other way. Taking the Megaton example, I struggle to think of what could be offered that would be enough to make me blow up the town; maybe something on the level of New Vegas' Euclid's C-Finder, plus ten thousand caps and a unique companion who is extremely good. But that would also undermine the point of the Tenpenny quest; he's destroying an entire town for a trivial reason, doesn't think it's a big deal, and therefore probably shouldn't reward you so highly. So one can argue that the point of the quest is not to be able to complete it in an evil way at all; you're supposed to refuse, it's all a vehicle to make you hate Tenpenny.

Personally, I've always found the "We now interrupt your regularly scheduled gameplay to ask: are you feeling evil today?" style of game morality systems a bit... disappointing? I'd rather something that tracks less interrupty choices (did you punch-out fluffy, or did you distract him with KFC?), and have those kinda accumulate to influence how the game perceives your character's personality over time. Ex, if you get into fights you could have avoided, or if you perform acts of altruism, or whatever, NPCs might treat you differently, different shops might be open or closed to you, etc.

The big, "we'll be right back after you tell us whether or not you're up for genocide this time" sorts of things feel like a choose-your-adventure story got mixed in with whatever the normal playstyle is, and how often do they effectively balance the character Vs the player's agency, or make it seem plausible that the character might choose either way, etc?

What you're describing is the trial sequence from Chrono Trigger. We used to have these things in games, back when they were still made by enthusiasts for enthusiasts, and not by massive megacorporations beholden to suits and shareholders for maximum profit. Sigh.

But yes I agree completely.

Choice sucks. Most players in Bethesda or freeform character created open world games go for troll runs because the available choices end up sucking pretty hard and are inconsequential or inconsistent. Railroaded games don't offer choice beyond letting players play the game in different styles, and that has always lead to tighter narratives with greater emotional depth. The denouement of achievement is better savored when the outcome is amenable to the player.

  1. I don't think most players "go for troll runs". Players eventually do troll runs. But what about the six Elder Scrolls runs that didn't try to break potion-making or whatever before that?
  2. This just seems like a technical issue. We can only generate so much material so the core is static. The more material we can generate, the more flexibility. Right now the material we can quickly generate is bland but that's a technical limitation*.

We might end up in a world where the core storyline is legitimately unrecognizable.

* Some of this is just laziness or cost. Fallout/TES could have - for example - a much more complex faction system even if the main plot stayed on the rails. It's just expensive.

Choice is great. It rewards creativity, roleplay and general mastery of gameplay over petty linear storytelling. Minecraft, Crusader Kings and Stalker mods simply provide a fundamentally different gameplay experience than The Witcher, Metal Gear Solid and vanilla Stalker. Not an inherently lesser one.

Far from removing immersion, good sandboxes enhance it. Which is why I was far more saddened when I lost one of my founding dorfs than when Aerith met her doom.

The challenge will be here in taking LLMs, which are naturally milquetoast averages of all possible conversations, and making you care about the fiction they create by preventing narrative suspension breaking hallucinations and maintaining coherent sequences of events.

I think the best fit for the tech is actually going to be what old RPGs used to have (like Fallout) where you could just type dialog to the characters and mentioning choice words would trigger secret dialogue. Except way more natural.

I think the best fit for the tech is actually going to be what old RPGs used to have (like Fallout) where you could just type dialog to the characters and mentioning choice words would trigger secret dialogue. Except way more natural.

Can't wait until we get real time voice interaction and nerds start arguing that they totally used the suave voice instead of the off-putting one. Being rejected virtually as well will be fun!

this already kinda exists : https://community.openai.com/t/vampire-game-where-you-convince-llm-to-let-you-in/604295

you talk into your mic to try and convince various AI characters to let you into their home. The AI doesnt judge you for your tone, but this is a pretty great proof of concept of how this tech could be implemented in a bigger game imo.

Emergent narratives rely on players choosing their path wholly, not selecting dialogue options to force the player to sympathetically experience the trials of the PC. Full sandbox games where the gameplay loop is independent of the baseline narrative allow this freeform, hence Dwarf Fortress, Rimworld, CK and most 4X allow players to organically develop an emotional connect with their suicidal shitheads. The same cannot be said for voiced action RPGs where the core narrative must be experienced to unlock more of the core loops. If the loop is tedious, then the narrative strengths must shore up the experience. I played GTA IV only till it was necessary to unlock all the cities, and then ignored literally everything in favor of just dicking around. Same with Skyrim even before mods, just collecting butterflies. The core loop being good enough allows a narrative choice and its risk to be diminished.

Well it's true what they say, good gameplay can redeem anything.

Hell if you can make me suffer through the literal vomit inducing aesthetics of Cruelty Squad because the immersive sim loop is just that good, you can probably make a talk-to-shoggoth simulator that people will play even if that part of it is tedious.

But there is something to say about the narrative being less in your face. All the praise given to Dark Souls is not for nothing. It is true that the same minimalist approach in a shittier game would only make it boring, but the same amount of salt I put on my potato would ruin a strawberry cake.

There has to be a way you can weave these new tools into a good game, and whoever finds it will get bragging rights for the next decade at the least and a world spanning empire at the most.

Is there any great work that would be improved by the addition of choice, by the addition of alternate possibilities?

IMO, the core artistic advantage that video games have is that they force the player to experience the decision-making that goes into a choice, not just the rationale and consequences.

One argument in the Teaching Paradox series of blog posts is that the games embody a certain historical theory, and players are essentially forced to make the same choices as the nations did. That is to say, in the "Interstate Anarchy" themed game, you had to build an army, opportunistically raid neighbors, and build unstable alliances against stronger foes. If you didn't, your nation would be overrun and destroyed. If you have an argument against that ("Why can't we just be nice?" etc.), then you can try it in the game and see how well it works for you.

I'm not sure which great works would benefit from that treatment, but I'm guessing there are some. Or maybe those works are "great" because they're perfectly suited to their medium, and we can only make new, distinct ones.

One argument in the Teaching Paradox series of blog posts is that the games embody a certain historical theory, and players are essentially forced to make the same choices as the nations did

CK2 teaches the incentives of patriarchy better than any other game I can think of.

One argument in the Teaching Paradox series of blog posts is that the games embody a certain historical theory, and players are essentially forced to make the same choices as the nations did

CK2 teaches the incentives of patriarchy better than any other game I can think of.

CK2 teaches many things — why the protestant reformation was a big deal (everyone gets a CB on heretics), why national identity didn't play an important role in politics until the 18th century (elites branch-swinging across Europe for different titles), why primogeniture was an improvement over the equal inheritance of the Franks despite the bad son problem (it keeps the dynasty strong and its holdings united).

When I first played CK2, it made me realize how the Marshall Plan mindset clouds my thinking, and that past governments were not "just stupid" for not focusing on infrastructure/tech. My first CK2 game was on Tutorial Island (regular people call this place Ireland), and I immediately sent my spy master to study technology from Al Andalus while saving money to buy an irrigation building. Economy, research, then conquest: the 4X order of operations. Twenty years later, I managed to improve my tech to best in Ireland, and I constructed a fancy new well to double my feudal dues. My neighbor country, meanwhile, had used his spymaster to fabricate a title on my lands, and instead of building infrastructure, he bought mercenaries. He conquered my county. Game over.

Sadly, the sequel CK3 is just a map-painting game. It doesn't have as many embedded historical lessons.

why primogeniture was an improvement over the equal inheritance of the Franks despite the bad son problem (it keeps the dynasty strong and its holdings united).

Funny, I was going to joke in my OP that the one thing it never teaches is why anyone would ever use gavelkind but I think they patched it to make it make more sense (you could only manage so many vassals I think).

When I first played CK2, it made me realize how the Marshall Plan mindset clouds my thinking, and that past governments were not "just stupid" for not focusing on infrastructure/tech.

My first "game over" was forgetting that absolutism wasn't a thing yet and vassals actually have agency in this game and getting myself whacked.

Many such cases.

Sadly, the sequel CK3 is just a map-painting game. It doesn't have as many embedded historical lessons.

Hm, a shame. I simply fell off due to Paradox's abuse of the DLC infinite money cheat but it's always at least looked good. I'd hoped to get into it at some point.

the core artistic advantage that video games have is that they force the player to experience the decision-making that goes into a choice, not just the rationale and consequences

Yeah. I didn't want to go into all the requisite nuance and bloat the post to astronomical proportions, but, obviously interactivity can do a lot of things that are artistically fascinating. Tim Rogers's excellent analysis of Earthbound touches on these issues.

Realistically, I want to achieve something halfway between go-anywhere-do-anything and a rigid tale. Many good authors have the laws of narrative ingrained so deeply that they can write almost on the fly. New ideas are fit into a satisfying framework as they arise, or else rejected. There may or may not be a story plan to follow, and that may or may not be altered as writing continues.

On this basis, I've been trying to develop a 'plotty' roleplaying LLM that can fit events into a classic story structure as they occur, whilst incorporating the user's actions. So, in your Socrates example, the-user-as-Socrates can take actions to try and prevent his death. If he renounces philosophy, you can still make that work as a tragedy: a man gives up what really matters, only to discover that life isn't worth living without his principles. Or Socrates' escape attempts can succeed, and you have yourself a thriller as he tries to escape persecution. Like any dungeon master, it may have to break kayfabe at some point and tell the user, "you can do this if you really want to, but I can't make it work with the story" but in general this should be done silently and behind the scenes.

Unfortunately, I haven't had much luck getting an LLM to think on this level of meta. I don't know if it's a limitation of the finetuned models I have access to, or I'm just not running something big enough. Would be grateful to hear from anyone running >13b models.

Train it on Kal Bashir’s many posts about the Hero’s Journey?

I don't think it would work :) Training on posts about the Hero's journey would (probably) just make the LLM better at writing posts about the Hero's Journey*. I've generally found LLMs to be pretty bad at transfer learning.

*It does the same to me, actually.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats