Future of Coding - Is the Whole Universe a Computer™?

Episode Date: January 5, 2025

"Is the whole universe a computer?", ask Jack Copeland, Mark Sprevak, and Oron Shagrir in chapter 41 of the book The Turing Guide. They split this question in two, first asking whether the universe itself is a computer, then whether the universe could even be computed. These are lofty, unanswerable questions, sure, but they encroach on our territory — philosophy, automata, nonsense. So, in our usual reverent style and with attentive pacing, the three of us explore the paper, the questions, the answers they choose to highlight, and even share a few perfectly reasonable answers of our own. Links $ Patreon In no particular order: Carl Sagan What it means to be open was Lu's talk at Heart of Clojure Rudy Rucker Is the brain a digital computer? by John Searle The Plankth of time. C is Not a Low-Level Language by David Chisnall The Connection Machine!! Komgloverav… Komolgorov… Outer Worlds? Outer Wilds? The Witness Music featured in this episode: No! That's a spoiler. No way I'm telling you. ! Send us email, share your ideas in the Slack, and catch us while you still can: Carl: Mastodon • Website Jimmy: Mastodon • Website Lu: Mastodon • Website See you in the future! https://futureofcoding.org/episodes/74Support us on Patreon: https://www.patreon.com/futureofcodingSee omnystudio.com/listener for privacy information.

ivan loves to lie to us about what things are working in his audio setups he has a very good reason to do so definitely not uh no incentive whatsoever to lie to us about it i even sent me a dress and he said where this is important for the audio setup true story i told steve krauss to try and find an environment like a closet to record in to dampen all the echo because he was in a very echoey room and he love him sent me recordings wearing a blanket and not wearing a blanket to see if that made a difference like just like sitting in his normal desk but like with a blanket over him yeah uh to see if that helped and unfortunately it did help nice yeah yeah i mean you can try to do some janky stuff impractpractical. Have you met Steve Krause in person? No, no, not yet.

That's crazy. I finally met him in person a couple months ago, and I apologized for parodying him on one of the earlier things. That's what happens when I meet most people at this point. You apologize for parodying them. Parodying them. people at this point this is you apologize for parading them pirating them so not related to programming at all i've been playing the new 2d zelda game which i've never been a 2d zelda person but this one looked really cool the zelda the zelda zelda game yeah have you have you all yeah the zelda's yeah where you make beds and trampolines and pile them on top of each other. Yeah. Uh-huh. Yep.

That looks fun. Yeah, yeah. It reminds me of Baba is You for some reason. Is it like that at all? No, it's not like Baba is You at all. If it was anything, it would be like Scribblenauts or something, right? Because there's all the different things in the world that you can use to solve problems

and then you only end up using one of them, just like scribble knots that's um but it's uh it's actually pretty neat i i was i think it's a good mix of this like open world but linear that i think breath of the wild didn't do well and so i'm hoping that like the next generation of zelda games kind of borrows more from this because it feels feels more traditional it feels less just huge open world with nothing to do in it besides play around with the physics oh i love that though breath of the wild tears of the kingdom i couldn't get into tears of kingdom because it just felt too much like doing the breath of the wild stuff but with fancy machines yeah but the thing is in my head those two games it's like one game you know it's like portal one and portal two it's

this one game portal one's the demo it's like mario galaxy one and mario galaxy two no mario galaxy one is just like the intro you know that's those comparisons actually hold up like portal one portal two i haven't played mario galaxy 2 it's like a complete package it's really good but it's not just that it's it's also like a team does something that's kind of new and different and challenging and it's well received and then they get the opportunity to go back and give it a like another run with everything they've learned from the first one and so you get this like much more polished experience where in both portal 2 and and tears of the kingdom you get like much better writing and much better use of music and sound design and much better like mission structure and like all of the

sort of bits that kind of weave around the the core concept are much more fleshed out and they find ways to work them together and so it's a it's a more cohesive experience which is like i enjoy that but i also enjoy just the like rougher smaller sort of like you know awkward first steps i think they're both kind of beautiful in their own way what i wanted from the breath the wild the tears of kingdom was like ocarina of time to majora's mask uh like significantly weirder yeah like okay like you can just definitely tell i feel like it was a little bit no it was the same game but with new ways of moving around yeah but they were so good

yeah but it wasn't weird i i and they like removed the mute like i i feel like okay tears of the kingdom breath of wild great games yeah fantastic games not great zelda games yeah like whoa i don't know no true zelda it's like no true scotsman yeah no but it's like no true zelda really if you think about it there's a lot of things about the zelda franchise that i feel like were left out like i like for me one of the things i loved growing up and like i i think it was a really important part of a lot of the at least 3d zelda games for sure because that's mostly what i played we like the musical aspects of it like in-game music not like just the soundtrack the soundtrack of course was beautiful yeah but like

the ocarina the wind waker that you know like having musical instruments as a part of the gameplay and having some magical system with music is just gone in those two games was really just those two though right it was like ocarina and majora and then uh like wind waker wind waker has has music as well uh yeah but most of the other ones don't like uh link to the past didn't i was 3d games is what i'm talking about and uh so link to the past was my first zelda game so to me like the 3d games aren't true zelda games you know whereas breath of the wild and tears the kingdom they were like way they felt way more like link to the past especially like yeah like link to the past is so open and

and it really does feel at times like you don't know where to go what to do but you can just explore which is pretty cool the thing i do miss from uh especially link to the past is like you have you had so many items in that you could you had like so many items and i guess you kind of do in these but they're not really it's kind of different you have to use the environment yeah well tears tears has the thing where it's like oh if you get like the octa rock balloon or whatever um you can like attach that to something and make it float or if you get like the you know the jelly from the electro choo-choos or whatever that can be a source of electricity and like they they did i think in both uh breath but especially in tears

breath find ways for yeah for those but boat boat boat tour boat uh and toe took yes they found ways to um to make the like you know the the items that you get as like pickups from objects in the world play a role in the systems of the game uh but yeah they don't have the the variety of like equipment that link to the past had linked to the past had like all of these you know objects that feel like oh i've now like sort of like a metroidvania right like i've gotten this new ability and it lets me get through all of these gear gates and change my relationship with the world around me in the way that I don't think Breath of the Wild

and Tears had as much. Yeah, I think trying to make everything always open and available often makes it so that you can't really have the story progress in a way that makes more things open and available. Right. And so that's one of the things that I feel was lacking.

And I think Echoes of Wisdom does a pretty decent job of like having this, like it is open, but there are certain places that are closed off based on abilities. And like, you can go beyond, like, I ended up wandering off into an area where the the creatures there i can't end up using their abilities yet i killed one and i got its ability but like i can't use that ability interesting and so it's like it's open and you can totally go there and it was like a really silly fight that i like cheesed and got and then i was like oh i can't even use this yet so like i think it's a good mix of of trying to you can definitely see a lot of the inspiration from like breath of wild and tears of the kingdom brought

into this game but uh really enjoying it i mean the great thing about breath of wild and tears of the kingdom and maybe this one is that like the whole world in those games like the whole universe is like this big simulation you know it's like this big computation right so so like do you see which brings us to our paper today wow i feel like the new meta is lou is always trying to beat me to the segue but was that what you were thinking of or were you just wanting to chat uh no i was just literally chatting it's fine So now I feel a bit bad. But we can't... I need to tell you, I went to Heart of Closure, the conference, a couple weeks ago. And it was a really, really good conference, by the way.

That's like the best conference I've ever been to. It felt really special. Nice. But the reason I brought it up today is that I was actually pleasantly surprised by how many Future of Coding listeners I bumped into. Lots of people said that they listen to the show. It's awesome. I think for some reason that that community in particular, because I go to a lot of conferences but uh normally there's like maybe one person this one there was loads and i received lots of feedback on the show i'm i'm gonna guess it's because closureists are masochists and i insulted jimmy for being a closureist very early on in the

history of the show and so closureists sort of whispered oh look there's a new podcast that'll say cheeky things about rich hickey come on everybody get in right you see that definitely yeah i think the crowd at this particular conference would enjoy that but like and i did someone taught uh well actually a whole group of people taught me closure on my final day they helped me to make sand in closure nice and sort of like watched my pain with parentheses but But it was super fun. It was like really great. It was like a team effort. It was kind of fun to have like the masters of the community

teach idiot me how to close brackets, you know? So was that your first time writing Lisp? Not Lisp, no. I've done quite a bit of Lisp but not Clojure. But like I feel like the times I've, like I've written, I've made Lisp, you know, but that feels very different. I've only ever done it for like toy purposes, not actually building, you know, like where the output is just more Lisp.

You know, here it was actually, let's make a program, you know, let's make something real. Does that make sense? Oh, totally. Yeah. It felt very different, to be be honest even though it's not that different the the first version of hest was in lisp or sorry not lisp closure specifically really really yes that sounds tricky uh well i don't know it was it was good i i over engineered it or like over architected it i turned into an astronaut and then that doesn't sound it like you sound like you at all not at all no i mean it's crazy because

it makes you think differently like i i was like yeah thinking in javascript as i was writing closure and then it all went wrong um but it was it was super fun but yeah i got lots of people telling me things about the show and i thought i could maybe share some of that with you too i would love that that's a brilliant idea i'll start with the most commonly given uh statement that people said oh and the most common thing people said was are ivan and jimmy okay so apparently the bone these are people not only who listen but are patreon subscribers no no no no i asked oh did you listen to the the bonus episode you know where where you two you know went went against each other but they said no no no they hadn't

listened to that one yet but they but now i told them about that they're great i'm gonna listen yeah though the spiciest bits all got cut out so actually this time not just as a joke yeah actually actually i said uh no no no they're fine they're fine you know it's just you know it's just putting on a character for the show you know and so i remember there was one time where i said that and the other person said uh well it doesn't seem like it's just for the show it seems real that's and it's so funny because like i had a friend growing up uh like in high school where i would get that question all the time with the with my friend tanner they'd be like are you and tanner okay and i'm just like what what do you mean you all were just like arguing so vehemently like i just have fun doing that uh yeah same i think people enjoyed it i think people were like enjoying uh the drama

well we'll see about that and um but they did say that it was nice to have me there to sometimes uh bring it back down to earth uh-huh yeah and obviously they said to my face i don't know anyway so the next time someone asked me if you two are okay i said no i said it's terrible i'm trying to stop them getting a divorce and um they seem to enjoy that so you're sowing the seeds of chaos amongst our listenership that's that's brilliant i i'm fully endorse it the other thing this will be quick you know there's not that much but the other thing uh people said is like they're so long yeah they're so long yeah they are and there was there was one person in particular that said uh you know like ivan man he really likes to talk around the point you know there's a lot of like

how should we talk about the paper like this and like that? And he said like, come on, man, just do it. Yeah, no, I a hundred percent agree with this person. And then, you know, I told him that you once admitted that when you edit it,

it gets longer instead of shorter. Well, and he put his head in his hands. Right. Anyway, someone who was listening to that conversation came up to me afterwards and he, he told me, I actually like all the talking around, you know, like it's very meta. So like, thank you. someone who was listening to that conversation came up to me afterwards and he told me i actually

like all the talking around you know like it's very meta so like thank you person this is why like feedback you have to you know it what it what you have to see is does it reflect what you wanted to get out right and i i'm totally down with this feedback like yeah we talk around up things we do talk forever uh-huh we uh you know we can't stop talking uh i will say one of my friends who listens to the show uh matt has said that like hello lou you were the best addition oh well yeah uh that every time we're ivan and i are our big diatribe uh matt's always happy when when lou jumps in uh so nice but to keep that i need to let you go off a little bit because you know it makes me look good right exactly so the other thing

someone's a couple people said was like you guys are really harsh on brett victor and then and then they actually clarified they said uh no actually you in particular blue you are really harsh on brett vector and i said hey you know like hey you know like i actually was supporting him a lot you should have listened to the cut content from the bonus episode but um which are you know patreon.com future of coding okay okay separately i would add that some people i think sort of took uh this perception of us or me being harsh which i don't think is true right and they they had a full-on like rant about this was a number of people ranted to me about like uh the future of coding's tendency to sort of hold particular figures up on pedestals and worship them and they're typically men and it's not you know it's not just brett vector

it's other people too so i don't know it's a that people notice that the final thing someone told me was that the intercal episode was crazy because that was the best one so good it's also the only episode um so i'm glad they found it and listened to it the only episode yeah it's sort of like we we have a patreon there's only one tier oh wait what yeah we have a patreon there's only one tier you should no no go and pick your your one the one the one tier that is the only tier. Wait, you think if they... I didn't say tier, I said episode. Yeah, there is only one episode.

It's the Intercal episode. For all things future of coding, there's only one future. There's only one episode. There's only one tier, right? There's only one host. If you don't know what we're talking about, just go and sign up on the Patreon. Yeah. Right? The thing about the intercal episode was you know there was uh random interspersing of interruptions um from an

unnamed person um in this episode though there'll be random interrupt uh interruptions by carl sagan so that will make this one even weirder nice what i hope that's i hope that's a real threat that you're actually bringing some some cs energy into this because no no i just had to think of how ivan would edit this one to be like intercal and that would be a a fitting uh interruption here all right cool this is all getting cut out all right so that was jimmy's segue um my segue is i have a new uh coloring scheme that i'm using for this one what is it what is it so traditionally lou your coloring scheme is just everything's green if you like it it's green if you don't like it has it has been known to be that yeah yeah yeah uh jimmy i can't remember

what your coloring scheme is my coloring scheme is i have like one color for i like this usually you know it's green it's the color i like this there's a kind of a pinky red that i use for i don't like this there's like a purple that means hey this is interesting to talk about uh but this time it's just the highlight colors that are available in in iBooks uh when I read on my iPad but this time I was reading at night uh in um my daughter's bedroom as she was sleeping next to me and I needed the screen to be really dark so I put it in classic invert love the the classic invert uh which inverts all the colors and so this time uh we're we're gonna have i don't know which passages i like or dislike or wanted to comment on because all the colors when you invert them there's like two different ones that kind of

look red and a couple of ones that look kind of purpley brown so uh today it's i don't know what the colors mean i just have highlights and i don't know if i like them or not or want to talk about them so this is exactly the kind of content that I think that one viewer was talking, well, that one listener was talking about. Like this, you know, this is just not needed. Yeah. Yeah. But see, okay.

The thing is what we're doing here, right, is not, if we wanted to only have the content that was needed, this would turn into like a big lecture series. Yeah, we're not here to lecture anyone no we aren't i enjoy lectures personally i know that there's a lot of people who are like lectures are all bad and you should never do lectures and people don't learn from lectures blah blah i love lectures i think they're great as long as the speaker's not super boring i'm like a big in-person talk person like i like them but there's something to be said about a different format right and so what you're getting is the you know informal ad hoc conversation but

we just happen to have a paper out in front of us and so you're going to get digressions you're going to get jokes you're going to get completely irrelevant things and have divorces and if you don't like it uh keep listening to the show anyways yeah you know someone actually did tell me they uh listen whilst like doing their laundry and cleaning hoovering and stuff yeah that's me that's that's how i do it yeah wait you listen to the show like that yes i've actually started editing the show like that so i'm gonna let's let's do a type five on how i edit the show um i've got a new strategy for how I edit that. No, let's get into it.

All right. That's my segue. We are reading a chapter 41 from some book that I don't know the name of. Jimmy selected this reading as usual. Why did you select this? So we were having this whole discussion on now I can't remember which episode on. Yes. What things are computers it was a was it

a dave actually one the beyond efficiency one wasn't it more recent than that maybe was it that long ago now i'm blinking on which episode yeah we were talking about like a while ago you know all people are computers and the universe is a computer and it's it one of us on one of our recent episodes oh no it was on the bonus episode where we went through the history of computing that's what it was oh yeah that one has just gone out of my mind yeah that was such a good episode i erased that it was so good yeah so you might be like wait what conversation on the bonus episode where ivan wanted us to go through the history of computing without any notes. Ridiculous idea.

It's so good. Don't listen to that episode if you're wanting to get good content on our bonus content. It's so full of bad ideas. It's so good. It's such a great episode. We should do it again. This is one of those episodes where you'd be like, are Jimmy and Ivan okay?

I was trying to is this is one of those episodes where you'd be like are Jimmy and Ivan okay I was trying to walk us through yes like conventional computing and there was a lot of pushback and a lot of ideas about I have that highlighted conventional compute all this overtalk is staying we're getting better at the overtalk I think just keep going yeah yeah don't stop there was a lot of uh talk about how everything is computational everything is a computer so i think uh lou was you know saying that we could start at the very beginning with the big bang and then animals and blah blah i thought it would be fun to find a paper on this question like what is computation what things are computational etc and so i found this paper by jack copeland i wish you would have found a paper that was like right about it um i

intentionally found a paper that we would probably all disagree with excellent oh no no it's mean it's mean no there's nothing mean about it i think this is a great uh like exploration of the idea i don't agree with it yes and but i couldn't find if we picked a paper that was just my view that would be a little self-serving yes if i picked a paper that was just one of you you're all's views i would have to wait for you to write it because i don't know your views well it's coming out. That's a good point. Yes. And I thought would elicit our views.

Yes. From this paper. Right. That was the goal. So this is, is the whole universe a computer by Jack Copeland and et al? Is the whole universe a computer? That's, that's quite a question. The theory that the whole universe is a computer that's that's quite a question the theory that the whole universe is a computer

what is this what is killed him he's dead we did it what is this episode now why are we trying to over talk me every time i talk could we spontaneously read the whole quote together you can read it ivan please yeah do it you can do it read it, Ivan, please. Yeah, do it. You can do it. Read it, Ivan. No, you go ahead, Jimmy. Please, please. The theory.

I mean, I knew it was coming. I'm not. That's why I stopped. I stopped before your audio arrived. Yes, yes, yes. The theory. The latency. I stopped before that anything happened.

Universe is a computer, is a bold and striking one. It is a theory of everything. The entire universe is to be understood fundamentally in terms of the universal computing machine that Alan Turing introduced in 1936. We distinguish between two versions of this grand-scale theory and explain what the universe would have to be like for one or both versions to be true. Spoiler! The question is in fact wide open. Bullshit.

At present stage of science, nobody knows whether it's true or false that the whole universe is a computer. Still bullshit. But the stage of science, nobody knows whether it's true or false that the whole universe is a computer. Still bullshit. But the issues are as fascinating as they are important. Dubious. So it's certainly worthwhile discussing them. Well, we begin right

at the beginning. What exactly is a computer to start with the obvious your laptop you could stop stop sorry i really feel like if i don't interrupt you will just read the whole like 10 pages or whatever it is. Stop. So can I ask, like, I think we've got some clues to Ivan's opinions when he was reading that out. But like, well,

I don't know which way. Does he think it's obvious that the universe is a computer or obvious that it isn't? No, I don't think we know. Like we don't know what it is. That's what he's saying. Okay. But that's what they said. That's what the paragraph just said.

And you said bullshit to it. But you said and yes i intentionally said bullshit it's fine wait what so you said bullshit after you read out the bit you agreed yeah wait throw you off my trail this is ridiculous set you up so you don't know what i think oh my god okay okay hold on so we're supposed to be on the topic, but like, are we going to repeat the thing where Ivan pretends he hates a paper for a whole thing? Cause we already did that. We already did that.

Yeah, we already did that. I got it out of my system. I actually, I actually dug this. So, okay. Okay.

Listen, I think, I think it is interesting. It is fascinating. He says the issues are fascinating. Yeah. Right. I think it is an interesting topic of discussion but like what

do you think about like the question just like the premise of this paper first of all like we need to know i think it'd be good to go ahead and break down what the two sections of this paper are uh because i think that will help us so you're gonna ask is the universe a computer and there are two different ways of taking this question and the paper kind of splits in half more or less i think the second half is a little bit longer because it's one they the authors clearly take more seriously which is is the universe an actual computer not it could be conventional it could be cellular autonomous but like at the base is that how it's built?

Is it a computer? Is it computational? Are we inside a computer? Are we inside a computer? Or you could say, can you compute everything about the universe? Is everything about the universe computable? That's the second way of taking this. Which doesn't necessarily mean it is computed, but just that it could be computed exactly could you simulate the whole universe so like you need the second one for the first one yeah so part one am i getting this right part one is like are we in a computer yeah are we actually in a computer yeah and part two are we in something that is computable exactly could we could it be in a computer like if if

part one is true we are in a computer then part two has to be true yes the thing is with this question right like nobody knows whether it's true or false and that was a direct quote yeah which i disagree with right okay so but the other thing is i'm questioning what does that even mean because and and they break this down in the chapter as well like our understanding of what a computer is is very it seems very dynamic even in the past 100 years even in the past like 50 ish yeah maybe 100 years is the right number but yeah like what What a computer is, is kind of like whatever we want, you know, at a given time. It's meaning has changed. So it's just like without any extra context at this point, it's a very like vague question.

It's like, can it even be answered? Yes or no? Or is it kind of like what we do? It's about what we decide a computer is one of my quotes that's in purple for some reason is from a bit where they take alan k's oh my goodness they take they take one of the alans of all time alan turing alan turing alan k they take one of the alans brett victor they take brett victor's definition of what a computer is um and use that Alan Kay. quote um the summary is basically like a computer no matter how you make it whether it's electricity or it's like hydrodynamics or whether it's like you know oh i wove this pattern of hairs together

or whatever it is however you make the computer what it's doing is in some provable way equivalent to what a person could do if that person could just do like calculations with pencil and paper and repeat some kind of process like that. And so Turing's way of explaining this is, the idea behind digital computers may be explained by saying that these machines are intended to carry out any operations which could be done by a human computer. Or a human computer is like a calculating person, a person who, you know, just sits there and does math on paper. Yes, or to be more specific, in another quote, Alan Turing joins the grand tradition of referring to human beings as only using male connotations and i quote a man provided with paper pencil and rubber and subject to strict discipline is in effect a universal machine right um and it's interesting to see like

there's a there's a bit of history explained in this chapter around, you know, like how a computer, that word used to refer to a human being, right? If someone, and I quote, if someone spoke of a computer in the 19th century, or even in 1936, less than 100 years ago, I added that bit, they would have been taken to be referring to a human computer, which is like, it's kind of like, I think, you know, with Turing, there was like a war effort behind it. But like, always the purpose has been to replace human workers, like to replace labor, which is just like an interesting starting point for all of this. You know, like one thing I realized is that at this point, they were really focused on making computers that did what we could do. Whereas I think nowadays, we're often looking for computers to do things that we can't do.

Like we're hoping for them to do even harder stuff or even more boring stuff. I think this paper does a really good job laying out this beginning point about, you know, what computers are. Yeah. And so, you know, you get that they don't have to be electrical. It could be slime mold computers, DNA computers, evolutionary computers, computers that use billiard balls, and computers that use swarms of animals or insects to solve a problem. The list goes on. There could, in principle, even be a universal computer consisting entirely of mirrors and beams of light. It doesn't matter what the physical makeup is.

What matters is what they're doing. And so we get a little later here. It says, here then is the Turing-style answer to the question, what is a computer in the modern sense? And it's any physical mechanism that carries out the same work as the idealized human computer is itself a computer. So that's the answer that's given in this article, is that if you're doing the work that an idealized human computer idealized, there's no limit on time and all sorts of things like that, then you are... Pencils.

Yes, pencils. Then you are a computer uh right so we can make a computer out of anything it just has to do this work so like there's some uh i guess like more out there ideas there like even like a hey a dna computer or a dna machine can be a computer so like i'm i'm on board with that like i think that i think that's that's good i think that's like a fair definition of a computer what what do you two think are you on board with that or do you think that some of those things are not computers i think one of the things i just want to say is like you know maybe we could come up with a different way of talking about what a computer is. But here, you could really just recast this article at the top and just say, is the whole universe a computer TM? Or TM doesn't stand for trademark, but Turing machine. Oh, you cheeky.

So that's what they're talking about right yeah they did a very good job of i think in the beginning making it clear that like that this is what they mean by computer yeah maybe there's some other definition you could want to give but they want to be pretty precise right yeah yeah and so and that's going to be really important the reason i'm trying to like reinforce this idea is it's going to be really important as we get into is the universe computable yes because when we say is it computable we mean on a by a turing machine by yeah or by an equivalent yeah yes yes by something that is isomorphic to a turing machine yeah i think this is a great definition personally like i think they did a really good job kind of walking through yeah and like making sure they stick with what you know is the conventional notion um i have some i differ

on some details that we can get into later i don't want this to be like a criticism of the paper because i think the paper does a really good job exploring on its own terms yeah but i think of computation a little bit differently. And it's just a slight tweak that I think undermines a lot of what the paper actually ends up going into. Spoilers. Because I think I could answer this question from the outset. That's the real value that we on the Future of Coding podcast bring. We'll select a paper with a definition of computer or something like that.

And we'll say that's a good definition but you got to keep listening to the full three hours to hear all these wild ideas for what other things a computer could be so there's one section there's one paragraph i guess towards the end of this intro we bit where they say right they say and i quote they say, and I quote, Is the universe a conventional computer? A cosmic version of your laptop? And then they refer to it as a name I can't pronounce of some

computer or anything. This seems to us logically possible, but not terribly likely. Where is the cosmic computer's CPU? Where is the cosmic computer's CPU? Where is the cosmic computer's memory? Where are the registers holding the operands and the registers in which the results of the CPU's operations are stored?

There's no evidence of the universe containing any of these core elements of the conventional computer. I just wanted to pull that out because I think that's the first time that they switch from defining stuff into saying their view. And they're basically saying, listen, if we can consider the universe a computer TM, then it's not going to be a conventional one. It's not like your laptop where you plug in a USB stick

or anything like that. They're saying, listen saying listen listen we're going beyond conventional computers you know that with cpu and ram and stuff we're talking about a different kind of computer here right and i i'm i'm i'm on board with them it's like yeah if the universe is a computer it's going to be a pretty weird one probably and what's interesting is this i think actually maybe makes it so that this question is not, are we in a simulation? Because imagine we're in a matrix-style simulation. We still wouldn't be able to discover the CPU and the memory and all of that.

Right, yes. It would be like, what are the software primitives that we somehow have access to? Not the hardware. Because you're not in the simulation. There is no CPU. So it's slightly a different question. It doesn't really matter for this article, but I thought that was an interesting point.

But yeah, I think it's very clear that the universe does not have a CPU. The universe does not have memory. And they define their conventional computer, just to be clear, as anything that derives itself from the 1945 first draft of a report on the EDVAC. So this like von Neumann architecture kind of set up.

There's also in this bit where it's talking about the difference between a conventional computer and an unconventional computer, slime mold DNA, that sort of thing. There's a bit here where it says, however, the Californian philosopher John Searle argues that even your garden wall is a conventional computer. And other philosophers maintain that a simple rock standing motionless on a beach is a conventional computer. And it references that there's a discussion of this somewhere else in this book. But I don't have the whole book. I just have this. but I'm curious, Jimmy,

if you know what that's about, that's the points I will bring in a little bit later. I think he really doesn't summarize John Searle's view very well at all. And I'll want to talk about that, but I think it will get us, as they say, it's a little bit away from what the topic of this paper is in some ways um but i i think we can come back to that after we kind of laid out the groundwork of what what this paper has but yeah i have a i have some quotes so on my screen here i have the this paper and then i have

john searle's paper uh is the brain a digital computer and there's a quote in here that i think really like pulls out my view i don't agree with john sirtle on all of his views there but his view of computation is really interesting and i don't think that this paper does it justice so we'll talk about that later so then it all goes downhill okay this good you say that because my highlighting color changes and i don't know what that meant so it's good to be reminded that it means oh it's getting worse okay good so maybe uphill maybe downhill so i'm curious are you saying it's downhill because of the assessment of the topic or because of the topic i'm ignoring your question and i'm gonna answer the one i wanted to be asked so i think perfect

so actually maybe i've maybe i've jumped in a bit too early i feel like there's some we're still in like the background this next section zeus's thesis gives some background of some popular cellular automatons that's why i was confused on why you're saying it goes downhill because i thought you'd be super excited for any cellular automaton well okay like but but the thing is i think we're getting towards downhill because there are some sections or some specific statements which i think really really really get it wrong great that's why i picked this paper with with cellular automata and i was like i was reading this and i was like wait wait, why? How have they got this wrong? How have they got this wrong? This will come in a bit.

But my impression is that they seem to have a pretty narrow view or a pretty narrow window onto what cellular automata are. And I was wondering about that. And then I realized I looked up the people behind this. And is it because they're like philosophers not computer scientists and they just know less about ca or cellular automata but yeah i don't i don't i was i was a little bit surprised well let's so this in this section zeus's thesis this is z-u-s-e uh not the Greek god. Yeah. Conrad Zeus. But they similarly are both wielders of lightning.

This Zeus in particular is sort of my sense of their character is that they're like a machinist or sort of engineering-minded person who gets into computer design and has some very specific opinions um uh you know i'm not going to read those quotes um and and eventually it gets to cellular automata for some reason this is the best summary ever uh and quote steven wolfram saying that uh cellular automata are lattice-like grids all all of whose properties are discrete.

They are, quote, systems in which space and time are discrete and physical quantities take on a finite set of discrete values. A cellular automata evolves in discrete time steps. Yeah, I want to hear Lou's take on this. So, for example, that's like just wrong from the outset, right? Oh, it's because i said a cellular automata and i'm sorry about that the singular is actually a cellular automaton which is what was written here i misread it i'm sorry right right which listen yeah yeah it was it was incorrect you're correct no no no that part's fine. I don't care about that. The funny thing with cellular automata is that lots of people who are cellular automata nerds love to point out when people get the plural and singular wrong.

So it's kind of enjoyable to get it purposefully wrong often. Should I reread the quote with it? No, you've said enough, right? Okay. So, but, but. Sorry, that sounded really mean. I didn't mean it. Please stay together.

We're okay. We're okay. So, yeah, this statement about cellular automata being discrete is just wrong, right? Like, typically cellular automata are discrete, but there is also like a whole, I guess, classification of cellular automata that are continuous. And you could argue that you could argue that, well, they're not really cellular automata, but they are very commonly referred to as continuous cellular automata. Right. So from the start, this is just wrong. This is a wrong statement continuous cellular automata right so from the start this is just wrong this is a wrong statement about cellular automata and but i know it's from steven wolfram and some people take him as like the uh the god-given authority on cellular automata but

it's just not truly representative of what's out there he's the brett victor of cellular automata do you know what i think that's unfair on brett victor i also think that's unfair to brett victor of cellular automata do you know what i think that's unfair on brett victor i also think that's unfair to brett victor um but it's just just wanted to point that out that there is a specific subset of cellular automata that are discrete it's not all of them i'm gonna go out on a limb and say like and this this paper we'll come back to it in a bit but it really gets into like discrete versus continuous when it starts talking about like hey if the universe is a computer what kind of computer is it oh computers are discrete but we think of the universe as continuous i'm actually i'm gonna go out on a limb and say there's no difference between discrete and continuous i'm gonna say that that is a that is a meaningless distinction what do you mean there's no different street discrete what are you talking

about what are you talking jimmy has got his head in his hand any time that we waste trying to say well this thing can't be that thing because one's discrete and one's continuous it's like you will find a continuous version of the discrete thing and you will find a discrete version of the continuous thing because actually there's like that i can get behind but they are different we know they're not the same well they're not the same the cardinality of the natural numbers is not the same as the cardinality of the reels the reels aren't continuous that's the that's the mistake you're making the reels are discrete um and then natural numbers are also discrete so do they have

the same cardinality uh uh one of them's uncountably infinite. So they're different things. They're different things. Yeah, they're different things, but they're both discrete. How are they both discrete? Because you can take a d-to-kind cut of any real number and discretize it to some arbitrary precision. And then you get approximations. No, no, no.

You get something that is effectively equivalent it's it's indistinguishable effectively he puts all these yeah it's he's just trying to troll now he's just he's just messing no no no no the point is that like if you're gonna hand ring over stuff being discrete or stuff being continuous don't because there's enough to go around there's enough enough for everyone. It really does matter. It really makes a huge difference on the properties of these physical systems. And I think, Lou, your point is well taken, that you could expand. It depends on if you're having this more restrictive or expansive notion of CA.

And I think it's totally fine to say that there are continuous CA. I think you could recast what he's saying here is, is it a, just, you know, a discrete cellular autonomous, which, you know, yeah,

I mean, it's, it's interesting though, because I think later on, like, you know, they're going to say, well,

you know, what Ivan was saying about how cellular autonomous, well, they, they are only discrete, so they can't apply here and that's that's just not true so i think like if but if you go into this argument thinking that they can only be discrete you're going to come out with a different answer at the end yeah i i found it i

just to show you both um i'm holding up my ipad to show here's a screen full of green remember that green i don't i don't think it actually means good and it's all the like let's try inverting uh first existing physics involves continuous quantities position energy velocity whereas cas and all digital computers only deal in discrete units not continuous quantities both sides of that are wrong no that's not time is traditionally regarded as being continuous no it isn't uh what about quantization right physics physicists have stopped regarding time as continuous for a very very long time now that's not true at all yeah quantization right energy only moves in discrete steps we think that

there's a like an equivalent plank unit for time time is still continuous by no no time is discrete wait why why i don't understand. Yeah, there's a Planck length of... There's a Planck unit for time that... Planck. Yeah, there's a Planck of time that is the smallest discernible unit of time, which is the discretization interval of the universe,

and yet it's phenomenologically continuous. There are, of course, physicists who think that time is discrete. Nine out of ten dentists think that time is discrete. I don't think that's true at all. Do you know what that value is? It's nine-tenths. The Planck's constant, symbolized as H, is a fundamental universal constant that defines the quantum nature of energy

and relates to the energy blah blah blah blah in the international system of units the constant value is 6.62607015 times 10 to the minus 34 joule hertz to the minus one or joule seconds right so that is great content the plank constant santa fe institute there's here's a post from the santa fe institute scientists believe that time is continuous not discrete right that is how this article starts there they're on the west coast they're all high on their weird drugs and no this is an article trying to argue that there's discrete time physics hiding inside our continuous time world and yet they're saying that scientists believe that time is continuous that is the

the consensus yes then i reject reality i think we've all made ourselves clear like come on ivan i really can't tell now if you're just trying to troll on this distinction here just for fun because you think it's being pedantic no no it is it is like 50 50 i think it's actively harmful to over index on discrete versus continuous i think that that is why why because we exist in a computing paradigm in a cultural practice and a set of norms that is entirely about discrete digital computers and our programming models reflect that discrete nature and the systems that we work with make it very hard to represent continuous quantities and continuous behaviors and processes evolving over

time which is the point being made here uh yes though they're they're the point here is the contrast between the physical world which has continuous quantities and digital computers which deal only in discrete units and the the digital computers that we have do not need to be discrete they could be continuous but we're like path dependent on them being discrete and so we mistake that and think that that means that they're inherently discrete which is not true if they're digital computers they would have to be discrete but digital because that's the definition of what digital is right that's what digital means yes but that is only because there's a very good hardware abstraction that lets you sort of like you know like a um c is not a low level language right

like branch prediction all that kind of stuff like we maintain this illusion that the computer is discrete but like it is not actually the computer is continuous it just you know most of the time you can pretend that it's discrete and we we live in that illusory world and i think we're worse for it okay so you're saying there is a distinction between continuous and discrete culturally not physically physically no culturally there's a distinction if the universe had discrete time yeah it would be a different universe from a universe that has continuous time if there's a distinction between those two things at the bottom level yes it's like a like a particle wave duality i think that there can be a discrete

continuous duality yes yes yeah okay okay i have no evidence for that but they don't need any evidence to to make this paper this is this we should keep going we should keep going i know this is great i love i love hearing ivan's crackpot physics ideas it's great yeah well just try to open your minds a little bit uh-huh uh third current physics allows for non-local connections between particles relationships without an intervening messenger this is known as quantum entanglement that's yet another bit of physics that they try and sprinkle in to say like, oh, how could a cellular automata have distant connections between cells of the grid? These people have never seen the connection machine, which is a giant grid computer where you have non-local connections. And it's awesome at simulating these kind of CAs with non-local connections. So they're the poorer for it.

Second, physics seems to involve non-deterministic random processes where CA behave in a completely deterministic way. I know that Lou is going to have some things to say here. So right. So right. There's no randomness. No. So right.

I know Lou is going to have some say. I want to just go ahead and say, we should make the distinction between pseudo random and the kind of randomness, complete non-determinism they're talking about here. Yes, this is an important distinction. Pseudo randomness and complete non-deterministic at the level of the universe are very different. Pseudo random are computable. Complete non-deterministic are not computable. So if we're talking about at the universe level something being cellular

automata. I can't out say the word and I'll just keep saying CA. Automata. If we're talking about something being CA at the fundamental level we would have to say that it's computable and so there could be pseudo randomness but there couldn't be this complete non-deterministic no no that's not true okay cool um all right okay so you think that like maybe we should just expand the definition of ca to say okay you could have rules that are completely non-deterministic. Yeah, you can have CA that are non-deterministic. You can have a non-deterministic rule as part of that CA, right? Full stop. It's just as simple as that.

The most popular CAs that we know of, just because I think people find it more elegant and beautiful, are deterministic. Game of Life life is deterministic all of wolfram's things are deterministic but um there's a whole huge class that are non-deterministic and you're right when we do them on our machines on the our computers in our physical universe in like you know rather than theoretical well i'm being more no i'm not going that big it's not that deep bro you know like i'm just saying like on my laptop like you know i use a pseudo random number generator um to simulate those kinds of non-deterministic cas because like we the way that we build our computers is to be as deterministic as possible

but like when we build a deterministic computer we're kind of like competing against like the non-deterministic non-deterministic world we live in they have so much like redundancy in them that makes them as deterministic as possible but to be honest like deterministic execution of code and this is a quote deterministic execution of code has always been an illusion. You know, there's always the possibility of cosmic rays coming in and flipping a bit, say, and that does happen sometimes, and that's cited. But we know that we can engineer traditional computer hardware so that the chance of that is small enough that we can usually ignore it. So,

okay, so can you tell me what you think is a genuinely deterministic computation that you can run on a computer? So hold on. I think we're mixing up multiple notions here. So I want to say, one, to make sure we're in agreement, if you had a truly non-deterministic CA, it would not be a computer TM. Yeah, I can agree with that. It has to be the case because turing machines can't do non-deterministic things yeah yeah like in an abstract like sense not we're not talking about like physical implementations of these we're talking about the abstract theoretical notion

cannot do anything non-deterministic yes i guess like one of the like remember these sort of some of the examples of computers earlier some of those being some of those unconventional ones right like the dna machine or dna computer was one of them right do we consider that to be like a computer tm yes because you can make a model of everything it's doing that only uses a deterministic turing machine i i'm not going to agree with that because that that presupposes that the universe is deterministic if we live in a non-deterministic universe then all physical computers like a slime mold computer a dna computer which are physical computers right those aren't abstract models yeah what we're saying is we can yes i agree i'm not that's why i tried to say we can make a model right i'm not saying we can't know for any computer whether

physical computer okay i'm sure yes we got to say physical or abstract or theoretical we gotta like for any physical machine we can't know if its effects are completely deterministic unless the universe were completely determinist that's right that was what i just said yeah right right all in agreement but when we make a program and we run it on non-conventional hardware or conventional hardware the algorithm we're computing we can model using a deter idealized deterministic Turing machine. That's right. So like the non-deterministic, back to the CA, right?

So the non-deterministic CA are sort of considered like this, I don't know what the right word is, but like higher class, you know, classification. Exactly. Exactly. So like the thing is non-deterministic CAs, in my sense, they're pure, you know classification exactly exactly right so like the thing is non-deterministic cas in my sense that's pure you know they're better because like you can uh implement a deterministic ca inside a fully non-deterministic ca you cannot uh simulate a non-deterministic deterministic ca inside a deterministic ca or sorry, simulates the wrong word. Exactly.

Sorry. Maybe, maybe you can simulate, you know. Yeah. You could try to make something, but it would lose something. It's a simulation. You cannot implement it in a deterministic CA or it's like your best. It will be like a 99% implementation, right? Exactly.

It won't be isomorphic to the non-deterministic CA. Yeah. And this is something that like, I have to tell people when they say that, like, well, it's not like a, why aren't you doing a deterministic CA, Lou? You know, but I have to say, well, like, no, this is bigger, this is better. Because, you know, like, you know, because you can do you can do anything in it, right? Yes. And it's like, in a way, it's like a step closer to at least how we see our world sometimes.

And so if the world were a non-deterministic CA, but just to take that as a given for a second, it wouldn't be a computer TM. Yeah, right. Right. And that's the question that we have in this article, which is why we're only considering deterministic CAs, because only deterministic CAs are computers, whereas non-deterministic CAs are a level above what computers can do. Well, you know, so I think I think you're right to me, but I question like whether that's why they said that, you know, like, you know what I mean? Like, I feel like they're simply not aware you know they're not saying so we're not counting those cas they're saying that those cas don't exist you know so i feel like it's interesting because like the horizons of the chapter are slightly closed off yeah yeah that's a good point and that's perhaps also why we're only considering

a specific class of computation which is a a computation TM or whatever, computer TM, right? Because this could be, you know, like it sort of raises the question, well, if you knew about this type of non-deterministic computation, then maybe this chapter would have ended up looking slightly different right but you're right we have this case where we're not even considering those things um so i don't know i don't know maybe for for what it's worth jack copeland is one of the coiners of the term hyper computation which looks at models that go beyond Turing machines. So like Jack Copeland, one of the first author listed, is a big fan of things that are not simulatable by Turing machines. I think the reason why this article

is probably going to be a little bit more focused on that is honestly actually in service of, and maybe we need models beyond Turing machines. So that might be part of why you're seeing such a narrow focus. Yeah, I don't know why. But it was just interesting to see, like, yeah, hey, CA are only discrete. CA are only deterministic.

When I think many people would say, well, that's not the full picture. I think that's a great point. Absolutely. Yeah. So yeah, I might be imputing something that's not in the... I think you're right the article doesn't mention these things but i i hope we can be on the same page that like a non-deterministic ca would not be a computer tm so if the universe is a non-deterministic ca

the universe is not a computer right but you know like the tm you have to say the tm loud i think you know what i mean yeah yeah just in the scope from now on when we're talking about computers we mean computer tm right but you know it's like it's like you like asterisk now we've never been able to make a non-tm computer right wait what we do not have any computers we have made that we know that we can prove can compute something beyond a turing machine we'll talk about how we could make one later in this article but we do not have one define make make physically or make a model like a theoretical model make physically sorry, sorry. Make physically. Well, if the universe is non-deterministic,

then all physical computers are beyond a TM, right? No, that does not follow. Why not? Because we're asking, can we make it compute something that a Turing machine couldn't compute? Just because it does things that a Turing machine doesn't do doesn't mean we've gotten it to

compute something that was uncomputable. And we'll talk about a system where we can't, where we could theoretically make something where we can get that computer to compute something that's uncomputable. But we haven't done that yet. Not with quantum computers, not with anything. There's nothing where we can say, here is this uncomputable function, uncomputable being a technical term in the Turing machine sense, and we were able to make a physical thing that can give us that answer.

Is that possible? We don't know or not, because it all depends on factors of the universe, and that's what the rest of this article is going to go into. I want to close the chapter on the the zeus ca though before we get there well okay but just one thing like i'm worried i don't think we should open too much of that because that's like scope explosion right like of because i think it's

helpful to to know that we're talking about computer tm right like the turing yes right so uh like i i make no comment on non-tm computers this point yeah later in the article well that we have one where there's a discussion about how we could build a theoretical idea of how we can build a computer plus theoretical relativity stuff that would go beyond a turing machine. Interesting. Yeah. As far as the article is concerned, if we're talking about, yes, probably, I think good points, Lou, that it maybe should have said non-deterministic versus deterministic. So I'm just going to say a deterministic cellular autonomy, the author thinks has some problems here of the universe being that thing if if the universe were that thing

it becomes a little difficult to say like well how exactly would it work uh so the first point is like what though would implement the cellular autonomous what would be the underlying hardware that the cellular autonoma is like being run on this is like one problem that we have with this yeah because you know like i said earlier it's all a lie deterministic computation is you know it's it's not attainable yeah the point is like if you try to say well it's like the atoms or it's you know the quarks or whatever yeah well those are the things that we're trying to explain how they came about so it can't be any of those physical things so it would have to be a non-physical thing and there's a suggestion here that the physical universe consists of abstract mathematical objects

that feels a little weird how do those give rise to physical objects? It's really this article kind of basically like putters out on this point of just saying like, I don't understand it. I think would really be what the author is saying here. I found this entire section deeply unsatisfying. And I want to know if either of you feel the same way uh i mean i found it unsatisfying because of what i already said about sort of like the narrow uh viewpoint of what a ca can be but the fact the point it makes about like deterministic computations don't satisfy our need to like i guess model and interact with the wider messy sloppy non-deterministic world like that i fully agree with and that's why that's one of the reasons why pretty much all the stuff i make is a non-deterministic or at least you know when

i run it in my browser it's my computer does its best job at being deterministic which then i trick into being non-deterministic uh but actually it's not but actually it kind of is because it's all running on this machine that i could smash with a hammer at any time right it's just like we're sort of yeah i i i feel like i i do a similar thing to them where it's like wait wait, wait, what comes first, like the chicken or the egg? Does it? We have a, we're in a non-deterministic world, and we make something deterministic, and then

we make it simulate non-determinism. But really, it's kind of deterministic, but really, it's non-deterministic, right? I could follow all those steps through, but I've... You're sort of leaning into the vibration between determinism and non-determinism. It's kind of like weird. It's a bit weird. It's a bit weird. It's clear that the authors don't take this idea seriously.

It's clear that, like, they don't give it its due of, like, could the universe actually be under the hood a CA? Yeah. Because they can't make sense of it. It is honestly, like, I think, I think, more or less what they say. They even quote, Wittgenstein said in his usual pithy way, whereof one cannot speak, therefore one must be silent.

Talking about, hey, if this were the case, the CA, if all of physics were implemented as a CA but not physical items themselves being a CA, then how could we talk about them? They'd be beyond physics. There are people who take this seriously. They're very few and far in between, just to be honest. Wolfram is one of them. Stephen Wolfram is now having his new physics project where he's trying to not use CAs, but use something fairly, a family resemblance to CAs to kind of undergird all of physics and explain all of physics that way. Yeah, like you could still pretty cool those things

that Wolfram's working on CAs in some form. It's an even more abstracted view of what a CA is, but I think people consider the things he's working on to be CAs. But I agree with the authors that that doesn't really work. But I think they would have a very different answer here if they were willing to let go of the requirement for determinism, to let go of the TM from the computer, and consider non-deterministic cas non-deterministic

cas are not some crazy thing it's very they can be very simple and easy to to express but i understand that they're not looking at that here yeah i think they should have had a section on non-deterministic cas it might have been short just saying, if that were the case, then they still wouldn't be computable, and they wouldn't be computers TM, and that's the restriction of our scope for this article. And so if the universe is a non-deterministic CA, it's not a computer TM, right? I think that would have been a reasonable addition. I want the sequel, you know? I want the sequel. I want the sequel where they uh they look at and i think maybe they like you were saying jimmy they may be hinting at needing like something else they're they're taking a fairly large issue on in this paper like it's a pretty big question it touches on a lot of things and the way that

they approach that question is by sort of setting up a single chain of like well what about this possibility well let's consider that and that possibility doesn't work out so that has this implication and they kind of there's a lot of places where they could branch off and talk about a lot of variations of these questions but they don't so for instance one of their objections to thinking of the universe as being modeled as a dis as a deterministic cellular automata is that physics involves non-determinism. Another objection is the distant connections between cells. Another objection is the theories that physics has produced that are our current best models for explaining phenomena are general relativity and quantum mechanics, which appear to be incompatible.

And they say, how to unify general relativity and quantum mechanics is the hardest problem in current physics, but this is exactly what would need to be done by an underlying computational theory. No easy task and i want to call bullshit on that because they're basically saying that that incompatibility between those theories um is something that needs to be solved by cas because cas would need to sit underneath those two theories in some sense and what i would counter is there could be some unifying theory that we come up with that unifies those two things that does not involve computation. And the computation could sit under that thing. And so saying that we would need to come up with a computational solution to unifying those two physics theories is just a way of sort of saying, like, like well we're not going to explore much deeper beyond that we're gonna like say ah here's a problem in physics if the universe is going to be a computer

it needs to solve that problem and i just i don't like that um requirement being placed on the question i agree that that was a little bit of a silly aside that didn't seem necessary to yeah i think the logic's supposed to be something like i'm not saying i agree with this i think the idea is that if you came up with a concrete you know computational theory it would end up it has to emerge out that it if it was right that it solves this incompatibility but like that seems a little silly because what if we just come up with a equally wrong theory a computational theory that also makes it so that there's an incompatibility between general relativity and quantum mechanics but it's

computational and it's just as good as what we currently have that also seems like a possibility i agree that seems a little like uh okay but, but you can give that criticism to anyone trying to make a unified theory of anything. are not critiques at all they're just distractions um but it's fair because we don't have answers for that kind of stuff or at least if we do they're not presented here they're sort of saying well it would be hard to do that so it can't be true you know yeah or like here's a related problem that needs to be solved and we haven't solved that problem therefore we can't solve this other related problem of you know the universe being a computer to be fair to them they don't solved that problem. Therefore, we can't solve this other related problem of the universe being a computer. To be fair to them, they don't say that it's false. They do say that it faces big challenges.

Yeah, fair enough, fair enough. And they say that nobody knows whether that's the case or not. They don't say that it's false. Yeah. Right, I'm not saying that they're saying it's false. I'm saying that they're dismissing the, they're cutting off that branch of exploration at that point they don't explore it any further than that and they could but you know they had deadlines yeah and that would be a much longer chapter and

and i'm not going to fault them for taking a you know for not exploring everything right that's an absurd yeah they do have some good citations to papers yes of course these things of course the fact that there's no difference between continuous and discrete phenomena culturally yeah culturally speaking sure that comes back in the next section where they're talking about we're getting into like what is the actual hardware that could implement the universal computer right because if the universe is actually a computer if the universe is a computer it what is the implementation medium and they say oh it could be mathematical objects like we talked about a moment ago um uh i just got a signal message from a french saying oh my god it's a three-hour podcast i don't know how he knows what we're doing

but uh all right weird what um yeah but uh then they get into talking about um the difference between math and the universe in a way that i want to pull on i want to pull on this thread they talk about like okay maybe the universe is made up of these abstract mathematical objects the problem with abstract mathematical objects is that mathematical entities, and I directly quote, mathematical entities don't seem to be the right kind of things to implement computation. Time and change are essential to implementing a computation. Computation is a process that unfolds through time, during which the hardware undergoes a series of changes, flip-flops, for example, and these mathematical objects exist timelessly and unchangeably. Skipping ahead.

Unchanging mathematical objects are just not the right kinds of things to implement a computation. And then it keeps going. We know that something must implement the universe's computation, but we should admit that we know nothing and can know nothing about this shadowy substratum. A, shadowy substratum. Beautiful. Love it. Culturally. B, how do you two feel about this? Because I've got some big feelings about it. I'm curious if this rang any bells for either of you. Big feelings about maths. Can I make one real quick. Yeah. Minor point.

Yeah. Because you skipped this like a multiple times. It's Tegmark's mathematical objects exist timelessly and unchangingly. So they're talking about a specific proposal from somebody who is saying these are the objects and they exist timelessly and unchangingly. Totally. Yes. I appreciate that kind of clarification. I mean, like, I'm sure there's a way i'm sure it's like such a okay like i guess the thing the thing i'm thinking is about all like reading all of this it's like it's such a like abstract hypothetical notion right like come on there's probably there's

probably some way to make a computer out of abstract mathematical objects yeah some way to what some way to do it i want to i think uh i want to if i'm going to comment on it i want to continue reading a little bit more yeah sure yeah we can go ahead and come back well no i'm saying on this same topic here i just think you skipped a part that i would say oh sure uh so you know it's saying we can't know about this shadowy substratum in the next paragraph if however you think that there is something unsatisfying about restricting the scope of physics in this way then you are not alone red-blooded physicists want to know everything about the universe and will not take well to this idea

that the universe contains a fundamental substratum that must always remain beyond the reach of physics love it i have it highlighted in blue it's it's yeah red-blooded physicists this is my problem with that this attitude where okay let's imagine this were true. Let's just take for a second that the universe really was somehow these mathematical objects. And now we're saying, well, we can't allow that kind of theory because our current ways of doing physics wouldn't allow us to explain that. And so we have to rule out the theory altogether.

It feels like putting the cart before the horse in an awkward way right if that like if we want to actually know the truth it doesn't matter what our current mechanisms are able to let us discover maybe we need to invent new ones maybe we need to understand more things and so ruling it out because it wouldn't let current physical physics uh for current physics would have to remain silent about it, feels like the wrong reason to rule it out. Specifically the line, computation is a process that unfolds through time during which the hardware undergoes a series of changes.

Is that true? I would say yes. Even in the theoretical? We're not talking about theoretical here. We're asking asking is the universe a computation a computer yeah physical like a no is that is the universe akin to a physical computer is it physically instantiated as a computer that's what we're asking is the universe physically instantiated as a computer right but in this section we're getting into like okay what if

we relax the requirements on physicality a little bit right abstract mathematical objects aren't physical but they could implement a computer according to this theory so if we relax start loosening what it means for something to be physical what remains i think i do i think i don't quite accept this right so the the full quote and you may have read this let's read it again yeah let's do it again time and change are essential to implementing a computation computation is a process that unfolds through time during which the hardware hardware undergoes a series of changes and flip flops flip for example neurons fire and go quiet plastic counters appear and disappear on a go board and so on so there's a couple things here it is talking about hardware it's not talking about like a

hypothetical thing right it's saying that time and change are essential to implementing a computation and when we talk about implementing we're talking about like physical instantiation. Yeah. Doing our best effort at implementing a Turing machine, a computer TM, according to them. Right. And what I think is that I don't think time and change are essential because I think there are some types of computer that they just happen instantly. Right. types of computer that that they just they just happen instantly right like an analog computer

i think you could for example like set up a really intricate way of calculating like a multiplication or something right like that just happens yeah but it happens because of physics right right the physics that make it do that multiplication involve a process changing in time to get there, right? Even gravity is, you know, going through time. Right, so you, like, move the knobs in a specific way, and then the other knobs ping up or whatever. Yes, exactly, right? So even the Turing machine has, and the idealized theoretical idea has a theoretical modeled notion of time. Right.

Right. There is a series of progressions. And so like, just to be clear, like when we're talking about these like abstract mathematical objects, they are completely immutable, completely unchanging, completely causally inert.

There are no causations happening between them they cannot change they cannot move they cannot do anything whatsoever this is the model it's very similar to how uh certain medieval concepts of god being completely unchanging immutable etc this is like these are very akin to each other and so you get this problem of like if i think it's more clear when you put it in God terms. So that's why I'm going to put it this way. If God were completely immutable, unchanging, et cetera, how would he interact with a universe

in time at discrete moments? And this is like the same kind of problem. But now you're saying it's this complete unchanging thing. And how would you have it create the universe i do think there's answers here right yeah yeah yeah but the answers would be probably unsatisfying to a lot of people right i mean like back to the ca right if you look at a ca say do like a Wolfram cellular automata, it involves some change over time. There's changes from one state to the next. But interestingly, like the way it's often presented is this static object, this image where each state of the universe is presented on a different line,

and you end up with something static, you know, sort of there's ways of going i think between static and non-static right in terms of representation right so like your mathematical objects could produce something that is more akin to like a list that list doesn't get built up though right like that list just exists in its final form as soon as you construct the formula that produces it, right? Right, right, right. So maybe you need something more. Maybe you need something more than that. Is that what they say? Like you need something to go along the list. If you just have the end artifact, you haven't done the computation. Well, you've done it because you've got the end artifact. The end artifact isn't the doing of of the computation how you got to the end artifact is the question right i mean this is also abstract

isn't it what does this even mean you know i this is this is my problem with this paper right this is my problem it's like they've just created this is why i love this paper this is they just created this like what if we make what if we ask a question that no one wants to know the answer to right and then we just like fiddle around you claimed in our bonus content that the universe is a computer yes you asked the question you made the declarative statement yes but i didn't say tm right i didn't say TM. I said the universe is a computer. And so now we're learning more about your statement that if the universe is a computer,

it might not be a computer TM. Right, right. All right. So we're just getting more clear on your bold proclamation that the universe is a computer. Listen, flattery can get you only so far it is quite interesting though but it is total bullshit it's just like pure like abstract you know philosophy nonsense anyway let's keep going i love it no no no let's not keep going

i would say he wanted this time he wanted to elicit something from the time yes changeable and computation i want to hear it yeah i wanted to elicit something from the time and changeable and computation. I want to hear it. Yeah, I wanted to dig into this question. Like, computation is a process that unfolds through time. Time and change are essential to implementing a computation. I want to find out, we've heard how Lou feels about this. Jimmy, how do you feel about that? Is that a fair statement?

What qualifiers do you want to put on it? By the way, I think Jimmy convinced me about this this i like whatever i said earlier i've gone i take back all right let's go what is it jimmy i think i think implementing a computation yes yeah to implement a computation involves a series of steps and steps involve a progression of time. Yeah. We can talk about relational versus non-relational theories of time, but especially on a relational theory of time, it involves time. Don't know what these terms mean. I know about the relational algebra.

Don't know what this is. You know about relational issues. So imagine you have a fictional universe in which there are five objects. Imagine you have a fictional universe. Like this one five objects imagine you have a fictional universe like this one and there are only five objects yes and those five objects undergo no change okay the question is is there time in that universe and some people you might say well i don't know you didn't tell me if there's time in that universe some people would say no there isn't because time just is the

relationships changing over over like the relationship changing i can't say over time the relationship changing yeah okay yeah yeah i see those two definitions yeah i see that definitional distinction that's cool so ivan i'm guessing you think that there can be computations without time and change? I'm bringing this up because I want to understand, especially you, Jimmy, when you say that you think that implementing a computation requires time, I'm assuming that that applies in the theoretical world, right? Like you're not just talking about like physically implementing a computation, but also theoretically implementing a computation.

Is that a fair thing to say? I think it gets a little complicated. I think here we're talking, we can't say physically, but like concretely implementing a computation, right? Because we are not saying these things are physical, but concretely implementing a computation, meaning making it real in the real world, not theoretical, I think has a different notion of time and change than what it would in a model. Right. Now, maybe in a model, you could try to say, I have a model of computation that does not involve time. That feels a little arbitrary because it's definitely going to have to involve steps. Yeah.

Right. And again, that's why I was saying like on like a relational theory of time or something like having those steps is to have a progression. You might not want to call it time, but that's, I think what you would have to mean by time in a theoretical sense. So I can't think of any theoretical models that wouldn't involve steps for computation and therefore have some notion of time but i think the instance here is not talking about theoretical it's talking about concrete implementing yeah but which is like a

i would say um if the theoretical model of computation involves some kind of succession of states some kind of notion of time then I would think that the physical one has to. Like, there's no way to have the physical one not if the theoretical one does. Exactly. And what I find interesting about this is that this, to me, suggests that computation, you know, or the implementation of a computation is inherently temporal. It is a process yeah so it would be fair to say that it's inherently temporal i know where he's going with this because he has a little smirk on his face and he's going to say then it's inherently spatial as well right it has to be no, this is where we're trying to be pedantic. I was very clear in our topic. A computation and a program are two different things.

Yeah, and what I realized is my statement about programs, programming, that kind of thing was wrong. I phrased it incorrectly. But what I'm trying to get at is what was my underlying belief that motivated me to say that? This I think is great. Cool. A computational process is inherently spatial and temporal. I'm down. Yes.

So you're down with spatial too, because I was going to say if we can get temporal, we can get spatial from temporal. Yeah. I'm down with a computational process is i i might not say inherently i will make a distinction a physical computational process is inherently spatial and temporal for sure yeah i don't know that a theoretical one because i don't think there's any sense of spatial in the that i can make sense of but there's temporal yeah well if you have a succession of states and if those states are distinguishable then i would say it's the

distinguishable um lambda calculus does not have a space in which it happens but it has separate terms yeah but i would i just would not call those spatial. I'd call them temporal. Like the succession of states. The reduction is temporal. Right. The problem with lambda calculus is it always reduces to a single thing, right? Or can you have a lambda calculus No, you can have simply type lambda calculus always

reduces, but untyped lambda calculus can go forever. I just don't, I think you could try to stretch it and say that it's spatial. I would not put those terms because i think space has involves extent yeah and you would want to say that you know these things have a size i don't i don't know when i say spatial i don't i do i'm not committing to something quite that strong so that's like definitionally we're using i will but i will just say like physical physical computation yeah of course it's inherently spatial and all right listen listen i i gotta answer you gotta let

me interrupt because i i gotta make a really important point um listen i've never done psychedelics uh and i don't think i ever will but like if i did it's like i kind of imagine sitting down saying like you know like from state changes and time and temporal and concrete and you know like all these words are just like going around and it's like how did we end up here that's this beauty of this paper did you when you said you said concrete earlier instead of physical like do you do they mean the same thing concrete and physical i don't even know yeah so we'll get into this in the next section of the paper where they make sure to make the distinct i think it's the next section where

they make sure to make the distinction between like is the universe completely physical all right let's do it but let me just pop some mushrooms first all right well is the universe completely physical is not something they're just assuming they're just gonna say all we're talking about is the physical universe but in this context we had to not say physical and the reason we have to not say physical is because abstract objects implement all physical objects right and so we're talking about things that are not physical that implement the physical or we're considering them at least yes yes that's in this conversation wait so you have objects some of them are physical yep like my phone or like the hole of a donut

right okay and uh some are abstract uh--huh, like numbers. Right. And abstract. So for the universe to be computable, abstract objects need to implement all physical objects. Yes, that's the question. Yeah? That's what they're asking, is could that happen? Happy to translate that to myself.

Yep, that's a lot of what we're doing yeah and so the reason i say concrete instead of saying that they're physical is i don't know that the implementation itself would be physical it creates the physical there's like a floor to physicality it's the thing that underlies the physical oh because we're we're oh because we're implementing the universe that's right we can't implement the universe in the universe yes well well well that's a question that they don't wait no wait wait wait engage with yes they should have engaged with that question that's a fascinating question but that's not what we're talking about right so they're saying we could talk about it it's like we need need another word because we're already using physical

to describe what we're implementing. Yeah, to mean the stuff in the universe. So, but how do we actually implement the implementation? Well, we can't use physical, so we call it concrete. That's right. Well, yeah, so an example of something that could be concrete but not physical. I'm not saying these things exist.'m just giving examples souls oh triangles triangles are abstract they're not concrete triangles are concrete they're abstract they're frigid

that's a throwback a soul would be something that is not physical like descartes soul would be something that is not physical. Like Descartes' soul would be something not physical, but concrete. It would not be abstract like the number two. Why is a soul concrete? I don't get that. That doesn't make sense. My soul is leaving my body in this conversation, honestly. Abstract are unchanging, timeless, causally inert things causally in a and descartes soul is not any of those things uh so it's it's not like souls in the in the um dark souls

metaphysical sense it's souls and like the you know descartes has this definition of a soul no met for souls in the metaphysical senses is descartes souls yeah so but souls exist in time right they undergo changes and they can affect something you are using a i i am unaware of the soul as in mind if you want to like yeah descartes idea of yeah the animating essence the non-physical part i've got a mind to move on to the next section do i have a mind mind as distinct from brain yeah right that's what we're talking about well the next section talks about brains we're at the halfway mark is no way yes we're at the halfway mark two hours in i should have taken so and it never mind they're kicking in is the universe computable is the second question yes right so we had is the universe a computer well we did not we did just determine

wait wait you need to say tm you need to say tm right tm yeah yeah yeah from now on if i say computer i only mean tm because that's all i think well i don't not for me i will i would never i am saying it for me um is the universe computable now computable here is computable by a turing machine right but we can i'm going to start probably quoting this a computable system is a system whose behavior could be computed by an idealized human computer this is the definition that's what tm means yes is computable by a idealized right but wait wait and i idealized does that mean like like perfect yeah they're gonna unpack that well we've already unpacked that oh they unpack it more yes i'm sure that they do but uh i want to like idealized as in has an infinite amount of time etc that's that's

really what we're talking about here doesn't get tired that human computer the human computer is idealized in that no limit is placed on the amount of time available to the human computer nor on the quality of the paper and pencils available idealized human computers live indefinitely long and never get bored. Which I just love, that they never get bored. It's not that they, it's not that they never just stop, right?

You could have just said, they never stop doing their computations. They didn't stop. Because they could just like, still keep going, but be bored, but not. They're also having fun.

That's a key. If your computer's not having fun, it's not a TM. Right. There's actually quite a few of these sort of like flavorful you know dehumanizing descriptions of the of the idealized human computer they're funny they're funny yeah but this i find this really confusing when it does this because like i'm like i don't think ah because we so much we're talking about well a computer tm needs to be deterministic right

but then it says like for example like an idealized human computer right yeah and like is a human deterministic okay hold on the human is following a series of steps and those steps are the things that are deterministic not the person themselves that's a good distinction okay the steps are deterministic so a step could be pick a random number between one and ten no because that would be a non-deterministic thing you could say do this pseudo random computational process and that's what we do with our computers. We give a pseudo-random computational process. That's fine.

That's allowable. What you couldn't say is sample the current, you know, we have these like true random ways or whatever from the universe. If you tried to say sample this random thing from these sorts of things, you would now end up having to go, well, is that deterministic or not? Well, it depends on if the universe is deterministic, right?

So it could be allowable if the universe is deterministic, but we don't know. So I'm like, that's fair enough. I get that. So why? Yeah. Okay.

I just want to give it like a justification for why here. Cause I feel like this is something that's a sticking point that may not be clear. So one reason that we're sticking to this is because we're trying to be a little bit historically accurate and trying to say what Turing said. That's one reason. Not a very good reason, but one reason here.

The other reason is that this definition that Turing offered us ends up going into so many different areas that really matter for mathematics, for practical upshots of what we can do. Yeah, yeah, yeah. That makes sense. There's lots of things that if they are computable, we are able to know them and find answers to them that if they are not computable, we have not been them and find answers to them that if they are not computable we have not been able to find answers to them so this actually has like a real practical import yeah yeah when i say computable i am always meaning the definition here of like turing's

lou was making him a bim bam reference i don't even know what that is arcade just laughed there There you go. That's all you need to know. Okay. Nice. In jokes. Like I said before, when we find something that's uncomputable,

it means practically we have not been able, if it's uncomputable, we can find approximations of it. We can do things like that, but we're not able to actually find an answer to those uncomputable problems. It's like computational complexity, where if something's quadratic, it's very hard to find an answer for. This is taking that one step further.

We can't find an answer for it. Yeah. How do you say the name Kolomorov, Kolomogorov of kolmogorov kolob yes until you said it i could have told you but now that you've put that in my head poisoned his mind yeah i find that that stuff fastened like and we're at the two hour mark so we might as well wrap up uh this i i really enjoyed this paper. It like branched off into complexity theory in a way that I...

I can speed run through this section and then we can talk about... Yeah, do it, because this second section sucks. The second section sucks. The second section is great. I have it all highlighted in green, it turns out.

So the question is, and again, every time I use the word computable, I mean computable TM, because that's how every mathematical thing uses the word computable. So, computable, we ask, is the

universe computable? And so what we're saying is, could we, given the, are the current physics, is the, you know, universe as it is, actually computable, meaning we could make a Turing machine program that would give us the whole answers, that could simulate the universe. And it goes through a series of ideas about, well, what if you could have a lamp here, is one of them, that does something uncomputable yeah it kind of like goes

off on these little tangents i i think they're not quite important to like us to go through the details so one example that they give is a lamp that is not a computer but it's flashing behavior is computable you can imagine a a lamp that is just flashing lights, and it's flashing a certain sequence. I'm imagining it right now. That lamp isn't itself computing anything. Let's just assume that there's something else turning it on and off at certain frequencies. But we could look at it and figure out, is this computable? Or we could just say we know the description. So for example, the lamp could be actually flashing the numbers of pi. So it waits a certain amount of time and

ends up flashing pi. And pi is a computable number. And so while the lamp itself is not a computer, its flashings are computable because it's flashing pi but if the the flashings were completely random literally random in the biggest sense you can mean by random not pseudo random they would not be computable because there's no way that you could simulate that using a turing machine simulate is, it's a dangerous word, I think. The way they put it, I'll not use the word simulate. I just think it's weird. Randomness is one form of uncomputability. If the lamp were flashing randomly, its behavior would not be computable because if the human clerk could always predict the

behavior at the next second, then the behavior would not be random. So if something is truly random, there is no algorithm we could write, no machine that we could do that would predict exactly when it's going to happen. And so if it's completely random, we could not make a machine that could predict when the next flash is going to happen with 100% certainty. That's what we're talking about here. So Turing machines can only, you can only predict deterministic behavior. You can't predict undeterministic behavior,

and Turing machines aren't the exception. They can't all of a sudden figure out undeterministic things. So now the question is, is the physical universe computable so the answer depends on what can actually happen in the physical universe if the physical universe has true randomness then it's not computable that's just plain and simple so we have to go ahead and just assume if we want to say that the universe is computable it's going to have to be deterministic ivan's breathing heavy at me but this is literally what is i'm i'm summarizing the article to be clear this is what the article says are you disagreeing this is what the article says or you know i'm not disagreeing with what the article says yeah i'm

disagreeing with the article okay let me summarize what the article says. Yeah, I'm disagreeing with the article. Okay, let me summarize what the article says. You are summarizing what the article says. Okay, I'm not stating my opinions. So if it's to be computable, it would have to be deterministic. Let's just assume that for a second, because if it's not, then it's not computable. All right, so let's assume a deterministic physical universe. Is it computable? Well, there's actually this really interesting idea called accelerated Turing machines.

And accelerated Turing machines are a Turing machine that you make a Turing machine, and each step that it takes starts getting faster and faster and faster so that it could do infinite computation in a finite amount of time. And then the question is, can you make an accelerated Turing machine? And there's this crazy speculative idea by a physicist, Patowski, who says, yes, you could. That if you take advantage of certain things in, it was originally special relativity, but we've moved now to also

making it work with general relativity and you took this black hole known as a slow curr hole and you sent a computer into this black hole and the black hole it's able to exist in there because it's a slow curr one this this paper is crazy right we don't know for sure that they exist but we have astronomical evidence that they might so anyways anyways, if you sent it in there, relative to the observer... Some outer world shit. Time would continue to get faster and faster and faster. So you could... Did you mean outer worlds?

Send in a computer that, assuming you can keep it up and running and blah, blah, blah. We're just going to assume that for a second. What did I say? It could do, relative to some observer, an infinite amount of computation in a finite amount of time. You said outer worlds. And so you could compute uncomputable things. Outer worlds, yeah. Yes, it's pretty cool, right?

And so if this is true, if this is physically realizable... The game, that is. The universe is not computable, which is fascinating. Okay, and what do you think jimmy okay is it okay it is actually like this this paper is like really enjoyably trippy right and i think i think like um i think my mistake going into this recording is like i wasn't in a trippy enough state you know and like i said i listen i don't do drugs, but there is something real crazy about this. Like, you know, we could send a computer into a black hole, right?

And we don't know if they're real. But it's a lot of fun. And I want to know what you think, Jimmy. What do you think? All right, so I did say I want to talk about this John Searle thing. From six hours ago. I think it's kind of weird the way that this paper put John Searle's thing. And so I just want to kind of read, you know, just a little couple quotes here from the John Searle paper that I think is the one they cited.

If it's not, then it's a very related paper. So John Searle has a bunch of ideas on the mind, and this is a paper called, Is the Brain a Digital Computer? But we're not going to talk about minds for a second. We're just going to read this quote, and I'll explain it for sure. There is no way you could discover that something is intrinsically a digital computer because the characterization of it as a digital computer is always relative to an observer who assigns syntactical interpretations to the purely physical features of the system. This point has to be understood precisely. I am not saying that there are a priori limits on patterns we could discover in the universe or in nature. We could no doubt discover a pattern of

events in my brain that was isomorphic to the implementation of the VI program on this computer. But to say that something is functioning as a computational process is to say something more than that a pattern of physical events is occurring. It requires the assignment of a computational interpretation by some agent. So if I had to summarize this, I would say that the universe can't be a computer, but we can think of it computationally, right? The point is nothing in and of itself. So when we're having this bonus discussion

about are looms computers, is the universe computer, is an organism a computer? How I was taking this is that nothing is intrinsically a computer, but anything can be treated computationally. So we can look at things and see the ways in which we can discover a pattern of events in them and look at those using our computational lens, right? We can think of them as a ca etc wait wait wait i didn't think you were gonna say i didn't think you were gonna say that

wait so wait wait wait wait wait where do you say computational lens there's actually no tm there right no tm no tm so you have to specify no tm Yes, Notiem. We can look at them as a deterministic, a non-deterministic CA, right? We can look at them using all the tools that we have at our disposal as programmers, and we can interpret it that way. And that is one sometimes useful, sometimes interesting frame to look at it.

But that doesn't, what I don't want is to say that that is the essence of what that thing is right so we can look at music in a computational lens but that doesn't mean music just is computation right that was a point that i made before in the bonus episode was that these are ways in which we can we can say that i am looking at this and that's true even of modern computers modern computers are not themselves intrinsically computational they are just they know even in the broader sense right they are just physical things whose outputs we interpret as doing computations but in and of

themselves they're just physical things doing stuff in the world i love that phrase and i'm fully with you and also i just love because that's like a pretty decent phrase we're all just ain't that the truth we're all just physical things doing things in the world and so yeah for me i love this paper i think that the answer there the question they're asking is this very narrow very technical very like you know uh nerdy logician philosopher computational science kind of person but like i i like it i like all the the you know bits in here that they're diving into all these speculations and all these crazy things. They're fun for me. I love that kind of nitty gritty stuff.

Even though I think it's just the wrong question because I think nothing's computational at the bottom level. It's just how you interpret it. Wait, so like, yeah, okay. It's given me like a very different perspective on everything you've said so far to hear you say this now which is kind of cool yeah i've been trying to inhabit the world of the the paper that's what i try to do with these things right just take on the author's presuppositions and

there's lots of papers we've read that i do not care about their topic that they're talking about at all i don't i think they've completely missed the boat but like i try to as best i can even at least talk on their terms rather than my own right right and i fail at that a lot but i i try uh well i like i completely agree right like with with with you jimmy but like it, that it feels like such a narrow viewpoint to be stuck in the deterministic computing TM world, right? That it's just... And I think you can see that throughout the paper. The whole paper is this struggle with determinism

or struggle against determinism this is or struggle against determinism and and i feel like it's just this great sign that that is like you can solve this struggle by by letting go of it a bit which is completely fine because that's you know that's what the paper was trying to do but it it makes me yearn for like a more non-deterministic view of computation and i know and i know that's going against the grain i know the the grain of the field is to you know think about deterministic turing machines but for me this shows why that's sometimes so hard when trying to pair things up with the real world. We even saw it back in myths and misconceptions. But by the way, I know the answer to the question. Is the whole universe a computer? I know. I know the answer. And they don't know. Do you know the answer? I don't know. What know the answer. And they don't know. Do you know the answer?

I don't know. What's the answer? Well, okay. Let me ask you this. If I could show you a working implementation of the universe, would you think that the universe is a computer? Like you have one? You're going to send it to me? Is it the universe? Yeah, it's the universe itself. We don't need to figure out if this is implementable. Because we're living in an implementation of the universe. Me, you, Jimmy, all of us here, we're implemented. We are here.

I think. Physical things. Doing things. Trying our best. But are we luring complete? Well, you know, I'll leave that as an exercise for the listener. Yes. Yes.

I think we'll. I think the universe is Yes, yes. I think the universe is luring complete. I'm pretty happy in life with things. Could always ask for more, but luring completeness means do I get what I want

from it? I think so. I have concluding thoughts uh no um this the the paper it's like equally exciting and aggravating to me because the part of it that I find exciting is the psychedelia of it all. I like that you brought that frame of understanding it into the picture, Lou, because that's something I hadn't put my finger on, but it is how I took it and the parts of it that resonated with me like the the bit earlier 35 minutes ago um based on how long i expect this episode to be um where i was talking about you know hey there is no difference between discrete and continuous that that's just sort of like a phenomenological thing and that's not like fundamentally there's no difference like that to me is the kind of psychedelia that this paper and this sort of line of inquiry invites. And I enjoy that very much. I enjoy wondering what are the kinds of experiments you can do to determine if the universe is computable, right? I think their experiments

sucked. I think that the experiments that they were proposing were bad experiments the follow-up thought to that was there are other experiments that are more interesting that are more exciting there's a lot of a lot of stuff in this space that i find really appealing and i talked about rudy rucker earlier there's like this whole category of sort of like you know brain teasers or like puzzles that emerge from questioning what is the nature of a computer and what is the nature of reality and how do those things intersect there's like a whole little you know phd world of of inquiry one can one can do in that space. And it made me also realize that that is sort of the way that people find their calling in life. For at least the people who pursue science and math and sort of theoretical science at least or like physics or broader than that. Like I'm not into category

theory and I'm not into proofs and I'm not into abstract algebra. I'm into the sort of human experiential parts of using a computer. I'm interested in finding a way of making programming that feels different, that there's a whole bunch of physical sensations that come from programming. And I want to understand those sensations and find other ways of getting different sensations when you program. And it made me realize that a way of looking at all of those things is those are all just like different kinds of puzzles or brain teasers or like little amusements that people gravitate towards in the same way that some people really like doing the the daily crossword some people really like doing puzzle

games like the witness that are systems driven some people really like you know the kind of building a semi-lattice of terms and you know doing doing that kind of formal mathematical stuff, that formal reasoning. And it's cool that there's this little cluster of people in this book, this chapter from this book, this paper, who are like, hey, here's a little collection of puzzles about the universe and about the computer and it's fun because i i enjoy those puzzles a little bit and i enjoy like squabbling with their particular way that they put those puzzles together in this paper but zooming out a step i think that those kind of puzzles are you know really fulfilling and i like that that way of looking at all the things that we do. It's like finding the kind of puzzles that we most enjoy and teasing them apart. Even if the puzzle is just like, what does it mean to be a human? How do you get the feelings that you want to have? How do you get people to feel a certain way together? I think when I first started listening to this podcast, I remember like binging through it and listening to the No Silver Bullet episode.

And it's like talking about coding. And this is, I promise this is related to exactly what you just said. It's not an offshoot and it's like talking about these like super fundamental ideas like are we gonna be able to solve our problems with one is there an answer you know to our problems can we find a silver bullet and obviously it's in the context of coding and programming but it feels if it feels like such a fundamental question to like, for all people, for all humans, right? Like, is there going to be an answer, like, to your problems? Is there going to be a single answer to your problems? Or maybe more than one answer, but is there an answer that solves them?

And it's like so fascinating to me that like, oh, if someone answers that question, it's like such a personal question. I feel, you know, like, right. Like it's like tells some tells you if someone's like an optimist or a pessimist or something else. And like what you're saying there, Ivan, it's like it's the same thing here. Right. We're talking about computation. We're talking about coding. But it's like it's the same thing here right we're talking about computation we're talking about coding but it's like it feels bigger than that you know like it feels like this huge metaphor i don't know when i first started listening to the show i was like enjoying it because it

to answer some of these questions about programming you have to like answer some actual like tough questions about the world how you think about the world about yourself and this paper is like super fascinating because it it really doesn't shy away from that right it's like we're gonna we're gonna ask we're gonna do it we're gonna say like computing or computation the universe people let's smush them together are they the same thing are they different and it's like confronting you with this question like is the universe a computer and they like answering perhaps the boldest way of all right and saying i don't know i don't know i don't know i don't know i don't know i don't know. Wonder. Wonder.

Wonder is the spark that ignites the human spirit. It is the feeling of awe at the stars scattered across a velvet sky. A quiet thrill of holding a fossil that once pulsed with prehistoric life. The first breath that you take when you come

into the world. And the slow dissolve into the ether when you leave it. Wonder reminds us that we are at once infinitesimal and enormous. Our bodies comprise more than a septillion molecules and are yet a mere droplet in the endless phony soup of space-time. Our lives last for but a fraction of a blink of the eye of the universe.

I wonder, as you do, has been a stalwart guide. Through the ages, we've progressively mapped out the most gnarled twists of the natural world around us. The boundary between knowledge and mystery has been pushed ever outward. But mystery is an inexhaustible supply. Each new discovery only increases the surface area of the bubble of our knowledge, exposing us to that much more of the unknown, and perhaps unknowable. What is to be done with all this mystery? For tens of thousands of years, humans have crafted myth and folklore

to help salve the itch of our wonder about the unknown. As our understanding of the world has grown, we've come up with ever more sophisticated tools we can use to bend and theorize about each new mystery before us. Fire was one of humanity's first great technologies. When the titan Prometheus defied the Olympian gods to give fire to humanity, he ignited

the kindling of curiosity, diminishing the dominion of the unknown. As folkloric murmur gave way to the bustle of civilization, humanity came to see the universe as a construct of steam and clockwork. During the Enlightenment, Isaac Newton's laws of motion revealed that the universe operated like a precise machine, advancing predictably with each passing moment. Even God himself, once the active artisan of every new day, was relegated to a distant role of setting the initial conditions from which, in his absence, all existence cascades forwards like the tumbling dominoes disturbed by an inopportune bump.

In our modern world, the frontier of technology is the computer. And in keeping with humanity's longest tradition, we find ourselves today at this very millisecond seeking to understand the natural world in terms of the computer. We can wonder like never before. Could computer simulation breathe new life into our animation of fish? With modern memory-safe languages, is there anything left to be considered harmful? If I spell the 216-digit name of Yehovah with all caps, will my code run this time? Somewhere between the thunderous throbbing of the nebula and the impregnable silence of my bedroom lies the most tantalizing question of all. Could the universe itself be a computer?

An intricate pattern of computation woven into the very fabric of reality. A computer processes information. Inputs are transformed into outputs with perfect correctness and efficiency. This process can be represented as a succession of states that proceed through the application of logical rules. The universe, too, proceeds according to rules we call physics. Every subatomic particle and galactic supercluster alike is governed by these rules, and each moment of time captures the state of all things. The laws of physics describe nature's underlying order,

but could they be more than just descriptions? Some scientists suggest they might be code, deeply fundamental instructions that govern reality itself. Consider the Planck scale, the universe's tiniest units of time and space. At this scale, the universe appears pixelated, discreet, much like the pixels on a screen. But a pixel is not a little square. Rather, it is a sample of a continuous function space, one that is discretized and interpolated, again and again and again, many times as it passes from the central processing unit, through the light-emitting display, into your eyes, and the ysiriant vacuum of your mind.

So too is the universe itself, on one hand sliced by quantum mechanics into electron energy levels and distinctly measurable planks of time, yet 4D manifold of the reals. The greatest thinkers of our generation bicker tiresomely, asserting that the universe is Four truths held in superposition Four arms of a ballet dancer twirling in pirouette Four sides of the same coin In the 21st century, mathematician John von Neumann studied systems called cellular automata, grids that evolve according to simple rules. Could the cosmos, in all its vastness, be built upon similar principles?

Unfortunately, no. There is no chance for this to be true, as cellular automata are inherently deterministic. As Einstein himself said, God does not play dice with the universe. So with the inherent non-determinism of the universe, we can therefore assert that there is no God, and thus no automaton. To say, as modern myth-makers of the Silicon Valley are keen to do, that the universe is a computer, is not to diminish its majesty. Instead, it reminds us of our interconnectedness. If the universe computes, then we are all a part of the greatest program that we will ever know. We are sandwiched together like electrons in the layers of a complementary metal oxide semiconductor field effect transistor. When you close your eyes, the cosmic computer retains a reference to your state,

so that the world is here to meet you when you open your eyes again. When you look up, you feel the same amazement that was felt by the earliest peoples when they gazed upon the heavens, the very same amazement felt by every statement and expression that gazes upon the kernel. The undocumented API of all things. The feeling of wonder. Wonder is a thirst that can never be quenched.

It is the spark that ignites the human spirit that burns us all. When, on a clear summer night, we look up at that onyx black space between the stars. When we craft a telescope to see further than our eyes alone allow. When we measure to the picosecond and the parsec. When we make the model. When we power on the cyclotron. When we simulate beyond the standard model, when we no longer need eyes to peer into the shadowy substratum, when we band together and devote our lives to pushing tirelessly at the boundary between our knowledge and the unknown.

You learn the age of the universe. You learn that there are things so far away that their light will never reach you. You learn that you are large enough to contain an immeasurable multitude of particles, quarks, tensor fields, and strings. You learn that you are so small that you barely even exist. That your life is a fraction of a blink of the eye of the universe. That all of this yearning you feel, all of the work that you do, standing on the shoulders of giants and becoming the giant yourself, will only ever leave you with more questions than answers. You wear it on your face. You hear it in silence. You will never know. You will never know. You will never know. You will never know. You will never know. You will never know.

You will never know. You will never know. You will never know.