Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Dark Forest Theory of the Internet

onezero.medium.com

Post reply on HN
Print view View original

171–180 of 182 posts

Re: The Dark Forest Theory of the Internet

#171

Post by dorgo »

YeGoblynQueenne wrote: post #164
Freaky wrote: post #157

Earlier quoted context omitted.

> And it's hard sci-fi because they were AIs made of protons, not actual fairies Does mentioning your faeries are made out of atoms stop them being fantasy? > That's how it goes with sci-fi, innit. You can come up with anything you like as long as it's obvious that it's just advanced technology, not magic. Sure, I mean, even Star Wars is still considered within the genre despite it all being completely made-up - but…

My opinion on this might be controversial but I think there's very little difference between even "hard" sci-fi and all-out fantasy (but, with spaceships). There is very little space left to write a story that is interesting and compelling if one wishes to respect the bounds of what is scientifically plausible. Then again, if one starts to stretch the definition of "scientifically plausible" there's all sorts of thin…

>There is very little space left to write a story that is interesting and compelling if one wishes to respect the bounds of what is scientifically plausible.

Are we living in the same world? Quantum mechanics, relativistic effects, mathematics ( game theory anyone? ) are blowing my mind. The world is more phantastic than anything one could imagine and we know that we don't know everything. The constraint to stay in this world is the least limiting for an interesting and compelling story.

Re: The Dark Forest Theory of the Internet

#172

Post by notduncansmith »

ColanR wrote: post #170
notduncansmith wrote: post #166

Earlier quoted context omitted.

The idea is not to assume that everyone will be superrational or trustworthy at the beginning of the game. I believe it’s possible for the superrational players in a game to become the majority by expressing the superrational strategy in terms that will appeal to rationally self-interested players and demonstrating a superrational strategy working at some non-trivial scale.

Even so, that idea merely expresses the hope that the prisoners can be educated out of their mistrust - which still seems like saying, 'the dillemma doesn't exist, if we fix the prisoners'. (for the sake of argument, it looks like we're both assuming a n>2 group of prisoners.) Also, the point of the dillemma is that only one prisoner has to act selfishly. If a subset of the group does cooperate, it doesn't matter bec…

In the prisoners’ dilemma, it is true that even a single bad actor has the ability to disrupt the rest.

One limitation of that thought experiment is that it assumes defection is guaranteed to achieve a certain outcome.

I think it’s reasonable to assume that any cooperative strategy has some computable “defection tolerance” and a variant on the prisoners’ dilemma that accounts for this could have very different outcomes than the classical formulation.

Furthermore, once you start playing multiple times with the same players, you establish an algebra of trust that also drastically changes the game. The ability to influence other players between rounds also changes the game.

Re: The Dark Forest Theory of the Internet

#173

Post by ethbro »

sandworm101 wrote: post #125
ethbro wrote: post #74

Earlier quoted context omitted.

I've never understood the association with relative time in regards to science fiction. SF is typically futuristic, but that bears jack all as to whether it takes place in the past, present, future, or even anything relative to our planet. Science, in that it pledges greater allegiance to, extrapolates, or invents in an internally consistent way. But not time.

Science. Scifi is fiction about where science might take us, how modern tech might impact society. It is foreshadowing a potential future. So stuff set in the past, generally, isnt considered scifi unless it somehow describes a potential future. This separates scifi from technothrillers (tom clancy et al) that are tech-heavy but set today rather than in the future.

> So stuff set in the past, generally, isnt considered scifi unless it somehow describes a potential future.

This makes literally no sense to me. Any definition of the future that doesn't include the past is just fiction, and consequently, any past is a potential future.

What you're saying sounds more like "If it doesn't have greater than present levels of 'technology', then it's not science fiction."

Which seems incredibly limiting and ignoring of the central genesis of the genre, as you've pointed out -- science.

Science is a verb, not a noun.

The idea of pointing at something in a book and saying "That's science" (and correspondingly, pointing at something in other fiction and saying "That's not science") is odd.

Where do Mary Shelley, Jules Verne, or H.G. Wells fit in your definition?

Re: The Dark Forest Theory of the Internet

#174

Post by ethbro »

mikeash wrote: post #90
ethbro wrote: post #75

Earlier quoted context omitted.

I was taking issue with the medium post. "In his sci-fi trilogy The Three Body Problem, author Liu Cixin presents the dark forest theory of the universe." "Liu invites us to think about this a different way."

I don’t see that as implying he came up with the idea. They’re just referencing a recent use of it with a catchy name.

A singular name. It would have been pretty simple to be clear with "and others." Instead, the obvious conclusion drawn by any reader is that the idea is the principle creation of the only author mentioned.

Re: The Dark Forest Theory of the Internet

#175

Post by jlawson »

KirinDave wrote: post #155
jlawson wrote: post #150

Earlier quoted context omitted.

> If this isn't violence, then why shouldn't I stigmatize you? What do you care? ..... If speech, association and the implication of violence are so trivial that it cannot constitute violence You seem to have the idea that any act/speech which isn't literally violence must thus be trivial/acceptable/unobjectionable. Do I even need to explain why this is wrong? Obviously there are many acts which are unacceptable and…

> You seem to have the idea that any act/speech which isn't literally violence must thus be trivial/acceptable/unobjectionable. I haven't put that forth, but thank you for your concern. What I'm noting is that you're essentially decrying the labeling of some types of speech-as-violence as a tactic that according to you is being espoused by people to somehow cover the idea that they want to use speech tactics against…

>Why would folks ostensibly wanting to use the sharper sides of self-expression go and label their own actions as violence?

Nobody labels their own actions as violence. They label others' actions as violence, and either don't notice or hope to avoid the logical conclusion. People are not good at consistency.

People label others' words as violence in order to justify their own actual violence in return. E.g.:

"That speaker is racist and his words are violence, so we're justified in forcing open the doors and screaming and stomping to stop the speech."

>It seems strange to suggest that a declaration stripping you of many of your rights and property is not a violent act simply because the lawmaker themselves is a flaccid old man.

It really doesn't, any more than a restaurant recommendation is not an act of eating. Words have meanings.

A law to reduce pollution standards is not climate change, it is a law that will cause climate change. A law to increase government spending isn't a budget deficit, it's a law that will cause a budget deficit.

>I could set up a bomb to explode outside your car when the door opens and then claim I didn't open the door. If I undertake actions that are certain to lead to physical harm towards another how am I not in some way morally responsible?

You are morally responsible, because you're responsible for the outcomes of your acts.

This is so bizarrely simple, and I know you understand it, because you can't function in the world if you literally don't understand the difference between a cause and an outcome.

An act that leads to a result is not the same as that result. The initial act makes the actor responsible for the result. It doesn't mean that the initial act _is_ the result. This is a simple distinction present in all human languages for a reason.

E.g. If I recommend you eat at a restaurant, and you do, and enjoy it, I get some credit for that enjoyment, even though I didn't eat at the restaurant, I caused you to do it.

This is not a hair-fine semantic game, it's basic language understood by children.

Your final veiled threat is childish.

Re: The Dark Forest Theory of the Internet

#176

Post by ColanR »

notduncansmith wrote: post #172
ColanR wrote: post #170

Earlier quoted context omitted.

Even so, that idea merely expresses the hope that the prisoners can be educated out of their mistrust - which still seems like saying, 'the dillemma doesn't exist, if we fix the prisoners'. (for the sake of argument, it looks like we're both assuming a n>2 group of prisoners.) Also, the point of the dillemma is that only one prisoner has to act selfishly. If a subset of the group does cooperate, it doesn't matter bec…

In the prisoners’ dilemma, it is true that even a single bad actor has the ability to disrupt the rest. One limitation of that thought experiment is that it assumes defection is guaranteed to achieve a certain outcome. I think it’s reasonable to assume that any cooperative strategy has some computable “defection tolerance” and a variant on the prisoners’ dilemma that accounts for this could have very different outcom…

> One limitation of that thought experiment is that it assumes defection is guaranteed to achieve a certain outcome.

That's a premise, and it's what creates the dillema in the first place.

> Furthermore, once you start playing multiple times with the same players, you establish an algebra of trust that also drastically changes the game. The ability to influence other players between rounds also changes the game.

Actually, the wikipedia article on the dilemma discusses this. The nash equilibrium is actually for all players to defect. There is no engendered trust. [1]

[1] https://en.wikipedia.org/wiki/Prisoner%27s_dilemma#The_itera...

Re: The Dark Forest Theory of the Internet

#177

Post by ethbro »

TeMPOraL wrote: post #107
ethbro wrote: post #36

Earlier quoted context omitted.

No. There are about 100 other parameters whose relative balance determine game theoretical solutions to the problem. And as with many things, Cixin Liu handwaves away all that complexity in service of narrative. Don't confuse fiction with reality. Look at the Drake Equation as a much simpler example. Even given its simplicity, small tweaks in assumptions produce wildly different projections.

> There are about 100 other parameters whose relative balance determine game theoretical solutions to the problem. Name three. I'm by no means an expert on game theory, but accepting the two axioms that are quite explicitly stated in the story, "dark forest" seems to follow quite naturally.

1. The value of alien ideas being introduced into a culture.

2. The rarity of habital planetoids.

3. How quickly civilizations spread within their local area.

Re: The Dark Forest Theory of the Internet

#178

Post by notduncansmith »

ColanR wrote: post #176
notduncansmith wrote: post #172

Earlier quoted context omitted.

In the prisoners’ dilemma, it is true that even a single bad actor has the ability to disrupt the rest. One limitation of that thought experiment is that it assumes defection is guaranteed to achieve a certain outcome. I think it’s reasonable to assume that any cooperative strategy has some computable “defection tolerance” and a variant on the prisoners’ dilemma that accounts for this could have very different outcom…

> One limitation of that thought experiment is that it assumes defection is guaranteed to achieve a certain outcome. That's a premise, and it's what creates the dillema in the first place. > Furthermore, once you start playing multiple times with the same players, you establish an algebra of trust that also drastically changes the game. The ability to influence other players between rounds also changes the game. Actu…

> The nash equilibrium is actually for all players to defect.

Only in games where players know the number of rounds (e.g. a suicide bomber defects because they’ve decided to cap the number of rounds they’re going to play).

The same Wikipedia article describes that in studies where the number of rounds is unknown to players, superrational strategies are sustainable and that the “always defect” strategy becomes only one of several paths to Nash equilibrium.

Re: The Dark Forest Theory of the Internet

#179

Post by KirinDave »

matz1 wrote: post #152
KirinDave wrote: post #148

Earlier quoted context omitted.

I agree that there are an awful lot of people who'd like to put forth your point as a means to defend their actions. And like them, you're the one seeking to erase senses from the dictionary by pretending they aren't there to begin with.

Yes, I too agree that there are an awful lot of people to put forth your point as a means to defend their actions. And like them, you're the one who contribute the support to replace the current definition.

It's pretty amazing how you started out arguing for the dictionary definition, suggesting that's how we know common usage. But here we are with you, days later, still completely unable to incorporate that fact into your worldview. Still insisting that your desire to erase the other senses of "violence" is somehow the status quo.

The one who wants to erase bits that have been in the dictionary for decades and ignore the real world use of many senses of violence just to protect your political beliefs? In this thread: it's you

Re: The Dark Forest Theory of the Internet

#180

Post by YeGoblynQueenne »

dorgo wrote: post #171
YeGoblynQueenne wrote: post #164

Earlier quoted context omitted.

My opinion on this might be controversial but I think there's very little difference between even "hard" sci-fi and all-out fantasy (but, with spaceships). There is very little space left to write a story that is interesting and compelling if one wishes to respect the bounds of what is scientifically plausible. Then again, if one starts to stretch the definition of "scientifically plausible" there's all sorts of thin…

>There is very little space left to write a story that is interesting and compelling if one wishes to respect the bounds of what is scientifically plausible. Are we living in the same world? Quantum mechanics, relativistic effects, mathematics ( game theory anyone? ) are blowing my mind. The world is more phantastic than anything one could imagine and we know that we don't know everything. The constraint to stay in t…

But those are not science fiction. If you tell a story that is within the limits of what we know to be possible with the knowledge that we have right now, you will end up with a very boring story. If you start speculating about what _might_ be possible- you'll end up with time travel through black holes, like in Interstellar, which makes for a nice story, but not a very realistic one.

There is a trade-off between speculation and realism, that leaves a very, very narrow space for an interesting story.

Part of it is due to the fact that most of modern science has been mined mercilessly for "hard" sci-fi subjects, that have now become tropes that can't form the basis for an interesting new story anymore. For example, try to write a story where the entire premise is that someone manages to construct an Alcubierre drive. You'd get ...Star Trek: First Contact. Nice movie, hey. But nothing new, there.

Post reply on HN