No internal model of any aspect of the actual game.
— fishfry

I feel like you might have missed some important paragraphs in the article. Did you notice the heat map pictures? Did you read all the paragraphs around that? A huge part of the article is very much exploring the evidence that gpt really does model the game. — flannel jesus

I was especially impressed by the heat map data and I do believe I mentioned that in my earlier post. Indeed, I wrote:

Those heat map things are amazing. — fishfry

A little later in that same post, I wrote:

If you don't give it a mental model of the game space, it builds one of its own. — fishfry

That impressed me very much. That the programmers do not give it any knowledge of the game, and it builds a "mental" picture of the board on its own.

So I believe I already understood and articulated the point you thought I missed. I regret that I did not make my thoughts more clear.

LLMs have some capacities that "emerged" in the sense that they were acquired as a result of their training when it was not foreseen that they would acquire them. Retrospectively, it makes sense that the autoregressive transformer architecture would enable language models to acquired some of those high-level abilities since having them promotes the primary goal of the training, which was to improve their ability to predict the next token in texts from the training data. (Some of those emergent cognitive abilities are merely latent until they are being reinforced through training the base model into a chat or instruct variant). — Pierre-Normand

I just watched a bit of 3blue1brown's video on transformers. Will have to catch up on the concepts.

I confess to having my viewpoint totally turned around tonight. The chess-playing LLM has expanded my concept of what's going on in the space. I would even be willing to classify as emergence -- the exact kind of emergence I've been railing against -- the manner in which the LLM builds a mental map of the chess board, despite having no data structures or algorithms representing any aspect of the game.

Something about this has gotten my attention. Maybe I'll recover by morning. But there's something profound in the alien-ness of this particular approach to chess. A glimmer of how the machines will think in the future. Nothing like how we think.

I do not believe we should give these systems operational control of anything we care about!

One main point about describing properties or capabilities being emergent at a higher level of description is that they don't simply reduce to the functions that were implemented at the lower level of description. — Pierre-Normand

I'm perfectly willing to embrace the descriptive use of the term. I only object to it being used as a substitute for an explanation. People hear the description, and think it explains something. "Mind emerges from brain" as a conversation ender, as if no more needs to be said.

This is true regardless of there being an explanation available or not for their manifest emergence, — Pierre-Normand

Right. I just don't like to see emergence taken as an explanation, when it's actually only a description of the phenomenon of higher level behaviors not explainable by lower ones.

and it applies both to the mental abilities that human being have in relation to their brains and to the cognitive abilities that conversational AI agents have in relation to their underlying LLMs. — Pierre-Normand

Yes, and we should not strain the analogy! People love to make these mind/brain analogies with the neural nets. Brains have neurons and neural nets have weighted nodes, same difference, right? Never mind the sea of neurotransmitters in the synapses, they get abstracted away in the computer model because we don't understand them enough. Pressing that analogy too far can lead to some distorted thinking about the relation of minds to computers.

The main point is that just because conversational AI agents (or human beings) can do things that aren't easily explained as a function of what their underlying LLMs (or brains) do at a "fundamental" level of material realization, isn't a ground for denying that they are "really" doing those things. — Pierre-Normand

And not grounds for asserting it either! I'm still standing up for Team Human, even as that gets more difficult every day.

I agree that it lacks a sum, but do you think that terms like Cesàro summation and Ramanujan summation are completely misnomers? — keystone

I think they are chainsaws, not to be trifled with by the untrained masses.

Do you truly think that there's no meaningful way to assign a value of 1/2 to that divergent series? — keystone

I do. But I assume it's possible that there's some clever Ramanujan insight to get some deep mathematical context that makes the equation come out. And it might even mean something. But as a factoid to be dropped into casual mathematical conversation, no. Not any more than the execrable 1 + 2 + 3 + ... = -1/12. That's also "False as stated, and true only in rarified mathematical contexts of no relevance to a general audience."

I'm taken aback by this, — keystone

That I think there's no sensible way to assign a value to the series, regardless of whether there's some deep mathematical context? Why?

though perhaps debating Grandi's series is merely a distraction. — keystone

Yes.

I think there's a bit of confusion around what I mean by "bundle." — keystone

There are fiber bundles in math. A hairbrush with bristles sticking out is a fiber bundle. Off topic but reminded me of the name.

Let me explain using an analogy. GULP. Consider a fitness membership that includes access to cardio equipment, swimming pools, sauna rooms, group classes, and more. When you join the club, you pay a single price for this all-inclusive membership bundle. This means one price covers numerous amenities. There isn’t a separate charge for the sauna or the swimming pools. However, there should ideally be underlying individual prices, right? Like, when setting the bundle price, the gym owner should have calculated costs for each component. But what should have been done doesn't necessarily reflect what is—a single price for the entire bundle. — keystone

Ok, it's an aggregate price where the components haven't necessarily been priced. So you have aggregate lengths, but no individual ones. Something like that?

Similarly, in my scenario, the bundle of interest (a line) is represented simply as (0,2). Just as there's no itemized pricing for each gym amenity, there's no infinite set detailing every coordinate on the line. — keystone

You know, you can recover the real line just from its open intervals. Is that what you mean? In fact the rational intervals are good enough, the cover all the reals anyway.

I honestly think that what you are doing is coming to understand, in your own way, the nature of the real numbers. Which in my opinion is one of the most worthy endeavors a person can do. Clarifies so much formerly bad thinking about infinite processes.

Dedekind cuts have perfect precision. I claim that the best we can do is plan to cut an arbitrarily narrow line surrounding an irrational number. — keystone

Well sure, every irrational can be identified with a descending sequence of open intervals. I can locate pi in the sequence (3, 4), (3.1, 3.2), (3.14, 3.15), (3.141, 3.142), ...

Does that idea resonate with you?

What I mean "identified with" is this. I don't mean that my descending sequence of nested intervals "traps" the number pi or closes down on it.

No. I mean that the sequence itself IS the number pi. If pi didn't already exist, we'd just define it to be this sequence; or more precisely, the equivalence class of all such sequences. That's how mathematicians think. Pi doesn't have to be any particular thing. If you can define it as something that behaves the way pi is supposed to behave, then you might as well just consider it to be pi.

Does this make sense to you?

Which is why I say that once you grant me the rationals, you get the reals for free. If you believe in the rationals there's no point in trying to make the irrationals second class citizens in number country. Once you have the rationals the reals are already right there on equal footing. They're first class numbers.

You just have an ... ahem ... irrational prejudice against irrational numbers.

Cuts are used to decompose the bundle. Initially, the bundle price for the membership is established, and it's only afterwards that we attempt to deconstruct it into an itemized price list. Itemizing a membership can become an endless endeavor, breaking the price down into increasingly smaller fragments—from the cost of each toilet to each square of toilet paper, and even down to the cost of each atom in that toilet paper. Attempting to detail a gym membership to such minute components is a fool's errand. The same goes for breaking a line into individual points. — keystone

You do appear, to my humble and untrained eye, to be recapitulating the notion of descending down to points on the line by means of a sequence of downward nested open intervals; which is exactly the method of defining a real number as an equivalence class of Cauchy sequences. That's an alternative construction to Dedekind cuts but it gives you the same set of real numbers, since there is only one set of real numbers up to isomorphism. That's a handy thing to know. No matter how you conceive them, there is essentially only one set of real numbers.

I don't know what an "arbitrarily small cut] means. It conflicts with your previous use of cut.
— fishfry
The process of making cuts involves two distinct phases: (1) planning the cut and (2) executing the cut. — keystone

This bit about planning and execution is a little off the mark. I'm with you descending down to points via sequences of open intervals. In math when we conceive a thing it's automatically done. Would that the rest of the world were so simple!

(1) We can devise a perfect plan. During the planning phase, we don’t commit to specific values for epsilon; we only recognize that it can be arbitrarily small. This stage is the realm of mathematicians.

(2) Conversely, executing the cut requires selecting specific values for epsilon, which inevitably introduces some imprecision. Applied mathematicians handle the execution, often employing approximate values for irrationals like pi, such as 3.14. — keystone

I see this as making an infinite sequence of descending intervals as above.

pi = {(3, 4), (3.1, 3.2), (3.14, 3.15), (3.141, 3.142), ...}

Do you relate to this at all?

While this approach might seem dirty, it's also quick, and this has allowed applied mathematicians to significantly improve the world. — keystone

Does the world seem improved to you? I'm afraid that one of my bad habits is following the news, so I can't share your enthusiasm. And a lot of the trouble I see comes directly from the data mungers trying to eat us all. Hope you're glad you asked my opinion about that :-)

You’re correct that previously, I was focused on the execution, but I've realized that the planning phase is indeed more critical for this discussion. — keystone

In math you can do it all at once. You can decree that the number pi be located via a decreasing sequence of nested open sets converging on pi. And if you don't believe pi is really there, then no problem. You just define pi as the sequence of nested open intervals and you've got an object that, if it's not the "real" pi, is just as good. That's how they build the reals from the rationals.

I appreciate you taking the time to read it, and take it seriously. — flannel jesus

Beneath my skepticism of AI hype, I'm a big fan of the technology. Some of this stuff is amazing. Also frightening. Those heat map things are amazing. The way an AI trained for a specific task, maps out the space in its ... well, mind, as it were. I think reading that article convinced me that the AIs really are going to wipe out the human race. These things discern the most subtle n-th order patterns in behavior, and then act accordingly.

I am really bowled over that it can play chess and learn the rules just from auto-completing the game notation. But it makes sense ... as it trained on games it would figure out which moves are likely. It would learn to play normal chess with absolutely no programmed knowledge of the rules. Just statistical analysis of the text string completions. I think we're all doomed, don't you?

I will have to spend some more time with this article. A lot went over my head.

Ever since chat gpt gained huge popularity a year or two ago with 3.5, there have been people saying LLMs are "just this" or "just that", and I think most of those takes miss the mark a little bit. "It's just statistics" it "it's just compression". — flannel jesus

I was one of those five minutes ago. Am I overreacting to this article? I feel like it's turned my viewpoint around. The chess AI gained understanding it its own very strange way. I can see how people would say that it did something emergent, in the sense that we didn't previously know that an LLM could play chess. We thought that to program a computer to play chess, we had to give it an 8 by 8 array, and tell it what pieces are on each square, and all of that.

But it turns out that none of that is necessary! It doesn't have to know a thing about chess. If you don't give it a mental model of the game space, it builds one of its own. And all it needs to know is what strings statistically follow what other strings in a 5 million game dataset.

It makes me wonder what else LLMs can do. This article has softened my skepticism. I wonder what other aspects of life come down, in the end, to statistical pattern completion. Maybe the LLMs will achieve sentience after all. This one developed a higher level of understanding than it was programmed for, if you look at it that way.

Perhaps learning itself has a lot in common with compression - and it apparently turns out the best way to "compress" the knowledge of how to calculate the next string of a chess game is too actually understand chess! And that kinda makes sense, doesn't it? To guess the next move, it's more efficient to actually understand chess than to just memorize strings. — flannel jesus

It seems that in this instance, there's no need to understand the game at all. Just output the most likely string completion. Just as in the early decades of computer chess, brute force beat systems that tried to impart understanding of the game.

It seems that computers "think" very differently than we humans. In a sense, an LLM playing chess is to a traditional chess engine, as the modern engines are to humans. Another level of how computers play chess.

This story has definitely reframed my understanding of LLMs. And suddenly I'm an AI pessimist. They think so differently than we do, and they see very deep patterns. We are doomed.

And one important extra data point from that write up is the bits about unique games. Games become unique, on average, about 10 moves in, and even when a game is entirely unique and wasn't in chat gpts training set, it STILL calculates legal and reasonable moves. I think that speaks volumes. — flannel jesus

That's uncanny for sure. I really feel a disturbance in the force of my AI skepticism. Something about this datapoint. An LLM can play chess just by training on game scores. No internal model of any aspect of the actual game. That is so weird.

I still can't find it. I copied the quoted passage into my message, but not the commentary. Which is a pity. — Ludwig V

Lost in the ether, forever.

Can you clarify which sense you mean?
— fishfry

Metaphysical impossibility. Supertasks cannot be performed in any possible world. P3 is a tautology, P2 follows from P3, and so C1 is necessarily true. — Michael

Ok. Possible worlds. I actually took a class where we talked about that, but I have a hard time understanding the concept. There are people who think possible worlds are real. I'm not one of them. And the whole metaphor is lost on me.

But there's a problem. You are assuming that which you wish to prove. You say, "Supertasks cannot be performed in any possible world." But that's the thing you're trying to convince me of. Nevermind that I don't necessarily believe in supertasks myself, but regardless, you are trying to convince me that there are no supertasks. So you can't just state that there are no supertasks. Right? Just a basic point of logic.

And even then I reject the claim on its own merits. I could argue (not that I do, but that I could -- hope that's clear) that if time is modeled by the real numbers (agreed, that is a dubious assumption) then I perform a supertask every time I get up to go to the kitchen for a snack. I named my refrigerator Zeno.

If I can make that argument, then there is some possible world in which it's true. Namely, in any possible world in which time is accurately modeled by the standard real numbers, supertasks are commonplace, every day occurrences.

I argue -- in fact I believe I've made this argument several times already -- that because supertasks are abstractly conceivable, they are NOT metaphysically impossible.

A



In short, (1) You assumed what you're trying to prove; and (2) Your assumption is wrong. Supertasks are conceivable and as far as anyone knows, even physically possible. It's for the physics of the future to know.

Also, it's conceivable that future physics will incorporate physically instantiable infinities, and that supertasks will be possible. I just don't see how you can ignore that possibility.

Here are three distinct propositions:

a) 1/2 + 1/4 + 1/8 + ... = 1
b) there is a bijection between this geometric series and the natural numbers
c) it is metaphysically possible to recite the natural numbers at successively halved intervals of time

(a) and (b) are true and (c) is false. — Michael

(a) and (b) are mathematical truths we all agree to, I hope. But "metaphysically impossible," I don't even know what that means.

If I make the assumption of successive halving, I can recite all the numbers in finite time. I agree this is a purely fictitious, hypothetical game.

We agree that it is physically impossible, relative to currently known physics.

I simply do not understand why you jump to saying that means it's metaphysically impossible.

By analogy with previous scientific revolutions, we must allow for the possibility that actualized infinities could become part of physics; and that physical supertasks might be witnessed in the Superconducting Supertask Collider of the future. This is a grant proposal waiting to happen.

I just don't see how you can call this metaphysically impossible. So was heavier-than-air controlled flight, once.

Your argument rests on the assumption that (c) follows from (a) and (b), but it doesn't. — Michael

I don't even know what you are talking about. No argument I make assumes a and b and concludes c. (a) and (b) are trivialities. (a) is freshman calculus and (b) is the very definition of an infinite sequence. Together they don't imply anything. You're imagining some argument I didn't make and wouldn't make.

(c) is proven false by P3, as well as arguments like Thomson's lamp. — Michael

I wish we were having the same conversation. "My argument" assumes (a) and (b) and concludes (c)? I can't even relate that to anything I can think of. What do (a) and (b) have to do with what's metaphysically possible? And what does metaphysically possible even mean? I should have asked that earlier. I don't know what the phrase means to you. And it's not one I use in my own mind, myself. I think just about anything is metaphysically possible, if I were pressed to give an opinion on the subject. If not pressed, the thought of metaphysical possibility never enters my mind. So probably I have no idea what you mean.

You can continually assert that (a) and (b) are true, and I will continually agree, but until you can present actual evidence or reasoning to support (c), I will always reject it as per the above. — Michael

I have surely presented my reasoning, which I will repeat here.

P1) It is metaphysically possible that time is accurately modeled by the standard real numbers;

P2) 1/2 + 1/4 + 1/8 + ... = 1 in the real numbers;

C1) A Zeno supertask is possible, by moving in any direction for one second, and thereby traversing a countably infinite sequence of finite intervals of time [0, 1/2], [1/2, 3/4], [3/4, 7/8], ...

C2) Therefore supertasks are metaphysically conceivable.

Secondly:

P3) It is conceivable that future physics will allow for supertask; and therefore

C2) Supertasks are metaphysically conceivable.

You could probably help me out by clearly defining metaphysically impossible.

And please reframe your argument about my claiming (a) and (b) imply (c), when in fact I make no such argument.

Please allow me to refine and restate my position on reals.

Grandi's series has no sum but it should be 1/2. — keystone

Why on earth would you think that? It clearly has no sum, since the sequence of partial sums has no limit.

Analogously, I believe a line is not made of points but it should be made of 2ℵ0
2
ℵ
0
points. — keystone

It's not made of points but it's made of points? How am I supposed to understand that?

Analogously, I believe a line is not modeled by numbers but it should be modeled by the real numbers. — keystone

It's not but it is?

Just as Grandi's series only sums to 1/2 in a very particular light, — keystone

I can't imagine what that light is. The Wiki page is misleading on that point.

my view amounts to the belief that there is great mathematical value in irrationals, but that they only make sense in a very particular light - when considered collectively as bundles, rather than individual, isolated points. — keystone

Needs explanation.

This is the essence of the top-down view where we start with such a bundle of 2^aleph_0 points - a line in this case - and then we make cuts to selectively isolate segments of this line. I refer to any point nested within such a bundle, as opposed to being isolated, as a potential point. — keystone

Cuts as in Dedekind cuts? If you already have continuum-many points, why do you need cuts?

Revisiting the analogy above, when I utilize an interval to describe a range, I am referring to the underlying and singular continuous line between the endpoints, which should correspond to the set of real numbered points contained within these endpoints. — keystone

But that is exactly the standard view.

I believe performing an arbitrarily small 1D cut around φ — keystone

I don't know what an "arbitrarily small cut] means. It conflicts with your previous use of cut.

What do you think? — keystone

There are no arbitrarily small real numbers.

The golden ratio?

This isn't working for me.

↪Michael
OK. You and fishfry both believe that the supertask is impossible. But you believe that is because it is contradictory and fishfry believes that it is because the last step is not defined. Am I right about that? — Ludwig V

I have not said that. I have said that I have no strong opinion about supertasks and am entirely comfortable arguing either side.

I don't think this is a take that's likely correct. This super interesting writeup on an LLM learning to model and understand and play chess convinces me of the exact opposite of what you've said here:

https://www.lesswrong.com/posts/yzGDwpRBx6TEcdeA5/a-chess-gpt-linear-emergent-world-representation — flannel jesus

Less Wrong? Uh oh that's already a bad sign. I'll read it though. I do allow for the possibility that I could be wrong. I just finished replying to a lengthy screed from @Christopher so I'm willing to believe the worst about myself at this point. I'm neither exponential nor emergent nor multimodal so what the hell do I know. The Less Wrong crowd, that's too much Spock and not enough Kirk. Thanks for the link. I'm a little loopy at the moment from responding to Christopher.

ps -- Clicked the link. "This model is only trained to predict the next character in PGN strings (1.e4 e5 2.Nf3 ...) and is never explicitly given the state of the board or the rules of chess. Despite this, in order to better predict the next character, it learns to compute the state of the board at any point of the game, and learns a diverse set of rules, including check, checkmate, castling, en passant, promotion, pinned pieces, etc. "

I stand astonished. That's really amazing.

This is simply wrong. — Christoffer

I take emergence to be a synonym for, "We have no idea what's happening, but emergence is a cool word that obscures this fact."

These are examples of what I'm talking about:

https://hai.stanford.edu/news/examining-emergent-abilities-large-language-models
https://ar5iv.labs.arxiv.org/html/2206.07682
https://www.jasonwei.net/blog/emergence
https://www.assemblyai.com/blog/emergent-abilities-of-large-language-models/ — Christoffer

I'm sure these links exhibit educated and credentialed people using the word emergence to obscure the fact that they have no idea what they're talking about.

It's true that a big pile of flipping bits somehow implements a web browser or a chess program or a word processor or an LLM. But calling that emergence, as if that explains anything at all, is a cheat.

Emergence does not equal AGI or self-awareness, but they mimmick what many neuroscience papers are focused on in regards to how our brain manifest abilities out of increasing complexity. And we don't yet know how combined models will function. — Christoffer

"Mind emerges from the brain} explains nothing, provides no insight. It sounds superficially clever, but if you replace it with, "We have no idea how mind emerges from the brain," it becomes accurate and much, much more clear.

No one is claiming this. But equally, the problem is, how do you demonstrate it? Effectively the Chinese room problem. — Christoffer

Nobody knows how to demonstrate self-awareness of others. We agree on that. But calling it emergence is no help at all. It's harmful, because it gives the illusion of insight without providing insight.

There's no emergence in chatbots and there's no emergence in LLMs. Neural nets in general can never get us to AGI because they only look backward at their training data. They can tell you what's happened, but they can never tell you what's happening.
— fishfry

The current predictive skills are extremely limited and far from human abilities, but they're still showing up, prompting a foundation for further research. — Christoffer

I have no doubt that grants will be granted. That does not bear on what I said. Neural nets are a dead end for achieving AGI. That's what I said. The fact that everyone is out there building ever larger wings out of feathers and wax does not negate the point.

If you climb a tree, you are closer to the sky than you were before. But you can't reach the moon that way. That would be my point. No matter how much clever research is done.

A new idea is needed.

But no one has said that the current LLMs in of themselves will be able to reach AGI. Not sure why you strawman in such conclusions? — Christoffer

Plenty of people are saying that. I read the hype. If you did not say that, my apologies. But many people do think LLMs are a path to AGI.

Why does conventional hardware matter when it's the pathways in the network that is responsible for the computation? — Christoffer

I was arguing against something that's commonly said, that neural nets are complicated and mysterious and their programmers can't understand what they are doing. That is already true of most large commercial software systems. Neural nets are conventional programs. I used the example of political bias to show that their programmers understand them perfectly well, and can tune them in accordance with management's desires.

The difference here is basically that standard operation is binary in pursuit of accuracy, but these models operate on predictions, closer to how physical systems do, which means you increase the computational power with a slight loss of accuracy. That they operate on classical software underneath does not change the fact that they operate differently as a whole system. Otherwise, why would these models vastly outperform standard computation for protein folding predictions? — Christoffer

They're a very clever way to do data mining. I didn't say I wasn't impressed with their achievements. Only that (1) they are not the way to AGI or sentience; and (2) despite the mysterianism, they are conventional programs that could, in principle, be executed with pencil and paper, and that operate according to the standard rules of physical computation that were developed in the 1940s.

By mysterianism, I mean claims such as you just articulated: "they operate differently as a whole system ..." That means nothing. The chess program and the web browser on my computer operate differently too, but they are both conventional programs that ultimately do nothing more than flip bits.

I do oppose this mysterianistic attitude on the part of many neural net proponents. It clouds people's judgment. How did black George Washington show up on Google's AI? Not because it "operates different as a whole system." Rather, it's because management told the programmers to tune it that way, and they did.

Neural nets are deterministic programs operating via principles that were well understood 70 years ago.

Stop the neural net mysterianism! That's my motto for today.

they operate differently as a whole system
Yes, and why would a system that is specifically very good at handling extreme complexities, not begin to mimic complexities in the physical world?[/quote]

When did I ever claim that large, complex programs aren't good at mimicking the physical world? On the contrary, they're fantastic at it.

I don't mean to downplay the achievements of neural nets. Just want to try to get people to dial back the hype ("AGI is just around the corner") and the mysterianism ("they're black boxes and even their programmers can't understand them.")

https://www.mdpi.com/1099-4300/26/2/108
https://ar5iv.labs.arxiv.org/html/2205.11595 — Christoffer

Jeez man more emergence articles? Do you think I haven't been reading this sh*t for years?

Emergence means, "We don't understand what's going on, but emergence is a cool word that will foll people." And it does.

Seen as the current research in neuroscience points to emergence in complexities being partly responsible for much of how the brain operates, why wouldn't a complex computer system that simulate similar operation not form emergent phenomenas? — Christoffer

Emergence emergence emergence emergence emergence. Which means, you don't know. That's what the word means.

You claim that "emergence in complexities being partly responsible for much of how the brain operates" explains consciousness? Or what are you claiming, exactly? Save that kind of silly rhetoric for your next grant application. If it were me, I'd tell you to stop obfuscating. "emergence in complexities being partly responsible for much of how the brain operates". Means nothing. Means WE DON'T KNOW how the brain operates.

There's a huge difference between saying that "it forms intelligence and consciousness" and saying that "it generates emergent behaviors". There's no claim that any of these LLMs are conscious, that's not what this is about. And AGI does not mean conscious or intelligent either, only exponentially complex in behavior, which can form further emergent phenomenas that we haven't seen yet. I'm not sure why you confuse that with actual qualia? The only claim is that we don't know where increased complexity and multimodal versions will further lead emergent behaviors. — Christoffer

You speak in buzz phrases. It's not only emergent, it's exponential. Remember I'm a math guy. I know what the word exponential means. Like they say these days: "That word does not mean what you think it means."

So there's emergence, and then there's exponential, which means that it "can form further emergent phenomenas that we haven't seen yet."

You are speaking in entirely meaningless babble at this point. I don't mean that you're not educated. I mean that you have gotten lost in your own jargon. You have said nothing at all in this post.

This is just a false binary fallacy and also not correct. The programmable behavior is partly weights and biases within the training, but those are extremely basic and most specifics occur in operational filters before the output. If you prompt it for something, then there can be pages of instructions that it goes through in order to behave in a certain way. — Christoffer

Yes, that's how computers work. When I click on Amazon, whole pages of instructions get executed before the package arrives at my door. What point are you making?

In ChatGPT, you can even put in custom instructions that function as a pre-instruction that's always handled before the actual prompt, on top of what's already in hidden general functions. — Christoffer

You're agreeing with my point. Far from being black boxes, these programs are subject to the commands of programmers, who are subject to the whims of management.

That doesn't mean the black box is open. There's still a "black box" for the trained model in which it's impossible to peer into how it works as a neural system. — Christoffer

You say that, and I call it neural net mysterianism. You could take that black box, print out its source code, and execute it with pencil and paper. It's an entirely conventional computer program operating on principles well understood since the first electronic digital computers in the 1940s.

"Impossible to peer into." I call that bullpucky. Intimidation by obsurantism.

Every line of code was designed and written by programmers who entirely understood what they were doing.

And every highly complex program exhibits behaviors that surprise their coders. But you can tear it down and figure out what happened. That's what they do at the AI companies all day long. They do not go, "Oh, this black box is inscrutable, incomprehensible. We better just pray to the silicon god."

It doesn't work that way.

This further just illustrates the misunderstandings about the technology. Making conjectures about the entire system and the technology based on these company's bad handling of alignment does not reduce the complexity of the system itself or prove that it's "not a black box". It only proves that the practical application has problems, especially in the commercial realm. — Christoffer

You say it's a black box, and I point out that it does exactly what management tells the programmers to make it do, and you say "No, there's a secret INNER" black box."

I am not buying it. Not because I don't know that large, complex software systems don't often exhibit surprising behavior. But because I don't impute mystical incomprehensibility to computer programs.

Maybe read the entire argument first and sense the nuances. You're handling all of this as a binary agree or don't discussion, which I find a bit surface level. — Christoffer

Can we stipulate that you think I'm surface level, and I think you're so deep into hype, buzzwords, and black box mysterianism that you can't see straight?

That will save us both a lot of time.

I can't sense nuances. They're a black box. In fact they're an inner black box. An emergent, exponentail black box.

I know you take your ideas very seriously. That's why I'm pushing back. "Exponential emergence" is not a phrase that refers to anything at all.

Check the publications I linked to above. — Christoffer

I'll stipulate that intelligent and highly educated and credentialed people wrote things that I think are bullsh*t.

Do you understand what I mean by emergence? What it means in research of complex systems and chaos studies, especially related to neuroscience. — Christoffer

Yes. It means "We don't understand but if we say that we won't get our grant renewed, so let's call it emergence. Hell, let's call it exponential emergence, then we'll get a bigger grant."

Can't we at this point recognize each other's positions? You're not going to get me to agree with you if you just say emergence one more time.

Believe they start spouting racist gibberish to each other. I do assume you follow the AI news.
— fishfry

That's not what I'm talking about. I'm talking about multimodality. — Christoffer

Exponential emergent multimodality of the inner black box.

Do you have the slightest self-awareness that you are spouting meaningless buzzwords at this point?

Do you know what multimodal freight is? It's a technical term in the shipping industry that means trains, trucks, airplanes, and ships.

It's not deep.

Most "news" about AI is garbage on both sides. We either have the cryptobro-type dudes thinking we'll have a machine god a month from now, or the luddites on the other side who don't know anything about the technology but sure likes to cherry-pick the negatives and conclude the tech to be trash based on mostly just their negative feelings. — Christoffer

And then there are the over-educated buzzword spouters. Emergence. Exponential. It's a black box. But no it's not really a black box, but it's an inner black box. And it's multimodal. Here, have some academic links.

This is going nowhere.

I'm not interested in such surface level discussion about the technology. — Christoffer

Surface level is all you've got. Academic buzzwords. I am not the grant approval committee. Your jargon is wasted on me.

If you want to read more about emergence — Christoffer

Oh man you are killin' me.

Is there anything I've written that leads you to think that I want to read more about emergence?

in terms of the mind you can find my other posts around the forum about that. — Christoffer

Forgive me, I will probably not do that. But I don't want you to think I haven't read these arguments over the years. I have, and I find them wanting.

Emergent behaviors has its roots in neuroscience and the work on consciousness and the mind. — Christoffer

My point exactly. In this context, emergence means "We don't effing know." That's all it means.

And since machine learning to form neural patterns is inspired by neuroscience and the way neurons work, there's a rational deduction to be found in how emergent behaviors, even rudimentary ones that we see in these current AI models, are part of the formation of actual intelligence. — Christoffer

I was reading about the McCulloch-Pitts neuron while you were still working on your first buzzwords.

This, when combined with evidence that the brain may be critical, suggests that ‘consciousness’ may simply arise out of the tendency of the brain to self-organize towards criticality. — Christoffer

You write, "may simply arise out of the tendency of the brain to self-organize towards criticality" as iff you think that means anything.

The problem with your reasoning is that you use the lack of a final proven theory of the mind as proof against the most contemporary field of study in research about the mind and consciousness. — Christoffer

I'm expressing the opinion that neural nets are not, in the end, going to get us to AGI or a theory of mind.

I have no objection to neuroscience research. Just the hype, buzzwords, and exponentially emergent multimodal nonsense that often accompanies it.

It's still making more progress than any previous theories of the mind and connects to a universality about physical processes. Processes that are partly simulated within these machine learning systems. And further, the problem is that your reasoning is just binary; it's either intelligent with qualia, or it's just a stupid machine. That's not how these things work. — Christoffer

I have to apologize to you for making you think you need to expend so much energy on me. I'm a lost cause. It must be frustrating to you. I'm only expressing my opinions, which for what it's worth have been formed by several decades of casual awareness of the AI hype wars, the development of neural nets, and progress in neuroscience.

It would be easier for you to just write me off as a lost cause. I don't mean to bait you. It's just that when you try to convince me with meaningless jargon, you weaken your own case.

I would not dispute that. I would only reiterate the single short sentence that I wrote that you seem to take great exception too. Someone said AGI is imminent, and I said, "I'll take the other side of that bet." And I will.
— fishfry

I'm not saying AGI is imminent, but I wouldn't take the other side of the bet either. You have to be dead sure about a theory of the mind or theories of emergence to be able to claim either way, and since you don't seem to aspire to any theory of emergence, then what's the theory that you use as a premiss for concluding it "not possible"? — Christoffer

I wrote, "I'll take the other side of that bet," and that apparently pushed your buttons hard. I did not mean to incite you so, and I apologize for any of my worse excesses of snarkiness in this post.

But exponential emergence and multimodality, as substitutes for clear thinking -- You are the one stuck with this nonsense in your mind. You give the impression that perhaps you are involved with some of these fields professionally. If so, I can only urge to you get some clarity in your thinking. Stop using buzzwords and try to think clearly. Emergence does not explain anything. On the contrary, it's an admission that we don't understand something. Start there.

In my opinion, that is false. The reason is that neural nets look backward. You train them on a corpus of data, and that's all they know.
— fishfry

How is that different from a human mind? — Christoffer

Ah. The first good question you've posed to me. Note how jargon-free it was.

I don't know for sure. Nobody knows. But one statement I've made is that neural nets only know what's happened. Human minds are able to see what's happening. Humans can figure out what to do in entirely novel situations outside our training data.

But I can't give you proof. If tomorrow morning someone proves that humans are neural nets, or neural nets are conscious, I'll come back here and retract every word I've written. I don't happen to think there's much chance of that happening.

The only technical difference between a human brain and these systems in this context is that the AI systems are trained and locked into an unchanging neural map. The brain, however, is constantly shifting and training while operating. — Christoffer

Interesting idea. Do they have neural nets that do that? My understanding is that they train the net, and after that, the execution is deterministic. Perhaps you have a good research idea there. Nobody knows what the secret sauce of human minds is.

If a system is created that can, in real time, train on a constant flow of audiovisual and data information inputs, which in turn constantly reshape its neural map. What would be the technical difference? The research on this is going on right now. — Christoffer

Now THAT, I'd appreciate some links for. No more emergence please. But a neural net that updates its node weights in real time is an interesting idea.

They can't reason their way through a situation they haven't been trained on.
— fishfry

The same goes for humans. — Christoffer

How can you say that? Reasoning our way through novel situations and environments is exactly what humans do.

That's the trouble with the machine intelligence folks. Rather than uplift their machines, they need to downgrade humans. It's not that programs can't be human, it's that humans are computer programs.

How can you, a human with life experiences, claim that people don't reason their way through novel situations all the time?

since someone chooses what data to train them on
— fishfry

They're not picking and choosing data, they try to maximize the amount of data as more data means far better accuracy, just like any other probability system in math and physics. — Christoffer

Humans are not "probability systems in math or physics."

Neural nets will never produce AGI.
— fishfry

Based on what? Do you know something about multimodal systems that others don't? Do you have some publication that proves this impossibility? — Christoffer

Credentialism? That's your last and best argument? I could point at you and disprove credentialism based on the lack of clarity in your own thinking.

Again, how does a brain work? Is it using anything other than a rear view mirror for knowledge and past experiences? — Christoffer

Yes, but apparently you can't see that.

As far as I can see the most glaring difference is the real time re-structuring of the neural paths and multimodal behavior of our separate brain functions working together. No current AI system, at this time, operates based on those expanded parameters, which means that any positive or negative conclusion for that require further progress and development of these models. — Christoffer

I'm not the grant committee. But I am not opposed to scientific research. Only hype, mysterianism, and buzzwords as a substitute for clarity.

Bloggers usually don't know shit and they do not operate through any journalistic praxis. While the promoters and skeptics are just driving up the attention market through the shallow twitter brawls that pops up due to a trending topic. — Christoffer

Is that the standard? The ones I read do. Eric Hoel and Gary Marcus come to mind, also Michael Harris. They don't know shit? You sure about that? Why so dismissive? Why so crabby about all this? All I said was, "I'll take the other side of that bet." When you're at the racetrack you don't pick arguments with the people who bet differently than you, do you?

Are you seriously saying that this is the research basis for your conclusions and claims on a philosophy forum? :shade: — Christoffer

You're right, I lack exponential emergent multimodality.

I've spent several decades observing the field of AI and I have academic and professional experience in adjacent fields. What is, this credential day? What is your deal?

Maybe stop listening to bloggers and people on the attention market? — Christoffer

You've convinced me to stop listening to you.

I rather you bring me some actual scientific foundation for your next premises to your conclusions. — Christoffer

It's been nice miscommunicating with you. I'm sure you must feel the same about me.

tl;dr: Someone said AGI is imminent. I said I'd gladly take the other side of that bet. I reiterate that. Also, when it comes to AGI and a theory of mind, neural nets are like climbing a tree to reach the moon. You apparently seem to be getting closer, but it's a dead end. And, the most important point: buzzwords are a sign of fuzzy thinking.

I appreciate the chat, I will say that you did not move my position.

I see no contradiction in Thompson's lamp, only a failure to define the terminal state.

PS Since I started writing this, the link to the post that I copied this quotation from seems to have become non-functional. Very odd. — Ludwig V

Was this from you to me? That post of @Michael disappeared for me as well.

P1. If we can recite the natural numbers at successively halved intervals of time then we can recite every natural number in finite time
P2. We cannot recite every natural number in finite time
C1. Therefore, we cannot recite the natural numbers at successively halved intervals of time — Michael

I believe you're equivocating in the sense that you don't distinguish between the successively halved time intervals as:

a) An abstract, hypothetical, mathematical exercise; and

b) A physical situation that is incompatible with known physics and that may or may not be incompatible with the true nature of the world.

Can you clarify which sense you mean?

If you mean (a), then of course we CAN recite every natural number, so P2 is wrong; if (b), we can't currently and likely will never be able to.

When a mathematician says that 1/2 + 1/4 + 1/8 + ... = 1, they don't mean that you can perform this calculation with pencil and paper before lunchtime. They mean that the two expressions on either side of the equal sign denote the same real number.

I'm talking about an Ai that passes all the time, even against people who know how to trip up Ai's. We don't have anything like that yet. — RogueAI

Agreed. My point is that the humans are the weak link.

Another interesting point is deception. For Turing, the ability to fool people about one's true nature is the defining attribute of intelligence. That tells us more about Turing, a closeted gay man in 40's-50's England, than it does about machine intelligence.

What if we had a true AGI that happened to be honest? "Are you human?" "No, I'm an AI running such and so software on such and so hardware." It could never pass the test even it were self-aware.

Yes, I find that as well. I work round it by selecting only the quoted text, not including the link that gives the attribution. Then, you can hit "quote" and the system does pick up the attribution. Then, if you separately select the response, it is copied and attributed in the normal way. — Ludwig V

Thanks, guess I can no longer just copy/quote the whole response any more. Have to do it one paragraph at a time.

Neither am I, on reflection. I was trying to articulate the point that one can count forward, but not backward, so I don't think anything is at stake. — Ludwig V

Nothing is at stake for me or you, but our friend @Michael seems to regard it as important. It's obvilus that you can't count the natural numbers backwards.

"If you know what you're doing you're not learning anything." Think I read that somewhere.
— fishfry
Yes, I like that. I'm a bit of a contrarian, so I'm tempted to reply that I don't need my surgeon to learn anything while he's cutting me open. Indeed, I would be rather concerned if I thought he was. It applies better to artistic, experimental, open-ended activities - like philosophy and maybe mathematics, at least sometimes. — Ludwig V

Agreed! That's why job applicants should not say they're looking for a challenging position. Employers want someone who can do the job!

If you don't understand what realism vs anti-realism means, you have understood correctly - as I see it. — Ludwig V

LOL. I feel better then.

Some people would argue that the proposition that "2+2 = 4" does indeed only have a truth-value only when someone passes judgement on it but that 2+2 = 4 independently of anyone doing that i.e. is objectively true. There's a temptation to think that mathematical truth is eternal, i.e. always has been true, always will be true, whatever happens. But that's a mistake. It makes no sense to assign a place in the time series to 2+2 = 4; there is no meaningful way of doing that. (Grammarians recognize a tense that is called the timeless present which is exemplified in propositions like this.) — Ludwig V

We must be in agreement on that then.

Ok. Don't think I disagreed with anything you said.
— fishfry
I'm glad it made sense. — Ludwig V

Ok!!

If I try to simulate our actual world, I must approximate it since perfect simulation is impossible, requiring, among other things, infinite precision variables. So Lara Croft has, among other traits, square legs. All very crude. It gets better in later years, but still an approximation of what it wants to be. — noAxioms

This is very different than simulation theory as I understand it. Sim theory doesn't say the simulator approximates our world. Sim theory says the simulator creates or instantiates our world. Exactly.

If the simulator is only approximating our world, then what is the real world?

If I simulate Conway's game of life (not our actual world, but one with very simple rules), well, it necessarily would have bounds, but otherwise would not be done as an approximation. — noAxioms

Do you think (or does anyone think) the dots in Life have an interior monologue?

No, operates under the laws of computation as they (in the far future) understand them. Not under our current understandi — noAxioms

Bostrom asks, "Are you a COMPUTER simulation?" (my emphasis)

If he meant computer as understood by some future society but not by us, he would have said that. He didn't. Did he?

Agree with that. Hence my aversion to magic. — noAxioms

You agree with me on this point then, am I correct?

I agree with this. I'm certainly not promoting sim theory. — noAxioms

Ok. Can you remind me of what we're agreeing or disagreeing on then?

My only concerns with what you've said so far are:

1) That simulation theory claims the simulator approximates some deeper reality; and

2) That Ms. Pac-Man is an extension of my mind and can be said to have an inner life, namely mine.

I hope that we are agreed that a simulation of gravity or a simulation of the stock market is not the same use of the word as the GS simulating my mind for me. The latter is a complete instantiation, not just an approximation.
Both will always be an approximation. Any simulation of something 'real' must be. The physics of the simulation will be different than the physics of the GS. If the two are close enough, then the simulation can achieve its goals. Hence weather forecasting not being a total waste of time. — noAxioms

You are misconstruing what Bostrom and other simulationists believe, then. They're not saying we're an approximation. They're saying our exact reality is being instantiated by a computer.

That's fine, but none of those has actually produced a real game before it was played. Sure, it can be used to set odds. Sure, it gets the final score right sometimes, but never the way the score gets that way. Of course the stock market is similar, but one can simulate the effect of certain news on the market. It can simulate a panic, and help test methods to control such instabilities. — noAxioms

Right. That's what simulation is. Which is why simulation theory is the wrong word. It should be instantiation.

Don't need a theory. Just a simulator. If it works, I don't have to know how it works. If it can't work, then it wont. — noAxioms

The fact that you can't explain why you have a mind is not evidence for a simulator. This point I don't follow.

"of ancestral history". His words. A fictional one at best, just like the football simulator. It's not going to show any historical events we know unless you start the sim just before they happen. If they start the simulation far enough back, there won't even be humans in it, ever. — noAxioms

Ok. But why would they do that? We don't do ancestral simulations, though we have the computing power to make a first cut at such an enterprise.

Yea, I saw those links. I didn't watch the talk, because I don't get my physics from there.
I still have no idea what Smoot is proposing. — noAxioms

Smoot believes we're a computer simulation. A depressing claim for such an accomplished scientist to believe.

What's wrong with Youtube? Sean Carrol is on there. Ed Witten gets interviewed there. The famous Sabine Hossenfelder. Many others. There's a guy named Eigenchris who does lectures on the math of general relativity. Youtube is a fantastic resource. A person does not suddenly become disreputable by virtue of being filmed.

Bostrom thinks mind is computational. I see few detractors that claim that it cannot be, and thus he must be wrong. — noAxioms

Lack of detractors is not evidence for a theory.

I'm not sure if LARPing qualifies as a simulation. They all know it's an act. Nobody really wants to kill the opposing side.
It happens a lot by me since I'm in a USA town that regularly holds a celebration of the British destruction of the place. The LARP types (reenactors) love it because the red-coats hardly ever get to be the guys that win. The blue guys fire back, but lose, but in reality there was no resistance. Everybody skedaddled and the place was burned down. — noAxioms

I mentioned civil war reenactors because that's the only type of ancestor simulation we do. Why does Bostrom think our simulators would have an interest in ancestor simulation? More likely they'd be trying to crack this weekend's football games!

So the consciousness is a separate thing, not just a different process of the body that utilizes different noncomputational physics. If the latter were true, then the body would be liking the ice cream, just via a noncomputational mechanism. — noAxioms

I take your point. My body processes the ice cream. Chemicals in my limbic system produce a feeling of enjoyment. My consciousness is the thing that has the experience, and science has absolutely no explanation for that.

Even if pleasure is a chemical response in the brain, my experience is the pleasure. The chemicals in the brain don't have experiences, I do.
Funny, but I totally agree with that wording.

Only because you choose not to consider them to be part of you, just like when you say "Also my body". That's a choice to include that. — noAxioms

My mind is my experience of my body.

I would say that a thing with no understanding of chess would not be able to win the game. — noAxioms

My computer regularly beats me at chess, and it's just flipping bits as per its program. We'll have to agree to disagree if you believe otherwise.

It's no different than the moisture sensor in your clothes dryer. It does not "know" or "care" about your clothes, it's just a mechanical mechanism.

Again, the different in our views seems to be a language one. Two systems (black boxes) are doing the same thing, but the word 'understands' only applies if it's done the magic way and not the computational way. I take a more pragmatic definition: If it wins or even plays a plausible game, the word 'understands' is functionally applicable. — noAxioms

I take understanding to be a subjective mental mental state. We must agree to disagree on this. Do you think your dryer understands what it means for your clothes to be wet?

It would probably slaughter any human at Jeopardy or some other typical trivial game. But I agree, the word 'understands' is pretty inapplicable to the LLM. — noAxioms

Wait, you just got through emphasizing that functional behavior is understanding. If an LLM passes the bar exam, by your definition it understands the law. But now you are going back on that.

Am I correct that you contradicted yourself?

If you mean that a brain isn't implemented as a Turing machine, I agree, but neither is any computer anywhere. The circuits don't work that way. — noAxioms

A TM can emulate any physical computer in existence. It doesn't matter whether the circuits work that way. In theory, they could.

Also, a brain is just part of a person-system just like a CPU is part of a self-driving car.. A person is conscious, not a brain, — noAxioms

Brains are not conscious. Minds are. What the relation is, we don't know.

A person is neither. It can in theory be simulated by either of those, but it wouldn't be done by modeling the person as either of those. A person is no more a Turing machine than is the weather. — noAxioms

If you agree with me that a TM is not a person, why are we having this conversation? I have a feeling there's a simulationist out there laughing at this foolish conversation. What are we discussing?

A digital computer is a Von Neumann machine, and a person isn't one of those either. There are digital circuits involved however. Wires, on/off states, etc. No clock. No bus. No instructions. — noAxioms

A TM could emulate any physical computer in existence. In fact a TM could emulate any type of computation there is, according to the Church-Turing thesis.

Throughout our conversation, my perspective and how I express it have greatly developed, leading me to believe it's best to reformulate and clarify my position. I'll be on a short holiday for the next few days, and I'd also like to take the necessary time to gather my thoughts before responding. For now, let me make two points:

The essence of my perspective (top-down) remains the same, although it requires some minor adjustments.
Having to reformulate my view underscores the significant value I've derived from our conversation—thanks once more!

I'll reach out again in a few days. I look forward to continuing this discussion. Enjoy your weekend! — keystone

No prob, I'll be here.

Just remember: Once you stipulate to the rationals, you get the reals for free. I thought at one point that those were your fictional numbers, but then you said they weren't.

It <is> a truth predicate that would work because Truthbearer(L,x) ≡ (True(L,x) ∨ True(L,~x)) screens out epistemological antinomies that Tarski get stuck on.
— PL Olcott

@jgill@fishfry — Lionino

I'm familiar with Pete's work from other forums.

↪fishfry Don't you think we're pretty close to having something pass the Turing Test? — RogueAI

The Turing test was passed a number of years ago by a chatbot named Eugene Goostman.

The problem with the Turing test is that the humans are not sufficiently suspicious. When Joseph Weizenbaum invented the first chatbot, ELIZA, he did it to show that computers that emulate people aren't really intelligent.

But he was shocked to find that the department secretaries were telling it their innermost feelings.

Humans are the weak link in the Turing test. It's even worse now that the general public has been introduced to LLMs. People are all too willing to impute intelligence to chatbots.

Oh ok these definitions are changing.

Simulation, in the sense of simulation theory, means that my reality (VR) or my very self (Simulation) are exactly being created by the Great Simulator (GS from now on).
— fishfry
I put out some definitions in my topic
Simulation theory and VR theory are very different, but you seem to be using simulation for both. — noAxioms

I agree. It often doesn't matter. If the simulator only simulates my experience but my mind is outside that, it's VR. If the simulator also instantiates my mind, it's the simulator. Have I got that right?

I often shorten the former to 'sim'. I am OK with defining GS as the world running the sim or the VR. With VR, you are in the GS world (but not necessarily of it), and with sim you are not.
If the GS is only approximating me or my reality, what is being approximated?
Depends on if its a sim or a VR. My topic covers this. — noAxioms

Ok. Thought I read it but I could read it again. I did object to your idea of approximation. My understanding is that simulation theory creates reality, it does not approximate it. Which is why instantiation is a better word than simulation.

Well, in sim theory, it is a simulation of at least me, so I disagree with your assertion that there is no 'of' there. In VR theory, it is the creation of my artificial experience. — noAxioms

Creation of, yes. We agree on that. NOT an approximation of; but rather the creation of. The exact creation, not an approximation.

Fine. You don't buy into the possibility of simulation theory since it contradicts other values which you hold to be true. — noAxioms

I don't necessarily reject it, I just note that it's a claim that God created the heavens and the earth, but was constrained to be a Turing machine. So it's just a theological speculation, and not deserving of the TED talkers' reverence.

You say there might be 'future physics' discovered that completes your model, but the GS might already have that understanding, and might have built their sim in such a way as to leverage it. — noAxioms

Sure. But the simulation theorists like to imagine that the GS has motives we can understand ("ancestor simulations") and operates via the laws of computation as we understand them. And also the laws of physics. If you are saying that maybe the GS has physics that we don't know about, there's no telling what they could do. It's just a magical speculation at this point.

Ok, but that's not how the TED talkers would see [difference between ID and sim]. They'd mock intelligent design, yet believe in simulation.
Besides the ridicule fallacy, how does that differ from the way I see it? — noAxioms

I'm afraid I'm not sure exactly what thesis you are arguing. My apologies.

I do use the ridicule fallacy. The exact same people who disdain God love the Great Simulator. I find that viewpoint lacking in self-awareness.

I hope that we are agreed that a simulation of gravity or a simulation of the stock market is not the same use of the word as the GS simulating my mind for me. The latter is a complete instantiation, not just an approximation.

Well, you deny the possibility of the latter, but I find it to still be the same use of the world. A simulation of our physics is necessarily an approximation since there is no way to represent anything physical exactly, so for instance it is probably going to be discreet physics with a preferred frame of reference — noAxioms

Well God, or nature, has instantiated my mind. I have no doubt of that. Call it God, call it the GS. What difference does it make, except that God is not constrained by the limits of computation?

I perfectly well agree that my mind has somehow been instantiated. I simply object to the claim that my mind could have been instantiated by a digital computer.

Are you equivocating the word simulation? Simulation theory does not mean the same as when we simulate the Super Bowl to predict the winner.

How would anybody go about doing that? — noAxioms

https://www.betstatz.com/simulator

This was the first of many links I found. It's commonly done. Simulation to predict sporting events is done all the time. Economies, the stock market, etc. Standard technology.

The GS instantiates my mind and/or my experience.
In sim theory, there is no 'my mind' to instantiate. It is not necessarily a simulation of something that also exists in the GS world. Most simulations are of nonexistent things. I suppose the weather is an exception to this since the initial state is taken from the GS world, not as a work of intentional design. — noAxioms

You have a subjective experience of your mind. Any theory of mind has to explain it or hold it as a mystery.

Good, because I think Bostrom's hypothesis falls flat. — noAxioms

Ok then we're in total agreement.

That's well known. Godel showed it for instance.
I mean, they can, but at far slower efficiency. I wrote a program that essentially simulated itself for profiling purposes. You could simulate the execution of any code (including itself), but it ran at about a 1/10000th of the normal speed, and optimized that to about 1/40th the normal speed. That could simulate itself, but per Godel, it could not be used to see if it finishes. — noAxioms

Ok.

Bostrom is clear on this. It is a simulation of ancestral history. I mock that suggestion in my topic. — noAxioms

Me too. Perhaps we are entirely in agreement.

I can't see a simulation not having a model to run. There's always an 'of', else the task is undefined. So I could run a simulation of a three legged creature to see which kinds of gaits it might find natural. There is no creature in the GS matching the one being simulated, but there's still an 'of'. — noAxioms

Well that's the thing. Bostrom says that we are a simulation. And the question is, of what? If he said we are an instantiation, that would at least make more sense.

But again, I think we are agreeing.

Tyson just seems to ride on Bostrom's paper ("<-- what he said"), which I doubt he understands.
Smoot knows what he's talking about at least, but I could not find a paper/article with his hypothesis to get even a glimmer of what he's suggesting or what evidence he claims supports it. Perhaps something concerning the CMB. It's all you-tube, and I don't get my physics from you tube. — noAxioms

Smoot did a TED talk. I get all my physics from Youtube these days.

My understanding is that simulation theory claims it's all a computation.
Bostrom suggests that. A different sim theory might not. We know nothing of the GS, so I agree, it differs little from deism. Bostrom says we know everything about GS world since they us in 'the future'. — noAxioms

Well, I agree that if mind is physical but not computational, a new definition of computation must be in our future somewhere. Perhaps some kind of continuous or real number based computation. Turing machines are hopelessly limited.

And simulation theory is God restricted to our current notions of computation.
It is only this constricted if one presumes the GS world has the same constraints as the world we know. — noAxioms

I do believe there's a lot about the world we don't yet know, and may never know.

He says the GS is us, so of course they think and feel like us. But I agree, I see no reason why they would find a need to create a fictional world framed in some past century, a simulation of the scale he suggests. It's not like it would produce any actual events that took place in the history of the GS world. What would be the staring date of such a sim? Last Tuesday? — noAxioms

The funny thing is that, other than civil war re-enactors. WE don't do ancestor simulations.

Maybe we are characters in someone's video game

Not possible given your stated beliefs. Only the players can be conscious, not the NPCs. But actually, I have suggested similar things myself, claiming to be a p-zombie in a world where not all are, because I don't see this hard problem that so many others find so obvious. Clearly they have something I don't. So OK, I'm an NPC. — noAxioms

My beliefs can be wrong. I could be a Boltzmann brain.

So a sim run by a world devoid of sickness and war, but populated by sadists with a need to create ant farms to torture? I can't see a world populated by such beings being free of natural misery.[/quoe]

Didn't say they're devoid of sickness and war. They're imposing it on us. A hypothesis with plenty of evidentiary support.
— noAxioms

— noAxioms

But even a Boltzmann brain is not a mathematical structure.
It would be be part of one under MUH, just like one would be part of our universe if there were some out there. — noAxioms

It's not a mathematical structure. A mathematical structure is a set with some operations and properties. Purely abstract conception.

Does your physical body enjoy the ice cream? You didn't answer that question. I want to see if you're consistent. — noAxioms

Umm ... no. My body processes the nutrients in the ice cream. My consciousness experiences the enjoyment. Even if pleasure is a chemical response in the brain, my experience is the pleasure. The chemicals in the brain don't have experiences, I do. Well I see that I'm getting in trouble here.

By being an avatar of a mind, but that isn't panpsychism I think, but I don't really understand that view. I suppose the rock is no different from a chess piece. I cannot move it by mind alone, but that's also true of my fingers. — noAxioms

The chess pieces don't enjoy playing the game, I can't understand why you are seemingly promoting that view.

Yes, I can extend my definition of 'me' to any boundary I wish. It's mostly just a language designation. There are no physical rules about it. — noAxioms

The me is the me that experiences. Also my body, but if I grant you that you'll say that Ms. Pac-Man's part of my body too. Don't know.

Yes, the system understands Chinese. — noAxioms

Ok then this is our disagreement. I'm with Searle, I say the system understands nothing, any more than the computer running a chess program understands chess.

A part of the system doesn't necessarily understand it, just like the CPU of my computer doesn't know how to open a text document. That doesn't mean that the computer doesn't open the document, unless that you define 'to open a document' as only something a human does, and an unspecified alternate word must be used if the computer is doing the same thing. — noAxioms

It doesn't care about the document, it doesn't know the meaning of the document, it just flips bits.

The Chinese room, as described, seems to be in a sort of sensory deprivation environment. Surely there are questions you can ask it that bear this out. They have machines now that officially pass the Turing test, and some of the hardest questions are along such lines. — noAxioms

The Turing test is weak, and the weakest part is the humans. A basic chatbot will cause humans to trust it with their deepest secrets, as Weizenbaum discovered with ELIZA.

https://en.wikipedia.org/wiki/ELIZA

https://en.wikipedia.org/wiki/ELIZA
An LLM cannot pass a Turing test. Something like ChatGTP does not claim to understand language. It's not how they work, but maybe it's not how we work either. — noAxioms

An LLM passed the bar exam. That's impressive.

I thought VR is Descartes's clever deceiver, who gives me an illusion of all my experience, yet my mind is still mine. And Sim says that my minds also is simulated/instantiated by the GS so that there really is no me outside the GS.

Yes, like that. — noAxioms

Ok, I understand the distinction then.

If Pacman was fully immersive, and I had been playing all my life, then I am essentially a mind connected only to pacman. If the game is unplugged, then all the hookups are still there, but I am left in a sensory deprivation state. If not hooked to a different feed, then it stays that way. I of course have no control over it. I cannot take off the VR headset because the connections required to do so have been severed in order to connect fully to pacman. — noAxioms

Maybe I'm a brain in a vat. But that's a nihilistic idea.

If by "computer" Bostrom means something other than a computer as commonly understood, he should say that explicitly.

Pretty sure he means 'as commonly understood'. It doesn't mean that all sim theories suggest that, but with him it kind of does. — noAxioms

Right. That's my point. Simulator = God constrained to the rules of a Turing machine. A terribly limited conception.

One could argue that the claim that consciousness is not computational is the one in need of evidence. — noAxioms

There are no structures in the brain that implement Turing machines. The neurons don't work that way.

I mean, a perfect simulation of our physics is not computational, but consciousness seems to operate at a classical electro-chemical level, and that is computational. I don't assert it to be thus, so it's a possibility, not a hard claim. — noAxioms

As a Turing machine or a digital computer? I don't see how anyone could make that claim. Although I do see many people making that claim!

There's a confusion here. The remark you quoted, which the system attributed to me, is actually Lionino. I could claim academic sources from what I'm saying, but I read them a long time ago, and if you asked my for attributions, I would have to spend a long time looking them up. — Ludwig V

Sorry about that. I typically select the entire post and hit Quote, and it seems to lose a lot of the attribution.

By definition, an infinite sequence is a1,a2,a2,… It only goes forward. Though if the elements are decreasing (as 1, 1/2, 1/4, ...) the points go from right to left.
— fishfry

I take your point. So the dots reflect the lack of definition and trying to run it backward finds the dots at the "beginning", so the "beginning" is not defined. But one could define a similar sequence that runs (0, 1/2,1/4.... 1), couldn't one? That would not be the same sequence backwards, of course. — Ludwig V

The limit is not part of the sequence. so that doesn't run the sequence backward. I am not sure what point you are making about the sequence. The dots merely indicate that the sequence progresses indefinitely.

Welcome to my world. Being out of one's depth in it is almost a prerequisite of inhabiting it, so that's not a problem. It would probably unfair to say that people who think they are not out of their depth are always wrong (compare relativity and QM). But it is certainly true that you need to be a bit out of your depth to be doing any serious work. If you have everything sorted out and pinned down, you've lost your grip on the problem. (Wittgenstein again) — Ludwig V

"If you know what you're doing you're not learning anything." Think I read that somewhere.

Unfortunately "The world is what's real, what's physical" is a metaphysical remark (at least, it is if there are any philosophers around), so you've jumped into the water without, perhaps, intending to. The question is whether numbers, etc. are real things that are not physical; platonist-type theories see numbers as real things that "transcend" the physical world. Don't ask me what "transcend" means - or "thing", "entity", "object". They would probably prefer to tell you what transcendence etc. are. But that's the same question in a different mode. Their mode is metaphysics. Mine is linguistic. — Ludwig V

Ok.

What I was doing, in response to what Lionino was saying, was putting realism and anti-realism together - since they are defined in opposition to each other - and then asking what they disagree about. (There are many varieties of both sides of this coin, so I'm simplifying, and arguably distorting.) In particular, I'm trying to show that "real" is not 'really' in contention, since no-one could deny that numbers are real - what is at stake is different conceptions of reality. And you see how slippery this is because in mathematics, not only are some numbers real and some imaginary, other numbers (like transfinite ones) are neither. Worse still, the imaginary numbers are numbers and exist, so must be real - in the philosophical sense. (At least, you can put me right if I'm wrong here.) — Ludwig V

I'm out of my depth on that. Don't understand what's meant by realism or anti-realism. Simply don't believe that 2 + 2 = 4 has a truth value before some intelligent entity shows up to pass judgment.

What "real" means depends on the context in which you are using it. Some philosophers want to use "real" in a context-free sense. But that generates huge complications and confusion. Better to stick to contexts. (The same applies to "exists") That's why I try to avoid metaphysics and metaphysicians will classify me as a linguistic philosopher - and that is indeed where I learned philosophy. — Ludwig V

Ok.

All right. Those are good questions. They lead one in a certain direction. I am very sympathetic, so it would be better to let a platonist answer them directly. But I don't think that platonism needs to rule out the possibility that humans might be able to create some things, such as fictional stories - (although Plato was very scornful about such things on moral grounds, though he made liberal use of them himself.) - and games.
But in this field, it is as well to understand your opponent's (colleague, hopefully, in a joint attempt to discover truth) position. So consider. Games like chess are unlike games like football. Once they are defined, all the possible games are defined (so long as you limit the number of moves). So you could argue that the Sicilian defence, for example, was not created, but discovered. That's the germ of platonism.
In the end, I think, one has to see these arguments, not as simple question of truth and falsity, but of how you think about things. The answers, then, are quite likely to be pragmatic or even moral. — Ludwig V

Agreed.

My enemy in this field is dogmatism. — Ludwig V

Ok. Don't think I disagreed with anything you said.

I'll take the other side of that bet. I have 70 years of AI history and hype on my side. And neural nets are not the way. They only tell you what's happened, they can never tell you what's happening. You input training data and the network outputs a statistically likely response. Data mining on steroids. We need a new idea. And nobody knows what that would look like.
— fishfry

That doesn't explain emergent phenomenas in simple machine learnt neural networks. — Christoffer

Nothing "emerges" from neural nets. You train the net on a corpus of data, you tune the weightings of the nodes, and it spits out a likely response. There's no intelligence, let alone self-awareness being demonstrated.

There's no emergence in chatbots and there's no emergence in LLMs. Neural nets in general can never get us to AGI because they only look backward at their training data. They can tell you what's happened, but they can never tell you what's happening.

We don't know what happens at certain points of complexities, we don't know what emerges since we can't trace back to any certain origins in the "black box". — Christoffer

This common belief could not be more false. Neural nets are classical computer programs running on classical computer hardware. In principle you could print out their source code and execute their logic step by step with pencil and paper. Neural nets are a clever way to organize a computation (by analogy with the history of procedural programming, object-oriented programming, functional programming, etc.); but they ultimately flip bits and execute machine instructions on conventional hardware.

Their complexity makes them a black box, but the same is true for, say, the global supply chain, or any sufficiently complex piece of commercial software.

And consider this. We've seen examples of recent AI's exhibiting ridiculous political bias, such as Google AI's black George Washington. If AI is such a "black box," how is it that the programmers can so easily tune it to get politically biased results? Answer: It's not a black box. It's a conventional program that does what the programmers tell it to do.

While that doesn't mean any emergence of true AI, — Christoffer

So I didn't need to explain this, you already agree.

it still amounts to a behavior similar to ideas in neuroscience and emergence. How complex systems at certain criticalities emerge new behaviors. — Christoffer

Like what? What new behaviors? Black George Washington? That was not an emergent behavior, that was the result of deliberate programming of political bias.

What "new behaviors" to you refer to? A chatbot is a chatbot.

And we don't yet know how AGI compositions of standard neural systems interact with each other. What would happen when there are pathways between different operating models interlinking as a higher level neural system. — Christoffer

Believe they start spouting racist gibberish to each other. I do assume you follow the AI news.

We know we can generate an AGI as a "mechanical" simulation of generalized behavior, but we still don't know what emergent behaviors that arise from such a composition. — Christoffer

Well if we don't know, what are you claiming?

You've said "emergent" several times. That is the last refuge of people who have no better explanation. "Oh, mind is emergent from the brain." Which explains nothing at all. It's a word that means, "And here, a miracle occurs," as in the old joke showing two scientists at a chalkboard.

I find it logically reasonable that since ultra-complex systems in nature, like our brains, developed through extreme amount of iterations over long periods of time and through evolutionary changes based on different circumstances, it "grew" into existence rather than got directly formed. — Christoffer

I would not dispute that. I would only reiterate the single short sentence that I wrote that you seem to take great exception too. Someone said AGI is imminent, and I said, "I'll take the other side of that bet." And I will.

Even if the current forms of machine learning systems are rudimentary, it may still be the case that machine learning and neural networking is the way forward, but that we need to fine tune how they're formed in ways mimicking more natural progression and growth of naturally occuring complexities. — Christoffer

In my opinion, that is false. The reason is that neural nets look backward. You train them on a corpus of data, and that's all they know. They know everything that's happened, but nothing about what's happening. They can't reason their way through a situation they haven't been trained on.

And the training is necessarily biased, since someone chooses what data to train them on; and the node weighting is biased, as black George Washington shows.

Neural nets will never produce AGI.

That the problem isn't the technology or method itself, but rather the strategy of how to implement and use the technology for the end result to form in a similar high complexity but still aligned with what purpose we form it towards. — Christoffer

You can't make progress looking in the rear view mirror. You input all this training data and that's the entire basis for the neural net's output. AGI needs to be able to respond intelligently to a novel context, and that's a tough challenge for neural nets.

The problem is that most debates about AI online today just reference the past models and functions, but rarely look at the actual papers written out of the computer science that's going on. And with neuroscience beginning to see correlations between how these AI systems behave and our own neurological functions in our brains, there are similarities that we shouldn't just dismiss. — Christoffer

I don't read the papers, but I do read a number of AI bloggers, promoters and skeptics alike. I do keep up. I can't comment on "most debates," but I will stand behind my objection to the claim that AGI is imminent, and the claim that neural nets are anything other than a dead end and an interesting parlor trick.

There are many examples in science in which a rudimentary and common methods or things, in another context, revolutionized technology and society. That machine learning systems might very well be the exact way we achieve true AI, but that we don't know truly how yet and we're basically fumbling in the dark, waiting for the time when we accidentally leave the petri dish open over night to grow mold. — Christoffer

Neural nets are the wrong petri dish.

I appreciate your thoughtful comments, but I can't say you moved my position.

But that's YOUR hypothesis, not mine.
— fishfry

It's not mine. It's the hypothesis of those who claim that supertasks are possible. — Michael

Ok. Well I have no strong opinion about supertasks one way or the other, and can defend either position. Perhaps I'm misunderstanding what you're trying to say to me, and vice versa.

They try to use such things as the finite sum of a geometric series to resolve Zeno's paradox. — Michael

The swine!

They claim that because time is infinitely divisible it's possible for us to perform a succession of operations that correspond to a geometric series, and so it's possible to complete an infinite succession of operations in finite time. — Michael

Incredible. But why do you conflate me with them, whoever they are?

I have been arguing firstly that it hasn't been proven that time is infinitely divisible — Michael

I agree.

and secondly that if we assume such a possibility then contradictions such as Thomson's lamp follow. — Michael

I see no contradiction in Thompson's lamp, only a failure to define the terminal state.

I was very clear on this in my reply to you on page 4, 22 days ago: — Michael

Due to my misspent youth, my short-term memory only goes back 21 days.

We can determine whether or not something entails a contradiction. If time is infinitely divisible then supertasks are possible. Supertasks entail a contradiction. Therefore, time being infinitely divisible entails a contradiction.
— Michael — Michael

Is this argument supposedly in response to anything I've ever said? I no longer believe you and I are even having the same conversation.

Most of the last few pages has been me trying to re-explain this to you, e.g. 10 days ago: — Michael

Shame on my for not remembering.

These arguments only show that if I recite the natural numbers as described then I have recited all the natural numbers, but this does nothing to prove that the antecedent is possible, and it is the possibility of the antecedent that is being discussed.
— Michael — Michael

Well it's not my antecedent. And now you claim it's not your antecedent. So why are you and I arguing about it?

It was brought up for two reasons. The first was to address the flaw in your reasoning. — Michael

I may be mistaken on facts; but I doubt I've made any flaws in reasoning. On the contrary, one of my flaws in life is excessively precise reasoning.

That same post 10 days ago was very clear on this:

Argument 1
Premise: I said "0", 30 seconds after that I said "1", 15 seconds after that I said "2", 7.5 seconds after that I said "3", and so on ad infinitum.

What natural number did I not recite? — Michael

You recited them all. Are we agreed or disagreed on that?

Argument 2
Premise: I said "0", 30 seconds before that I said "1", 15 seconds before that I said "2", 7.5 seconds before that I said "3", and so on ad infinitum.

What natural number did I not recite? — Michael

You recited them all. Which number didn't you recite? Are we agreed or disagreed on that?

These arguments only show that if I recite the natural numbers as described then I have recited all the natural numbers, but this does nothing to prove that the antecedent is possible, and it is the possibility of the antecedent that is being discussed.
— Michael — Michael

Yes but it's an entirely hypothetical thought experiment. And not mine, either. It's Thompson's. Take it up with him.

If argument 1 is proof that it is possible to have recited the natural numbers in ascending order then argument 2 is proof that it is possible to have recited the natural numbers in descending order. — Michael

You are equivocating "possible." If by possible you mean relative to the entirely hypothetical, abstract premises, then it's possible. If you mean physically possible, then it's doubtful, and false under current physics.

It is impossible to have recited the natural numbers in descending order. — Michael

1, 1/2, 1/4, 1/8, ...

That's a perfectly sensible and well-known infinite sequence. If at each step you recite "1", "2", "3", and so forth, you'll get it done. But what of it?

Therefore, argument 2 is not proof that it is possible to have recited the natural numbers in descending order. — Michael

Well the natural numbers have no last element, so I have agreed -- REPEATEDLY -- that you can't recite the natural numbers backwards. I have stipulated to this at least half a dozen times.

Therefore, argument 1 is not proof that it is possible to have recited the natural numbers in ascending order. — Michael

Under the successively halved time interval hypothesis, it's clearly possible to recite the natural numbers in ascending order. Proof: You can recite 1. And, if you can recite n, you can recite n + 1. Therefore, by mathematical induction, you can recite them all. I've asked you many times now to engage with this argument and you won't.

The second reason I brought it up was a proof that it is impossible to have recited the natural numbers in ascending order. — Michael

I just showed how it is. And rather than engage with the argument I've given several times, you just repeat your false claim.

If it is possible to have recited the natural numbers in ascending order then it is possible to have recorded this and then replay it in reverse. — Michael

If you play the recording in reverse, your first step, no matter how small, must necessarily jump over all but finitely many numbers. I have also given THIS argument many times now, and you pointedly refuse to engage.

Replaying it in reverse is the same as reciting the natural numbers in descending order. — Michael

No it's not, as I've explained several times now.

Reciting the natural numbers in descending order is impossible. — Michael

I have agreed with this several times.

Therefore, it is impossible to have recited the natural numbers in ascending order. — Michael

I already showed that it's possible, under the hypothesis of successively halved time intervals.

Or if you don't like the specific example of a recording, then the metaphysical possibility of T-symmetry might suffice. — Michael

Lost me there.

Either way, the point is that it's special pleading to argue that it's possible to have recited the natural numbers in ascending order but not possible to have recited them in descending order. It's either both or neither, and it can't be both, therefore it's neither. — Michael

It's tedious at my end to dialog with someone who refuses to engage with any of my arguments, but nly keeps repeating their own talking points.

You seem sincere in your own confusions, but this is not productive and has not been for quite some time.

We'll have human-level Ai's before too long. — RogueAI

I'll take the other side of that bet. I have 70 years of AI history and hype on my side. And neural nets are not the way. They only tell you what's happened, they can never tell you what's happening. You input training data and the network outputs a statistically likely response. Data mining on steroids. We need a new idea. And nobody knows what that would look like.

Me? I make no such claim.
— fishfry
No, not you. No quote of yours was in the bit there to which I was replying. — noAxioms

Ok I might have been confused.

I say that consciousness is physical but not computational.

What do you mean by that? I mean, technically, none of physics is computational if done to a sufficient level of detail, but I don't think that level of detail is needed in a simulation.
Computation is classical and physics has been shown to be not. — noAxioms

Oh ok these definitions are changing.

Simulation, in the sense of simulation theory, means that my reality (VR) or my very self (Simulation) are exactly being created by the Great Simulator (GS from now on). If the GS is only approximating me or my reality, what is being approximated?

Simulation is in fact not the right word. We have simulations of gravity, simulations of star formation, simulations of elections, simulations of economies. The word simulation is always accompanied by "of." If there is a simulation, it's a simulation "of" something.

In simulation theory, there is no "of." The simulation is exact and perfect already. So the right word is instantiation, not simulation.

So when I say that intelligence, or mind, or the world, is physical but not computational, I mean that the universe does something that is physical -- involves the atoms and the quarks and the quantum fields and whatever future physics will be discovered -- that transcends our current understanding of computation. It's a Turing machine with an oracle for consciousness, maybe that's a good way to put it.

So my conception is not dualistic, not supernatural, not non-physical. It's entirely physical. But it's not subject to the profound limits of Turing machines, finite state automata, and any of the other things that we currently think of as "computation." The universe is transcending the [ul=https://en.wikipedia.org/wiki/Church%E2%80%93Turing_thesis]Church-Turing thesis[/url].

That's my view of it, my conception of what's going on. Physical, but not a computation as we currently understand it.

Of course I entirely agree with you that we can simulate the heck out of a lot of things via approximation and abstraction. But we can't instantiate consciousness with a computation, as computation is currently understood.

Is my position somewhat clear? The world and the mind are physical but not computational, in our present understanding of computation.

What's the difference between [ID]and sim theory?
Not too much. Both are deliberate choices of interesting mathematics. The vast majority of possible universe are not interesting. — noAxioms

Ok, but that's not how the TED talkers would see it. They'd mock intelligent design, yet believe in simulation.

[And again, I hope that we are agreed that a simulation of gravity or a simulation of the stock market is not the same use of the word as the GS simulating my mind for me. The latter is a complete instantiation, not just an approximation].

A simulation of gravity does not implement gravity. Simulations of brains therefore do not necessarily implement minds.
I didn't say implement them. I said that they would find the familiar pattern. If nothing is known about how that works, then you can't say it wouldn't happen with the sim. — noAxioms

Running gravity simulations does not attract nearby bowling balls. "Find the familiar pattern," not sure what you are getting at.

Are you equivocating the word simulation? Simulation theory does not mean the same as when we simulate the Super Bowl to predict the winner. It means to instantiate. The GS instantiates my mind and/or my experience.

The question is, why do we mock the Godly street preacher, and venerate the simulation theory TED talker?
There's a lot more veneration of the God talkers than you suggest here, and if Bostrom screamed his assertions from a box in a subway station, he'd get a lot less attention. He's getting mocked plenty in topics like this one. Bostrom is venerated at the Ted talk because the audience is full of people who's seen Inception and think that's what he's talking about. — noAxioms

I didn't see it, I thought it would just aggravate me. I have a tropism against trendiness.

Maybe I should just go read Bostrom. I did take a run at his paper once and gave up after a while. I should try again, I might learn something.

I'm gladly advocating it?? Bostrom claims we are in a sim of us: The world simulating us is the same as the one simulated. — noAxioms

Right. That is a confusing equivocation of the word. It's not like when the physicists simulate the early conditions of the universe. Simulation theory is what God does to instantiate the universe. Which is why simulation theory is essentially a theological speculation.

Apologies if I confused your views with Bostrom's.

A thing can't simulate itself. Unless, by that definition, I'm simulating myself just by sitting here and being me.

Simulation's a bad word. I always want to ask the simulation proponents, simulation of what?

I had a funny thought. Just as all waves must travel in a medium; yet light is a wave that does not require a medium; perhaps the GS simulates without the need for an "of" to simulate.

I think the word should be instantiate, because that is what's being conveyed.

That's not ID since the design is already made and it is just mimicry. But in general, if you admit that we know nothing of the world running the sim, then the idea is no different than deism. — noAxioms

We agree then. Neil deGrasse Tyson and George Smoot believe in simulation theory. And Tyson, at least, would object to the claim that "God does it all, and we are His flock of lambs." Bothers me that people who should know better advocate these ideas.

I see no difference between "God did it" and "The Great Simulator" did it, except that the GS is required to be a computation
Is it? If we can know nothing of those running it, how do we know it is a computation? At what point does it cease to be sim theory and just become straight up god:"whoomp, there it is" theory? — noAxioms

Ah. My understanding is that simulation theory claims it's all a computation. If I'm wrong about that I should correct my thinking. I will go look at Bostrom's paper, maybe he says something about that.

It seems a lot of my answers agree with yours, but your tone suggests disagreement with my replies. — noAxioms

I think we're in agreement. Apologies if I'm unnecessarily oppositional.

Simulation theory says we are computations. That can only be understood as computation as we currently understand it. Turing machines, finite state automata, etc.
OK. You have a tighter definition of the term. You must call it something else if it is done, but not done as computation as you currently understand it. Do quantum computers qualify? Are they (if one is actually created) beyond our current understanding? Can they run a simulation, or would a different world need to be used? Can a quantum computer solve the halting problem for a Turing machine? — noAxioms

As I understand it, quantum computers have no more computational power than classical ones. They run more efficiently in terms of complexity theory. But any function that a quantum computer can compute can already be computed by a classical computer. And the proof is that quantum computers are often simulated on classical hardware. They run slowly, but in principle they do the same things either way.

A quantum computer most definitely can not solve the Halting problem. They can do certain problems much faster than classical computers, but a classical computer could solve the same problem if given enough time.

I mean, the god people do it all the time. God created physics, be it computable or not. — noAxioms

Right. And simulation theory is God restricted to our current notions of computation. That's my argument. It makes God the more likely hypothesis, by virtue of requiring one less assumption.

Time as well, and general causality. That sounds an awful lot like a simulation mechanism to me. Old school says the sim began ~6000 years ago, but lately, in an attempt to avoid all out denial of science, they've backed off to a view of the project starting at the big bang, and perhaps with initial conditions that bring us about, because it's all about us after all.
That's a big difference BTW between god and a sim: A sim is run to see what happens, to gain information. God creates something where he knows exactly what will happen, and he wants that to happen. He gains no knowledge by running the universe experiment, at least not the god typically asserted. — noAxioms

Maybe God created us all then sits back and watches. Like a kid with an ant farm, do they still make those? Why yes, they do. https://www.antsalive.com/shop.htm

That's another thing about Bostrom's argument. That the GS runs "ancestor simulations." First, WE don't run ancestor simulations, though we certainly could. Get some programmers and some sociologists in a room and they could work it out. Like Sim City, the game.

But why should the GS run ancestor simulations? Isn't it rather arrogant of us to impute motives to the GS as if the GS thinks and feels like us? Maybe we are characters in someone's video game. We play video games for entertainment, not to gain insight into our ancestors.

Deism isn't theological. It would be if those running the simulation implemented say a moral code which they expect to be followed by the subjects being simulated, "or else ...". — noAxioms

Matter of definition I guess. I don't think theology necessarily needs to come with a moral code. Just needs a supernatural being that instantiates the world. Like the GS.

So the Great Simulator doesn't ask Abraham to kill his son?
That's messing with the simulation, violating the causality rules and such. If it works like that, then its a VR for the great simulator, and the rest of us are NPCs being asked to kill our sons. — noAxioms

I'm not sure I remember what my point was with that question. But why not? Bostrom's GS likes to run "ancestor simulations." Why on earth would they do that? Why on earth would WE imagine that they do that? Maybe they are the cause of sickness and war and suffering. Maybe they are sadists. That's a more realistic hypothesis than that we're an ancestor simulation.

But Tegmark's MUH is such a category error that I can't imagine he's serious.
Him redefining the categories is not a category error. You're begging a different definition. Mathematics is not a map in the view. — noAxioms

He says the world literally is a mathematical structure. I have had the opportunity to see a handful of modern mathematical structures, and I don't think the world is one of those. It's a category error. So I should go read Tegmark too, another paper I took a run at some years ago without feeling the magic.

The MUH predicts that the majority of consciousness are Boltzmann brains, reducing the hypothesis to where it cannot be simultaneously believed and justified. That's a huge hit to the idea, and one which he must be aware, and has perhaps attempted a refinement, but it wasn't addressed in the book. — noAxioms

But even a Boltzmann brain is not a mathematical structure. It's an implementation of one. Mathematical structures by definition are sets with operations and rules. They are abstract. They may have some earthly representations, but the representations are not the structures. A bag of groceries is not a set. Sets are far stranger than that, and sets do not exist in the physical world.

A simulation is a created thing. It exists in time. There's no evidence that our universe exists in time. — noAxioms

Block universe? Time is an illusion? Maybe. Can't refute it.

Oh I see your point! Thank you for explaining that. She gets her consciousness from me. I enjoy making Ms. Pac-Man eat the dots. I can see that. But Ms. Pac-Man does not have an inner life.
You see that Ms Pacman is you, but you still deny your inner life? — noAxioms

No, by the time I was done last night I rejected your concept. Same reason that my chess pieces do not care if they get captured or win or lose the game. I'm the one who cares. They're just inanimate pieces of plastic or bit patterns on my computer screen run by a chess algorithm.

A bit like you saying that your experience is the same experience had by the body of fishfry. Well, fishfry body doesn't have experience separate from 'you', and similarly Ms Paceman doesn't have separate experience. She does become a zombie while the game isn't being played, zooming around randomly and getting killed in short order. — noAxioms

I was momentarily happy that I understood the point you were making, but I now reject it completely. The chess pieces don't care. The baseball doesn't care when someone hits it out of the park. And which side is it on? The batter is happy but the pitcher is unhappy. And the baseball has no emotions at all.

Is this a form of pantheism? I enjoy throwing a rock, and by your theory, the rock enjoys being thrown.
It does? Where did I say anything like that? Because I intentionally caused it to move? That's different than me being the rock while doing so, making it move on its own. — noAxioms

I thought that is your thesis. That Ms. Pac-Man is imbued with my own pleasure in winning the game.

But of course that's not right. It's just a video game run by a circuit board. (And so are we, according to the computational theory of reality that I reject).

Pantheism? What's that got to do with it? Do you mean panpsychism? — noAxioms

Yes did I write pantheism? Well my fingers have a mind of their own, maybe you're right :-)

Yes pansychism. How else can the rock, the baseball, the chess pieces, and Ms. Pac-Man experience my pleasure in the game?

I think you are not convincing me of this thesis, though you did have me going for a while last night.

A dualist has a mind and a body, and typically the body has presumed boundaries which usually don't include the rock, but there's no actual hard definition of where the boundary is since there's nothing physical about it. So for instance, are the clothes I'm wearing part of me? The usual presumption is yes, despite that probably not being the answer if it is asked as a question.
"Where does 'you'" physically stop? It's more of a language thing than a physics thing. I typically don't include the rock as part of 'me', and you probably don't either. I could open an entire topic about this. — noAxioms

You seem to include Ms Pac-Man as you. Isn't that what you said? Am I perhaps not understanding?

Searle also plays the game of refusing to apply a word to something nonhuman doing exactly what the word means when a human does it. That's begging his conclusion. — noAxioms

Well yes, this comes up in these discussions. We compute, computers compute, so we're computers. The Chinese room speaks Chinese, who am I and who is Searle to say it doesn't understand what it's doing?

That's the argument against my position. My Chinese friend speaks Chinese and my Chinese room speaks Chinese, what's the difference.

I say there's a difference. Searle says there's a difference. Everyone agues with Searle and I.

I looked at the wiki page and the argument seems to have been updated. The guy doing it (instead of the computer) cannot pass the Turing test since speed is an issue. Somebody who takes 20 years to reply to 'hello' is probably not going to pass a Turing test. Speed up time in the box and this objection goes away. No, the man in the box does not understand the conversation any more than does the CPU in the AI or than does a human brain cell. — noAxioms

Well we're back to LLMs. They're not very bright but many people these days are impressed. Searle of course did not put his argument in terms of computers, but today we could literally program a Chinese room. It's Google translate. Which in general isn't very good, language translation is a hard problem. But I don't think we should let execution speed muddy the argument. I'm willing to stipulate that the Chinese room is as fast as it needs to be. It still doesn't understand Chinese. And a few million professional and amateur philosophers disagree with me.

Physics is the historically contingent human activity of Aristotle and Newton and Einstein explaining why bowling balls fall down.
Not talking about a human activity. I'm talking about the actual nature of the world, not how we describe that nature. — noAxioms

Ok.

With sim, the world behavior (physics) is primary, and things proceed according to the rules, without outside interference or intentionality. I have done both kinds. They're very different. — noAxioms

I may be missing some subtleties. I thought VR is Descartes's clever deceiver, who gives me an illusion of all my experience, yet my mind is still mine. And Sim says that my minds also is simulated/instantiated by the GS so that there really is no me outside the GS.

If there's a simulator, they may get bored of providing me with this interesting reality and unplug me, and I'll cease to be.
That sounds more like a sim, yes. If they unplug it, everything/everybody is gone, but perhaps still on disk somewhere. It could be restarted 2 years from now and the simulated beings would never notice the interruption. They very much would notice if it was a VR. — noAxioms

Yes right, so I believe I'm understanding you.

It would be like quitting PacMan. Devoid of experience of the pacman world, but not devoid of experience. — noAxioms

Memory?

Well this convo's getting long.

Apologies again if my tone is oppositional sometimes, it's how I try to figure out what I think.

PS -- I looked up Bostrom's paper.

https://simulation-argument.com/

The very title of the paper is: Are You Living In a Computer Simulation?.

If by "computer" Bostrom means something other than a computer as commonly understood, he should say that explicitly.

So my remarks on computability stand. Bostrom's thesis that the world and my mind are computational, as the word is understood today, is an unwarranted and probably false assumption; and in any event, needs evidence.

The world can not be simultaneously Euclidean and non-Euclidean.
— fishfry

I am not talking about the fabric of space-time. — Lionino

Perhaps I misunderstood. What then?

Nothing to do with the physical world.
— fishfry

Right, except for the kinds of realism that make it about the physical world, but that is one type among many. — Ludwig V

Not conversant with the technical meaning of realism in this context, probably can't hold up my end of this.

Maybe you are misunderstanding what "abstract" means in those quotations. It doesn't mean something that we conceive in our minds, but a real object that exists independently of any conscious being, but that is outside space and time. — Ludwig V

Is this a dualist point of view? What is outside of space and time?

I think the question, "Was 5 a prime number, before there was life in the universe?" is meaningless. Are you saying that a realist would say the answer is Yes? In that case, what is the difference between a realist and a Platonist?

I freely admit to my philosophical ignorance, so I am out of my depth in these matters.

Of course a single sentence doesn't represent a family of views. But one of the minimal characteristics of mathematical realism is that things such as "2+2=4" are true and they are true even if we are all dead — in other words, it is about the world. — Ludwig V

But no, that is not about the world. The world is what's real, what's physical.

So you (or the realists) are arguing a Platonic position then. I don't happen to think 2 + 2 = 4 is true before there was intelligence in the universe. Or, say, before the big bang. Or before the endless cycle of big bangs, "big bounce" speculative cosmology. I think it takes a human, or at least an intelligence, to give meaning to the proposition.

If we are all dead, if there is no life and no intelligence in the world, who can pass judgment on whether 2 + 2 = 4? How could it be assigned a truth value? Who would do the assigning, the agreeing and the disagreeing?

Again, I admit my thoughts in this area are naive and not the product of any directed study of what the smart folks think about these matters. So I don't put a lot of stock in my own opinions.

Let me ask you a different question. Before chess was invented, did all the games of the grandmasters exist "out there" in Platonic space? Did the collected games of Bobby Fischer exist before he played them? After all, each game could be encoded as a number, and the Platonists believe numbers exist independently of minds. I find that difficult to believe, that all the symbolic works of humanity exited before they were created.

Humans create. That's what we do. Humans are, if you like, the very mechanism by which the universe figures out if 2 + 2 = 4.

I hope not, my sources are academic. — Ludwig V

I have no doubt, and I hope I am sufficiently conveying the humble limits of my knowledge in this area.

But these academic sources, are they Platonists? Dualists? Surely there are those who strenuously disagree and take something approximating my own point of view. That 2 + 2 = 4 has no truth value before there is an intelligence to assign it one; and that the chess games of the grandmasters did not exist before they were played, even if the list of moves is an abstract sequence of symbols that a Platonist must believe existed before time itself.

I put this to ChatGPT4. Have a look at what it said. — Wayfarer

Impressive but meaningless, as all LLM output is.

Not because machines could never be conscious. I believe they can't, but I could be wrong.

Rather, because the very design of LLMs renders them literally moronic. They're autocomplete on steroids. They crunch a huge pile of text and make a big lookup table (conceptually) that says, "If you see this phrase, this other phrase is likely to follow."

Then the programmers decide whether they always want it to pick the most likely phrase, which gives a boring LLM; or the least likely phrase, which gives a crazy but interesting LLM. They typically tune it till it outputs seemingly interesting blocks of text that are -- by the very design of the system -- utterly meaningless if you are looking for insight, but superficially clever looking.

For factual data they can be excellent. "Write an essay on medieval sports." But for insight, "Are LLMs intelligent," they spout nonsense, no matter how seemingly clever.

I'm not saying anything that people don't already know about LLMs. What I wrote is commonplace knowledge among people who work with LLMs.

I agree with "a bat has.... what ultimate reality is" But then, I wonder what the status of "what's really going on in the world". Is that ultimate reality? From what you say, the answer is not clear. — Ludwig V

Don't know how to answer that. If there is an ultimate reality, whether we have access to it or not, that's what it is. If there isn't, then maybe it's all random. But then that would be the ultimate reality. God's point of view, as it were.

My concern is that both "ultimate reality" and "what's really going on in the world" are not defined in a way that reminds me of the way that the last step in a converging series is not defined - and cannot be defined. Yet, the sun is really shining at the moment and there really is a war in Ukraine - in short, we all (including bats and ants and slugs) live in the same world and interact in it. — Ludwig V

No, those are only the things we perceive. Ants don't see the war but they do experience the sun. Creatures at the bottom of the sea don't see much of the sun. Ultimate reality is whatever is really out there, if there even is such a thing.

But how can you say that an ant's view of the world is inaccurate? — Ludwig V

Limited is a better world. Our view is limited too.

I think I can grasp what you are getting at when you say that physics is inaccurate. It reflects the fact that physics is an on-going enterprise. "What if it's wildly inaccurate.." is a style of question that I'm very sceptical of. It reminds me of "what if everything's a simulation?" I classify it as a speculation and not capable of a meaningful answer. — Ludwig V

Ultimate reality is the thing that physicists believe (or used to believe) they are trying to understand and learn about. That is no longer a core objective in physics, since the "shut up and calculate" school of quantum physics won the day.

As I understand it, Tegmark believes the world is a mathematical structure, like a group o a topological space.
— fishfry
One might interpret that belief as a dramatic way of putting the point that we can find a mathematical structure that applies to the world. If he doesn't mean that, I want to know what he means by "is". — Ludwig V

He says that the world literally is a mathematical structure. Not "is described by," but literally is.

https://en.wikipedia.org/wiki/Mathematical_universe_hypothesis

I have the worst habit lately of only responding to my mentions and not reading the rest of these threads.
— fishfry
A very sensible policy. It is easy to drive oneself crazy by trying to respond to everything. But sometimes I can't resist intervening in discussions that haven't mentioned me. It doesn't always work, in the sense of developing into something interesting, but some times it does. — Ludwig V

I'm probably missing a lot.

I ended up spending all my time explaining the ordinals and that detracted from my resolution of the lamp.
— fishfry
That's my fault. Sorry. I did benefit very much. — Ludwig V

I love talking about the ordinals so if you enjoyed that I'm happy.

But the limit isn't defined in the lamp problem.
— fishfry
Yes, I understand that now. I was talking about the limit of the convergent series. The series "0,1,..." has no inherent limit. If it ever is limited, it is by some event "outside" the series. That's badly put. I just mean that I can stop following the instruction for any reason that seems good to me or even none at all. The series as defined is infinite. — Ludwig V

It's infinite in the sense of being defined at each of 1, 2, 3, 4, ..., but it is NOT defined at $\omega$, the "point at infinity." Or for a more concrete representation, it's defined at each of 1/2, 3/4, 7/8, ..., but it is not defined at 1. That's the point. The 0, 1, 0, 1, ... sequence only covers the infinite sequence. But there is no definition of the state at the limit point.

I'd say that the standard mathematical rules for dealing with infinity are perfectly clear, and do apply.
— fishfry
I didn't mean to suggest that wasn't the case. Thinking of the series backwards is a vague handwavy imagining. — Ludwig V

That's @Micheal's example. But the sequence 1, 1/2, 1/4, 1/8, ... is a familiar sequence with limit 0. If you plot the points on the real line, it does indeed go backward, from right to left.

That's all. I intended to contrast that with a series that can be defined forwards or backwards. It's odd, that's all. — Ludwig V

Not sure what a sequence (not series, a series is an infinite sum) that can be defined forwards or backwards means. By definition, an infinite sequence is $a_1, a_2, a_2, \dots$. It only goes forward. Though if the elements are decreasing (as 1, 1/2, 1/4, ...) the points go from right to left.

Yes, sure, a fixed body of knowledge evolves. But that body of knowledge is added to every day by every math journal and university colloquium.
— fishfry
Both sentences are true - the first sentence does not imply anything platonic, in my view. I think the difference between us is a question of emphasis rather than an actual disagreement. — Ludwig V

Well, human endeavors are never fixed, they always evolve. In the case of math, it's a philosophical question as to whether there's something "out there" that it's evolving to.

I believe I lost track of what this paragraph referred to, sorry.
— fishfry
Yes, that was a step too far, and it is very speculative, more a musing than a thought. I should not have pursued it. Let's just let it go. — Ludwig V

It looked interesting, I should take another look.

Agreeing with what follows if we can recite the natural numbers at successively halved intervals of time doesn't prove that we can recite the natural numbers at successively halved intervals of time. — Michael

Of course. Fully agreed. But that's YOUR hypothesis, not mine. Am I wrong about that? What am I missing here? Isn't that your example?

You have repeatedly asked me what happens if we go backwards, saying "1" at 60 seconds, "2" at 30 seconds, and so forth. That also is a purely hypothetical thought experiment. Why on earth are you proposing hypothetical non-physical thought experiments, then saying, "Oh that's impossible!" when I attempt to engage?

I also think that's what Peirce was getting but that's definitely not what I'm getting at. Remember when I "trolled" you by introducing a scenario involving infinitesmals? I believe that approach aligns with Peirce's thinking and I believe it's wrong. — keystone

It's the only way I can make sense of what you're saying.

You keep trying to concieve of my line as something built from smaller more fundamental elements (before points, now infinitesimals). It is not built from anything. (0,1) is one object - a line. The smaller elements emerge from the line, not the other way around. — keystone

[0,1] is standard mathematical notation for a particular set of real numbers. You can't fault me for bringing my preconceptions about that notation. You should use less suggestive notation.

How about "L". If you say, "I have a line, I call it L," then I can't come back and challenge you about that notation.

You want it both ways. You think you can traverse the line from 0 to 0.5 to 1, freely borrowing our high school intuitions about the real number line. And when I bite on that bait, you say, "Oh no it's not the real line!"

You call your line [0,1], you treat it as if it's the usual unit interval, and then you object when I believe you!

I'm not allowing a single real number. We can partition the S-B tree at a rational node (e.g. 1/2), but we cannot partition it at a real node (because real nodes don't exist). — keystone

Ok back to no real numbers.

Just as you don't grant infinity actual status as a natural number, — keystone

It's not a natural number. It's not 0 and it's not the successor of any number. I can PROVE "infinity", whatever you mean by that, is not a natural number.

I don't grant irrational points actual status as points. After all, infinity and irrational points are inseparably linked in the S-B tree, since irrational points become actual points at row infinity. If there is no actual row infinity, there are no actual irrational points. — keystone

Last post you started believing in the real numbers, but you called them fictitious. Now you're backing off even that.

The difference is that you believe individual irrationals can be isolated, whereas I think we can only access irrationals as continuous bundles of 2ℵ0 — keystone

Whatever that means.

fictional points. A mathematical 'quanta' if you will. In a 1D context, I refer to this continuous bundle as a line. And if we cut a line, we have two lines (i.e. two bundles of 2ℵ0 — keystone

Ok. Can't agree, can't disagree.

fictional points). No matter how many times we cut it, we will never reduce a bundle down into individual points. — keystone

That's true of the real numbers as well. You know, I think you are just coming to understand the nature of the standard mathematical real numbers.

If you start with the real line and cut it any number of times, you can never isolate a point that way. Do you agree?

Since we can only ever interact with these bundles, it is meaningless to discuss individual irrationals - they are fictions. The bundles are not. Do you see the distinction? — keystone

No, because the real numbers have the same property. I cut [0,1] in half, I get [0, 1/2) and [1/2, 1]. I'm arbitrarily placing 1/2 in the second segment.

If I cut [0, 1/2) in half, I get [0, 1/4) and [1/4, 1/2), and so forth.

No number of cuts will ever isolate a point.

So I think what is happening here is that you are coming to a better intuition of the standard real numbers. Because you can keep cutting the unit interval in half and you will never isolate a point.

Patterner bumped this old post, so I tracked down what was being referenced.

I'm not making any claim other than we know mind and consciousness exist. It's up to the people asserting mindless stuff (i.e., matter) exists and consciousness and mind emerge from it to prove it.
— RogueAI

Minds/consciousness can't come from matter, therefore simulation theory is false.
— RogueAI

How do you prove that?
— Benj96

Why is the burden of proof on me? We know mind and consciousness exist. The existence of mind-independent stuff is simply asserted. I would like to see a proof that this stuff exists. Something a little more robust than "go kick a rock".
— RogueAI

You're making the strong claim that mind/consciousness can't come from matter, so the burden of proof of that claim is definitely on you. If Bostrom makes the claim that mind/consciousness does emerge from matter, then the burden of proof of that is his. I'm not sure if he's making the claim directly, but his sim argument depends on it, and he's claiming the sim argument, so the burden is still there, as it is on you for your strong claim. — noAxioms

Me? I make no such claim. I've made the opposite claim. I say that consciousness is physical but not computational. I hope that's clear. Or was that for the other tagged handles?

You make a second claim, that sim theory is false if your assertion is true. To me, that's another thing in need of proof. You arrange matter into a person and somehow a mind thingy finds it. — noAxioms

Yeah that's a real puzzler. It clearly seems to be going on a lot, so it's true. We just don't know how it works. Sim theory is false because consciousness is not computational. You're right that I assert both those claims.

By the way, do you believe in intelligent design? If you said that in polite company you'd be shunned. What's the difference between than and sim theory?

What's different about the simulation that the same thing wouldn't happen, that the simulated thing would be conscious the same way you claim to be, despite it being attached only to a simulated physical? — noAxioms

When I run a simulation of gravity on my computer, nearby bowling balls are not suddenly attracted to it by any amount greater than what can be accounted for by the mass of the computer. A simulation of gravity does not implement gravity. Simulations of brains therefore do not necessarily implement minds. Not saying they don't, just that it's not the case that they necessarily do.

Why do people who reject God accept the Great Simulator?
— fishfry
More to the point, why would anybody (even Bostrom) accept the SH? People choose a view either because there is evidence or because they want it to be true. The former is a rational motivation and the latter is rationalized. Bostrom's argument seems to attempt to bend the facts horribly to make the hypothesis plausible. This suggests that he wants it to be for some reason, but I cannot fathom why somebody would want to actually believe that. OK, I see why one might want to appear to believe it: Because of the popularity of the idea from movie fiction. He has gained money/status/notoriety from pushing a view that nobody else is in a coherent manner. Elon Musk is a decent example of an incoherent hypothesis, and he's not doing it for the notoriety that he already had. Without knowing it, he pushes for VR, and I can see reasons why somebody might choose that. — noAxioms

Not the point I was trying to make. The question is, why do we mock the Godly street preacher, and venerate the simulation theory TED talker?

Again, do you believe in intelligent design? Nothing provokes scientists more than that idea, they hate it. While gladly advocating simulation theory.

Simulation theory is a theological belief. That's the point I was making. Not, why do people believe in simulation, but why do so many who believe in simulation reject intelligent design or God? I see no difference between "God did it" and "The Great Simulator" did it, except that the GS is required to be a computation, and that makes it less likely than God, because God requires one less assumption.

The GS is just God constrained to computability.
The world simulating us is not constrained to the computability laws that constrain our world. — noAxioms

This I utterly reject. Simulation theory says we are computations. That can only be understood as computation as we currently understand it. Turing machines, finite state automata, etc.

If someone is saying we're computations but computations defined differently than they are in computer science, the burden is on them to make that remark coherent.

It is thus constrained in Bostrom's view, but not in general. — noAxioms

At least Bostrom agrees with me in that regard, then. The word computation has a very specific meaning. For example if the Great Simulator can solve the Halting problem, it's not a computation. Maybe it's a computation with an oracle for the Halting problem. If so, then simulation proponents should make that assumption explicit.

It's sort of a computing version of deism. — noAxioms

Well then you are agreeing with me. It's a theological claim.

The creating simulator starts it up, but then steps back and never interferes and lays no demands on what the occupants do, nor does it make any promises to them. The typically posited god usually does have promises and demands, but not necessarily under deism. — noAxioms

So the Great Simulator doesn't ask Abraham to kill his son? Or not mess around with his neighbor's wife? You are stretching a bit now.

I laid out my case that Tegmark is a troll here ...
— fishfry
I haven't got round to replying to that endless topic yet, but Tegmark is more appropriately discussed here since it has little to do with supertasks. — noAxioms

As it happens, lately I just respond to my mentions and often have no idea what thread I'm in. But Tegmark's MUH is such a category error that I can't imagine he's serious. I'm sure others have made that point. That MUH is a category error, not necessarily that Tegmark's a troll.

You say category error: Please explain that without begging a different view. You do explain it there, but you are very much begging a different view when doing so. Tegmark is saying that mathematics (not any mental concept of it) IS the territory. Our abstract usage of mathematics is the map, but that abstraction is not what is the universe. — noAxioms

Well, the map is not the territory. Anyone claiming that the map is the territory is making a category error. I don't think I need to explain that, it's pretty clear. What map is its own territory? Is it a map a map of itself? This is word play. The map is not the territory. I hardly need to defend that proposition.

It's not much different than all these centuries where the universe was considered to be an 'object', a thing contained by time and in need of creation. — noAxioms

But now we know better. It doesn't need creation, only simulation!! /s

They all of a sudden a new view comes along and the category changes. It isn't an object created in time, but rather a structure that contains time. Most people still hold the 'contained by time' view since it is more intuitive. Tegmark is doing something similar: changing the categorical relations. Refute it from its own premises, but not by begging different ones. — noAxioms

Having seen what mathematicians mean by structures, from groups, topological spaces, measure spaces, and the like, I find it impossible to comprehend a point of view that claims that these things are flesh and blood. A mathematical structure is a set with some operations and some rules. They can model certain aspects of reality. They are not reality itself.

Your refusal to apply the language you use for human activities to something non-human doesn't mean that the non-human thing isn't doing them. — noAxioms

Your browser does not have an inner life. We've been having a very sensible conversation, but I can't join you in this belief of yours that Ms. Pac-Man experiences pleasure. I literally can't believe you are saying that to me.

Ms Pacman is you. — noAxioms

I just can't hold up my end of this part of the conversation. What are you talking about? Perhaps if you have been inspired by something you read to assert such an absurdity, you can give me a reference so that I can regain my bearings.

It's a VR game, and you enjoy eating the dots, else you'd not be cramming quarters into the machine. It is a straight up case of dualism. Ms Pacman's consciousness is yours. She is the avatar, who doesn't enjoy the dots any more than you claim your physical avatar enjoys the ice cream. — noAxioms

Oh I see your point! Thank you for explaining that. She gets her consciousness from me. I enjoy making Ms. Pac-Man eat the dots. I can see that. Ms. Pac-Man derives her inner life from mine.

That is a very interesting idea.

You raise a question that I'm not sure I can answer. What is the distinction between my enjoying myself playing the game, and Ms Pac-Man's enjoyment of eating dots? After all, I try to eat the dots and I try to avoid the monsters. I'm the one who has Ms. Pac-Man's experiences, but those are (in your view) her experiences nonetheless. My experience is her experience.

I haven't got the words or concepts to argue that point but it's a very interesting way to think about it.

At least I now see what you mean.

Is this a form of pantheism? I enjoy throwing a rock, and by your theory, the rock enjoys being thrown. Not because it has an inner life, but because it inherits or represents my enjoyment. Is that a fair characterization of your point? I must give this some more thought. I don't think I agree with you, but it is definitely interesting.

Searle says exactly that, since what your avatar does instantiates feeling in your mind. Intentionality comes from that mind and not from the avatar. Likewise, Ms Pacman makes no choices on her own, since the intentionality comes from the mind (you) who is obviously very much enjoying eating the dots. — noAxioms

But that's his argument against the Chinese room understanding Chinese. He says that we humans provide the meaning or intentionality. He says that the room does NOT have meaning or intentionality.

You are turning that argument on its head, are you not?

Perhaps this is the disconnect. In what way is Searle a physicalist? — noAxioms

I believe I saw him in a video lecture say that he thought there's something about life that gives rise to consciousness, but bit flipping is not sufficient. I might be misremembering or mischaracterizing his position.

Usually the term is used for a physical monist: All physics (including people) operate by the laws of physics, every bit of which is arguably computational — noAxioms

Oh no no no no no. Physics is arguably computational, but also arguably not computational. I argue that it is not. Physics is the wrong word here. Physics is the historically contingent human activity of Aristotle and Newton and Einstein explaining why bowling balls fall down. The ultimate nature of the world itself is not necessarily computational. I assume we're talking about the ultimate nature of the world, and not just our latest theory of physics.

Of course if I stipulated that the physical world is computational and that I am a physicalist, it would follow that I believe the world is computational. But I doubt very much that the world is computational. Computation is far too restrictive. Perhaps the world solves the Halting problem, perhaps it solves some other noncomputable problem, and that's how it manages to work. We just don't know.

.; Searle perhaps posits a different kind of matter that he still labels 'physics', but the physics community doesn't since there's been no demonstration of it. — noAxioms

Not at all. Plain old matter. But not computational. You at least qualified your claim by saying that physics is arguably computational. It's also arguably not.

I'm still disturbed by the things you claim to believe.
Have I claimed beliefs? Do I believe the rock exists independent of me? Do you know enough of my beliefs to answer that? — noAxioms

You succeeded in making me understand your point about Ms. Pac-Man and gave me something to think about. I'm no longer disturbed.

Anyway if simulation theory is true, we're all characters in a video game
No, that's if VR is true. SH is not modelled by a video game. — noAxioms

Ok ... not entirely sure about this. Isn't it the opposite? If my mind is primary and my experiences are an illusion, the illusion-giver, the simulator, may withdraw my reality at any moment. If there's a simulator, they may get bored of providing me with this interesting reality and unplug me, and I'll cease to be.

And if VR is true, the same thing might happen, but my untethered mind will remain, but devoid of experiences.

Have I got that right?

ps -- When I play chess, do you claim that the chess pieces care whether I win or lose? I'm the one who cares. They're just wood or plastic, or bit patterns in the computer.

I reject your Ms. Pac-Man thesis as well as your misinterpretation of the Chinese room. [But at least I now understand it!]

The humans care. The objects in a video game or chess game don't care. The Chinese room doesn't understand Chinese.

You are assuming a non-realist view of mathematical entities again. You can still have Euclidean and non-Euclidean facts in the world as different facts just like algebra and calculus are different facts. Many philosophers think mathematical objects are real objects that exist outside of space and time. — Lionino

The world can not be simultaneously Euclidean and non-Euclidean.

Algebra and calculus can both be true of the world. They don't contradict each other.

Not sure I'm following your analogy.

You can't have facts about the world that are in conflict with each other.

What does the realist say about that? I did a quick lookup of mathematical realism and it has nothing to do with the relationship of math to the world; rather, Google says, "Mathematical realism asserts that mathematical objects exist in the abstract world, and that a mathematical sentence is true or false, depending on whether the abstract world is as the mathematical sentence says it is." The Wiki article on the philosophy of math takes a similarly abstract view of mathematical realism. Nothing to do with the physical world.

That's a lot more subtle than saying that realism believes that math is literally true in the world.

I make no claim to expertise on these matters, actually I admit to ignorance. But I don't think you are using mathematical realism in the same sense as Google and Wikipedia.

As if reality is the limit of our theories.
— fishfry
Since I don't know what "reality" means in its philosophical sense (which I designate by "Reality", but I do know, roughly, what you mean by "the limit of our theories", I would prefer to say "The limit of our theories is Reality". I'm of the school that teaches that the philosophical sense is metaphysics, and nonsense. But, since I arrived on these forums, I've had to recognize that, in philosophical discourse, "Reality" is a term in regular use and with some level of common understanding. — Ludwig V

Reality is what's really going on in the world. Not sure why you regard that as problematic.

A bat has a particular view of the world, as does an ant, as does a sea slug. None of them, and that includes us, know what ultimate reality is. Not sure what your objection or concern is with this idea.

It's still a bit broad brush. I can understand it in the context of the inescapable inaccuracy of measurement in physics, etc, contrasted with the preternatural accuracy of (many, but not all) mathematical calculations. It's a version of Kant's regulative ideals and gives some content to phenomena/noumena and an explanation how they might be related. — Ludwig V

Physics is inaccurate, but what if it's wildly inaccurate, as inaccurate as an ant's view of the world relative to the real world? We like to think that we're "close" to knowing reality because our physics works so well, but that's arguable.

Well, I would certainly want to get him to explain what he means by "is". That might slow him down a bit. — Ludwig V

As I understand it, Tegmark believes the world is a mathematical structure, like a group o a topological space.

Intellectuals have human motivations and follies just like everyone else - and some of them would do well to acknowledge that. I understand also that it is irresistibly tempting to explain people's failures to recognize conclusive rational arguments in ways that they will not like. But one needs also to understand that can be a trap. Hence Plato turned a classification of the philosophers he disagreed with into a term of abuse - "sophist", "rhetoric". You may have noticed that I'm engaged in some discussion with Metaphysician Undercover about this issue in relation to Zeno. — Ludwig V

I have the worst habit lately of only responding to my mentions and not reading the rest of these threads.

They, and, apparently @noAxioms cannot believe that Zeno believed his own arguments - and that's not an irrational response because they are incredible. Nevertheless, I can't believe that they believe that. It's not easy. But I think it is important not to follow Plato's example in this respect. — Ludwig V

Zeno's arguments are far more sensible than Tegmark's.

The example is so familiar to me that I thought it would add clarity. To the extent it got in the way, perhaps I should rethink how I present the idea.
— fishfry
I don't think there was anything wrong with your explanation. There's no such thing as the bullet-proof, instantly comprehensible, explanation. On the contrary, it helps to allow people space to turn what you say round and poke it and prod it. It's part of the process of coming to understand a new idea. — Ludwig V

I ended up spending all my time explaining the ordinals and that detracted from my resolution of the lamp.

The lamp's defined at each point of the sequence, but it's not defined at the limit.
— fishfry
Quite so. It's a sequence, but also a chain, because each point of the sequence depends on its predecessor. The reason it's not defined at the limit is that we can never follow the chain to its' conclusion - even thought the conclusion, the end, the limit, is defined. — Ludwig V

But the limit isn't defined in the lamp problem. There is no limit to 0, 1, 0, 1, ..., and therefore no terminal state that's more natural than any other.

It seems paradoxical, because the limit is established before the chain can begin. The first step is to define the limit and the origin; that gives us something we can divide by 2 - and off we go.
This may not be mathematics. But I do maintain it is philosophy. — Ludwig V

There need not be any relationship or rule that defines the elements of a sequence, but the sequence can have a limit. But in the case of the lamp, there's no natural conclusion because 0, 1, 0, 1, ... has no limit. I'm repeating myself but that's the point of the lamp. There's no natural terminal state.

The consequence is that the series "vanishes" if we try to look back from the "end". It's existence depends on our point of view. I don't suppose that any mathematician would be comfortable with that, but I plead that we are talking about infinity and standard rules don't apply. — Ludwig V

I'd say that the standard mathematical rules for dealing with infinity are perfectly clear, and do apply. It's vague, handwavy imaginings that don't apply.

I'm asking, in what sense? Surely math has never been fixed. It's always changing. It's a human activity.
— fishfry
Originated as, yes. But that doesn't restrict how math is seen today.
— fishfry
I think you are agreeing with me. Abstract today, applied tomorrow. Or often the reverse. We invent new abstract math to help us understand some real world application. It goes back and forth.
— fishfry
I agree with all of that. But I think it is very, even hideously, complicated. — Ludwig V

Not following at all. Math is constantly changing, has been for thousands of years, and is changing even as we speak. New math papers are published every day.

Are you referring to some kind of Platonic math, God's math textbook? Is that what you mean by fixed?

I am not understanding your point, I don't see what's complicated about this at all. Can you explain what you mean?

It seems to me that we should always be specific about what is fixed and what is not. There may be disagreement about what goes in to which classification or what "fixed" means. But to say "math" without specifying further leads to confusion. — Ludwig V

I don't feel confused, so I must not be understanding you.

Arithmetic, for example, is (relatively) fixed, though it may be modified from time to time. The inclusion of 0 and 1 as numbers is an example. Number theory might count as another example - I'm not sure about that. But once the methods of calculation are defined, they are fixed and the results from them are fixed as well. — Ludwig V

Yes, sure, a fixed body of knowledge evolves. But that body of knowledge is added to every day by every math journal and university colloquium.

One could say, however, that both methods and results are discovered rather than defined, because there are ways of demonstrating whether a particular procedure gives the right result or not - through the application of the results or through the application of criteria like the consistency and completeness of the system. — Ludwig V

Not only new results, but new ways of thinking are constantly introduced.

Euclidean geometry is similar, so far as I'm aware.
Algebra, calculus, non-Euclidean geometry, infinity theory are all additions to mathematics, rather than replacements of anything. It is almost irresistible to speak of them as developed or created rather than discovered, but since they share something with arithmetic and geometry, there are some grounds for speaking of them as discovered, because they were always possibilities, in some sense. What is it that is shared? The best I can do is to say something like logic - a sense of what is possible, or permitted. — Ludwig V

So are you thinking of God's math book that humans slowly learn about? Well maybe that is fixed. So you're just regarding math as a Platonic body of knowledge that is "out there" somewhere. Is that what you mean? I'm not saying it's not true, I'm just asking if you're taking a Platonic viewpoint.

This is not irrelevant to this thread. Once we have realized that "+1"
can be applied to the result, it would not be wrong to say that the result of every step is fixed, whether or not we actually do add 1 to the 3,056th step. The result of each step is "always already" whatever it is. (I think it derives from Heidegger, but that doesn't prevent it from being helpful.) It captures the ambiguity between "+1" as something that we do and something that is done as soon as it is defined, or even before that.
As a result of the simple recognition of a possibility, we find ourselves plunged into a new and paradoxical world. I mean that it is simply not clear how the familiar rules are to be applied. Which makes it clear that we have to invent new ones - or are we discovering how the familiar rules apply or don't? I don't think there is a determinate answer and "always already" recognizes the ambiguity without resolving it.
When we refer to a step in the series, are we talking about something that we do (and may not do) and which actually takes time or something that is "always already" done, whether we actually ever do it or not? — Ludwig V

I believe I lost track of what this paragraph referred to, sorry.

But you just proved P2 yourself! You agreed that under the hypothesis of being able to recite a number at successively halved intervals of time, there is no number that is the first to not be recited.
— fishfry

I agreed that if P2 is true then C1 is true, as I have agreed from the beginning.

This doesn't prove that P2 is true. — Michael

You yourself proved P2 true, and I don't understand why you aren't even engaging with my argument supporting that claim.

Certainly mathematics is, in a sense, fixed. — Ludwig V

I'm asking, in what sense? Surely math has never been fixed. It's always changing. It's a human activity.

Do you believe that God's math book is already written? Is that how you are defining math, as a Platonic ideal "out there" that exists even before we've discovered it?

But what we are talking about it is applied mathematics. It seems pretty clear that arithmetic and geometry originated in severely practical needs of large empires. — Ludwig V

Originated as, yes. But that doesn't restrict how math is seen today.

But it does seem to have taken off on its own, as it were, as a theoretical enterprise. Here, we are talking about applied mathematics. — Ludwig V

I don't recall stipulating to any such restriction.

In any event, what is the difference between abstract and applied math? Only time and historical contingency. Non-Euclidean geometry was abstract and useless in 1840, and it became applied (to general relativity) in 1915. [Someone complained a while back about my phrasing. I could say Riemannian geometry, since there are other flavors of non-Euclidean geometry. I'm making a different point which I hope is clear].

Likewise number theory, regarded as supremely useless since the time of Diophantus; and now the basis of public key cryptography, the basis of modern digital commerce, since as recently as the 1980s. That's 2200 years of uselessness, only to become supremely useful.

The difference between abstract and applied math is time and history.

I think what fishfry means to say is that mathematics is the way the world is represented to us. That's the point of the comparison with what sound is to a bat. I would rather say that mathematics is the way we represent our world to ourselves. — Ludwig V

I could have said that. I could have said your formulation too. My only point is that how humans model the world and the world itself are two different things. Aren't those Kant's phenomena and noumena? I'm not a philosopher but that's what he meant, right? Humans try to explain the phenomena. We can't know the noumena. That's Plato's cave analogy again.

It's true that the mathematical techniques we use are fixed - though we also develop new techniques, as in 17th century calculus or non-Euclidean geometries. But we have to work out how they can be applied to specific phenomena. — Ludwig V

I think you are agreeing with me. Abstract today, applied tomorrow. Or often the reverse. We invent new abstract math to help us understand some real world application. It goes back and forth.

So you DO have axioms :-)
— fishfry
I hold them to be true out of necessity, not because they necessarily are. Another one then I forgot to list: No magic. "I don't know, needs more investigation" is a far better answer than the god of the gaps explanation. Every time one of those open questions finally gets answered, it's never magic. The magic explanation is thus far on the wrong end of a shutout. — noAxioms

Why do people who reject God accept the Great Simulator? The GS is just God constrained to computability.

Why does the sun cross the sky each day? God carries it thus. What's have we learned since? Clue: It isn't that Earth goes around the sun, since it doesn't do that each day, yet that's the rebuttal typically given. — noAxioms

One could write out the equations for a geocentric solar system -- hmm, I mean Earth system! -- and get the exact same predictions. It's just a matter of perspective and utility. A frame of reference problem.

So: "God does it -- harumphh, you ignorant peasant!" but "Oh the Great Simulator does it, you're a wise TED talker." That's the logic I disagree with. Simulation theory is just theology in a black turtleneck and jeans.

Likewise Tegmark's mathematical universe. An even more obvious troll.
— fishfry
You may not buy into Tegmark's suggestions, but that doesn't make him a troll. — noAxioms

I laid out my case that Tegmark is a troll here ...

https://thephilosophyforum.com/discussion/comment/904316

Saying that the world "is" math rather than is described or approximated by math is such a massive category error that there is no possibility that Tegmark isn't trolling us.

I don't agree with him either, but I still read the book and find it revolutionary. His attempts at empirical evidence are completely faulty, but one is expected to pony up evidence to bump the idea from interpretation to actual 'theory'. He doesn't call it that, only calling it 'hypothesis', but even that word implies falsifiability. — noAxioms

How does he get around the category error problem, confusing the map with the territory, or the program with its execution? My hat is off to you for having read the source material.

Is your web browser passing judgment on the opinions you post to this site?
— fishfry
Matter of time. Right now it only passes judgment on my choice of sites on which I choose to post my opinions. — noAxioms

You give your browser far too much credit. It passes no judgment on anything. You are the one who has judgment. The browser just flips bits on your computer to implement certain communication protocols that it uses to exchange data with a web server. And the data has no meaning, it's just a long string of bits. Humans give it interpretation and meaning.

Does Ms. Pac-Man experience pleasure eating white dots,
Obviously yes. — noAxioms

You can't believe that. Are you joking with me or making some kind of point I'm not understanding? It's not possible that you believe that literally.

As a Searle fan, you should know this. — noAxioms

Searle's rolling in his grave and he's not even dead. That's not true. Searle denies that bit-flipping instantiates intentionality or feelings like pleasure.

The question is does Blinky experience pleasure eating Ms Pacman? Blinky is an NPC. Ms Pacman is not. The answer there is no only because such experience would provide no benefit to Blinky, so there's no reason for it to be there. This would not be the case in Bostrom's sim, were it possible. — noAxioms

You are totally trolling me. I am really puzzled by this. You can't mean what you are writing.

the brain does not operate by the same principles as a Turing machine.
Agree, but a physicalist would say that the brain could be implemented by a Turing machine, just as it could be pencil and paper. Arguably, the latter might actually be more efficient. Turing machines are not designed for practicality. They're a model of computability. — noAxioms

No. A computationalist would say that. A physicalist, which Searle is, would say that something unique to life implements consciousness, but whatever it is, it's not computational. Or maybe he didn't say that, but I did. That's what I tend to believe. Physicalism but not computationalism.

I'm still disturbed by the things you claim to believe.

Ms. Pac-Man experiences pleasure? What on earth can you mean?

Anyway if simulation theory is true, we're all characters in a video game in an alien bar, and they're about to run out of quarters.

Didn't I ask you about this several posts ago? Ok, Euclid's line.
— fishfry

Sorry, I didn't appreciate the point when you first mentioned it. Yes, I'm starting from classical Euclidean geometry. — keystone

Ok. I wanted to say that I'll stipulate to your non-rigorous conception of a continuum of being made of tiny little continua "all the way down," with no need for actual points, if that's your idea. I think this is what Peirce is getting at.

In any event to save us some time, I'll stipulate to your vision, even if it's a bit contradictory and not totally clear.

So then what?

And by the way, what is this "+" symbol? Have you defined it? Is this the standard + of the rational numbers?
— fishfry

Yes. Formally the arithmetic is performed as described here (https://www.sciencedirect.com/science/article/pii/S1570866706000311) — keystone

LOL You are committed to that idea, I'll give you that.

but informally it's performed using the standard method we teach kids. The formal and informal results are equivalent. — keystone

I'll stipulate to arithmetic on the rationals, I think we can agree on that.

what does the notation (0,1) mean?
— fishfry
uu
It describes the line's potential. — keystone

Ok. With my earlier stipulation, I could say:

(a) What? That makes no sense; or

(b) I sort of get it. The line contains a frothing sea of tiny little micro-continua that are not points. Is that about right?

I'm going to provide a shorthand answer involving real numbers that I don't want you to take literally. If this explanation lands, great, otherwise forget it.

No points exist on lines, including the unit line (0,1). To put it another way, there are no 'actual points' present on that segment. (Actual vs. potential is discussed below).
Cutting line (0,1) in two will introduce an 'actual point' between the two resulting line segments. That point will have a rational coordinate between 0 and 1. — keystone

Well here you are in trouble. If you allow "cuts" then à la Dedekind we have the real numbers. But you don't want to go there so ok. There are cuts but not so many as to allow the reals.

In my last post, I noted that -inf and +inf are not 'actual points' but rather are used as helpful shorthand. I should have called them 'potential points'.
With a similar shorthand, we can say that on line (0,1) exist 2ℵ0
2
ℵ
0
'potential points', which have real number coordinates between 0 and 1. — keystone

Oh my. You are now allowing the reals? Ok. Maybe that's good.

The rational 'potential points' can become 'actual points' through cuts.
The irrational 'potential points' are permanently confined to their 'potential point' status.
I want to reiterate that 'potential points' don't actually exist. They're just a fiction that may help us comprehend the potential in continua. If you don't think potential points are a useful concept we can just drop.
The interval "(0,1)" describes the potential of the corresponding unit line. — keystone

I sort of get your thinking. Not sure where you're going but I'll stipulate to all this, even with the vagueness.

Of course all mathematical entities are fictional, so I can't see what the difference is between and actual and a fictional point. Once you stipulate to fictional points, they become actual by virtue of being used and accepted. Just as negative numbers and imaginary numbers once did.

The life cycle of a mathematical idea is is:

Regarded as impossible ---> Fictional but useful ---> Normal everyday stuff.

Same path taken by non-Euclidean geometry. Impossible, then Riemann's curiosity, then Einstein's platform for general relativity

So once you admit a "fictional" entity you might as well grant it actual status, since you will eventually.

Since your intervals are entirely made up of rationals, the total length must be 0. Where is the extra length coming from?
— fishfry

The length of a line comes from its potential. — keystone

But here's the thing. I said earlier that in the standard unit interval, the length is stored in the irrationals.

You are saying the exact same thing, but changing the name of irrationals to "fictionals." I don't see how that changes anything. You just changed their name but they're the same irrationals.

Sometimes it’s a bit frustrating when my explanations don’t connect, but this conversation is exactly what I need right now, so please don’t feel bad. I'm very appreciative that you've stuck around. — keystone

Ok. I'm adopting a less rigorous and more intuitive sense of what you are saying.

But you have also met me more than halfway. You have agreed finally that there ARE irrationals on the line, and that they carry, or store, the length. You just call them fictionals instead of reals. But they are the same thing.

[==== next post ====]

Path Length = Length of Lines + Length of Points
Path Length = Length of Lines + 0
Path Length = Length of Lines

So referring to row 3 of that figure...
Path Length = Length of Lines
1 = 1/2 + 1/4 + 1/4 — keystone

I'll stipulate that the length is stored in the fictionals, which I'll continue to think of as the reals till you claim otherwise.

Tegmark's trolling. And the world is mathematical to us just as it's sound to a bat. The world does whatever it's doing. We do the math.
— fishfry

That is the view that mathematical is somewhat of an empirical endeavor. Many disagree however, and think that mathematics is something fixed and representative of the world. — Lionino

Surely few if any people believe math is "fixed." Math is historically contingent and changes all time time, with a massive volume of new papers published every day.

If you are referring to some kind of Platonic math that's already known by God, that we are just discovering, that's an entirely different discussion.

Am I understanding you correctly?

Besides, math can't "represent the world," simply because there are Euclidean and non-Euclidean geometry. They can be used to represent the world; but they can't both be true, hence they can't both "represent the world." They can only be used to represent the world.

Math can not tell you what's true about the world. It can only be used to model various aspects of the world. That's different.

Given P2, what is the first natural number not recited? I seem to remember having asked you this several times already.
— fishfry
— Michael

There isn't one. I've answered this several times already. That's what it means for me to accept P1.

But you need to prove P2. You haven't done so. — Michael

But you just proved P2 yourself! You agreed that under the hypothesis of being able to recite a number at successively halved intervals of time, there is no number that is the first to not be recited.

This proves that all numbers are recited. This is a standard inductive proof that a high school student should be able to not only understand, but even figure out for themselves. If someone's high school didn't teach them mathematical induction perhaps they picked it up in Discrete Math class; and if not, then the writeup on Wikipedia would suffice.

You have proven P2 yourself simply by agreeing that there is no first number that is not recited.

If no number did not get recited, then they all did.

So we're back to my post here:

a. I said "0", 30 seconds after that I said "1", 15 seconds after that I said "2", 7.5 seconds after that I said "3", and so on ad infinitum — Michael

[details omitted]

You accept that (b) is impossible but you claim that (a) is possible. You have to prove this. P1 doesn't prove it.

Let's focus on one thing at a time. Regarding your example of counting the natural numbers backward, or letting the sequence get smaller when time goes forward, the 1, 1/2, 1/4, ... idea; I have repeatedly asked you if you understand and agree that any interval of real numbers containing the limit of a sequence, necessarily contains all but finitely members of the sequence.

I need you to understand that in order for me to explain to you how the backwards counting puzzle is resolved.

Since I've asked you several times to just tell me, yes or no, do you understand what I said, and you have repeatedly ignored me, conversational progress can not be made on this point.

So let's stick to the inductive proof, in which you yourself proved P2 is true. Let's get back to the backwards counting example after you tell me, yes or no, do you understand the property of limit points of sequences that I keep asking you about and that you keep not answering.

Believe it or not, that's an incredibly helpful remark. — Ludwig V

Thanks.

Not only do I understand and agree with it, but it also enables me to get a handle on what metaphysics is. Sorry, clarification - I am referring to the whole sentence, not just the last five words. — Ludwig V

Well metaphysics is just "What is reality?" And it can't exactly be our math, because we can see that it wasn't quite what Newton wrote down, and in the end it won't quite be what Einstein wrote down. It's actually kind of strange that math doesn't exactly describe reality, but so well approximates it. Our theories get better and better but never get there. As if reality is the limit of our theories.

Or worse. Our math is like the bat's echoes. Just the only tool we have to understand the world, but greatly limited. And we think we know everything.

I had to look Tegmark up. — Ludwig V

Goes a step farther. The universe isn't just described by math, it "is" math. Which is a category error so massive that Tegmark must be trolling. The equations of motion describe the planets, they aren't the planets themselves. The map is not the territory. Just as the source code for a program must be executed on hardware in order to do anything.

Tegmark must be trolling. There is no other explanation. That so many take him seriously is a good reason to be skeptical of experts, celebrity scientists, and "public intellectuals."

No disrespect, but he does illustrate the observation that intellectuals are not exempt from normal human desires for fame and fortune, no matter how much they protest the contrary. There's also a normal human pleasure in astonishing and shocking the tediously orthodox Establishment. — Ludwig V

We're in agreement. Bostrom (we're all sims) and Tegmark (we're all mathematical structures) must be enjoying themselves tremendously. Most likely when they write serious stuff, nobody pays attention.

That's why I prefer the 1/2, 3/4, 7/8, ... example. Same structure in more familiar clothing.
— fishfry
Yes, we had that discussion as well. You may remember that I had reservations. Same, but not identical, structures, I would say. But I don't expect you to like it. It doesn't matter until it becomes relevant to something. — Ludwig V

Well it's relevant to the Thompson lamp. It's a mathematical model of a sequence with its limit point adjoined. The example is so familiar to me that I thought it would add clarity. To the extent it got in the way, perhaps I should rethink how I present the idea.

My apologies. I should have restricted my remark to those who dream up paradoxes. — Ludwig V

Mostly philosophers who prefer to indulge in the vagueness of word games rather than the precision of math. But I concede that many smart people take these puzzles seriously. I respect that, but for some reason the fascination eludes me.

The lamp's defined at each point of the sequence, but it's not defined at the limit. There's no way to make the sequence continuous, se we are free to make the terminating state anything we like. There is no natural continuation. That seems perfectly clear to me. I don't know why it's not perfectly clear to everyone else. I actually have a difficult time seeing the other points of view.

Though perhaps even that is wrong. They may be exploiting the rules themselves, rather than merely breaking them. The mathematical rules for infinity don't seem particularly helpful in resolving these problems. — Ludwig V

Maybe we'll get some new infinitary physics some day.

unless you mean the original line of Euclid, "A line is breadthless length."
— fishfry
Yes!!! I agree with Euclid's definition of lines and points. I appreciate that he provides foundational definitions of both as separate, fundamental entities. Thanks for pointing this out. — keystone

Didn't I ask you about this several posts ago? Ok, Euclid's line.

What is a line? What does the notation [0, 0.5] mean?
— fishfry
Euclid also said that "The ends of a line are points." When I describe a path as 0 U (0,1) U 1:
(0,1) corresponds to the object of breadthless length and
0 and 1 correspond to the points at the end. — keystone

Ok so you are doing classical Euclidean geometry (not modern Euclidean geometry, please note).

It seems that some people intepret Euclid as saying that a line without endpoints extends to infinity. I do not think this is necessarily the case. While (-inf,+inf) is a valid line, I believe (0,1) is also a valid line in and of itself. — keystone

Euclid would not recognize that notation; and at this point in our conversation, neither do I. You have variously stated that (0,1) contains only rationals, or that it may even be empty.

In view of my new understanding that by line, you mean Euclid's line, what does the notation (0,1) mean? Euclid did not have numbers as we know them.

Please give the following figure a chance as it captures a lot of what I'm trying to say: — keystone

Utterly baffled. Utterly. Baffled. No idea what it means. 0, 0 + 0, 0 + 0 + 0, no idea what I am supposed to glean from that. And by the way, what is this "+" symbol? Have you defined it? Is this the standard + of the rational numbers?

I feel terrible ignoring these diagrams that you put so much work into, and that hold so much meaning for you. I wish I could be more helpful. I don't mean to just continue to snipe at you. It pains me. I just don't know what you are saying and have no idea how to respond.

I believe that someone even as intelligent and knowledgeable as yourself is not qualified to discuss the bottom-up philosophy of a continuum because it is flawed. — keystone

I never claimed to be able to discuss the philosophy of the continuum. On the contrary, I've admitted that I can't. Except, that I know a bit about the real numbers, and they are the standard mathematical model of the continuum. And that counts for something.

I'm 100% certain you have the capacity to understand, discuss, and criticize the top-down philosophy. — keystone

Possibly, but not the inclination. If I could dispatch a clone, I'd have him read Peirce. I'm not a philosopher of the continuum. I'm not a philosopher at all. I only come to this forum to clarify people's mathematical misunderstandings. And it's a full time job :-)

You're right, I did say that the endpoints were necessarily rational numbers. (-inf, +inf) has no endpoints. While there are scenarios where it is useful to include points at infinity, for this discussion, let's agree that the points at -inf and +inf are not real points. I'm only using infinity as a shorthand. I should have been clearer. — keystone

Ok. So far, your line is Euclid's original line. Leaving undefined, your notation (0,1), which you have repeatedly pointed out is NOT the open unit interval of real numbers.

ps -- Ok I took another look at your picture. You correctly note that the sum of the lengths of the points is 0. But then you say that the sum of the lengths is 1, and I'm not sure how that follows.

Since your intervals are entirely made up of rationals, the total length must be 0.

Where is the extra length coming from?

I'm willing to let you say that the length of the interval (0,1) is 1 even though it's only made of rationals. I'll stipulate that for sake of discussion, even though it's hard to understand how it works.

But what does it all mean? I'm lost and dispirited. It's not my role in life to feel bad about myself for endlessly sniping at your heartfelt ideas.