The following is a rough transcript which has not been revised by The Jim Rutt Show or Matthew Pirkowski. Please check with us before using any quotations from this transcript. Thank you.
Jim: Today’s guest is Matt Pirkowski. Matt’s been on the show before and we’ve had a very interesting conversation and we decided to get together and chat a little bit about the consensus mechanisms that work in the blockchain world. And Matt is a relevant guy to talk about this. He’s both a tech guy and a crypto guy and like me, he lives way the hell out in the woods. So he has a chance to actually think and not be overwhelmed by what I like to call the everyday, the narrative or the consensus reality. And I think to see these things clearly it’s best not to live in the Bay Area or in New York City or Miami, but be able to sit out in the world and think about things fairly clearly. So anyway, welcome back Matt.
Matthew: Thanks Jim, great to be here. Looking forward to unpacking this topic from a highly non-normative perspective, let’s say.
Jim: In our pregame conversation, you talked a little bit about some work you’re doing on active inference. Maybe you very briefly could tell us about that.
Matthew: Yeah, sure. So at a very high level, for those who are not familiar, active inference follows in the footsteps of the free energy principle and work done by Karl Friston in computational psychiatry, trying to understand the relationship between perception, modeling of the world and how our actions in the world allow us to facilitate that development of those models through our perception of the consequences of our actions. So I’ve been trying to use some of that along with a few other people and we’ve created something called BioForm Labs and we’re applying that to the understanding of emergent organizations and also beginning to look at this technology as a way of also perhaps creating something like, let’s say a membrane around emerging systems such as these Auto-GPT type agents.
Jim: Very cool. We’ll talk about that later. As regular listeners know I spend a fair amount of my time reading and thinking about theories of cognitive science and I have just started reading in the Free Energy Principle in Friston’s work and in a month or two I should be able to at least fake being halfway knowledgeable about it. So we may chat about that in the future. In the meantime-
Matthew: Sounds good.
Jim: … Today’s topic is going to be talking about the consensus mechanisms that secure the blockchains that we hear so much about out in the world. Things like Bitcoin, Ethereum, Cardano, Polkadot, et cetera. There’s hundreds of them and they use different kinds of consensus mechanisms. By far the dominant, in fact, I think probably the only ones that are actually in production today as far as I know, are proof of work and proof of stake. Maybe Matt, you could start us off by describing at the highest level the distinction between the two.
Matthew: Sure. Yeah. I mean at an extremely high level, I think we’ll go perhaps a little bit further afield into more conceptual definitions of this throughout the conversation. But in the most concrete definitions in the way that these are manifested in the crypto world, proof of work essentially relies upon performing an energetically difficult computation to generate the security for the Bitcoin network. So fundamentally, the act of mining requires that one consumes energy to run miners, to produce hashes against a mathematical formula that is asymmetric, so it’s very hard to solve. Requires basically doing fairly simple computation a bunch of times to get the right answer. Whoever gets the right answer is rewarded for performing that set of computations, that then is used to secure the next block on the chain. And this enables the ability to block after block after block.
The accumulation of all of this work that has been done means that, based on the underlying mathematics, it becomes increasingly difficult to fake or add a block that is not within this consensus, that isn’t provably an extension of this chain of energetic coupling, let’s say, to the world. And so I think the fundamental way to look at this, and this will relate to our future conversation on this. There’s the system itself, which is the protocol which is defined in reference to its own code and the systems that it will run on in the world, but it also has a very explicit mechanism by which it links up with the extrinsic reality in which it is nested. It’s fundamentally coupled to the physical world itself. If you can’t access energy, you can’t run Bitcoin. And that’s a very interesting property that we can unpack further later.
Switching gears to the proof of stake mechanisms, these mechanisms, we could talk about the complexities of it all day. And in fact, it’s one of the reasons why it took Ethereum so long to actually get their proof of stake mechanisms up and running because it is an extremely complex process to figure out how to create similar levels of guaranteed security over something like a blockchain. To generate the ability of that blockchain to maintain its consistency, to resist capture purely through the internal acknowledgements of the people running the system itself. And so there’s conscious signaling and the conscious incentive structures within the system itself. So I would say the highest order, or the simplest way of thinking about it is can you bootstrap a system that is secure without actually maintaining extrinsic energetic connections to some other frame of reference? Can you actually, within the system itself, within the agreements made by the users of the system, develop a set of relations between those or signals within that system that by their own nature, by their own capacity to signal the veracity of the chain, actually secure the chain?
Jim: Yeah, certainly the proof of stake concept is conceptually much more difficult. Though I don’t know as much about the Ethereum proof of stake. I do know a little bit about the Cardano and Polkadot proof of steak, the so-called ouroboros protocols. And not being a mathematician at that level I’ll take people’s words for it, but the claim has been made that despite the complexity of the apparatus, it is provably secure. Do you have a view on that?
Matthew: It’s been quite a while since I looked into the underlying mathematics of this. So basically in these networks, even though they will claim to be proof of stake mechanism, typically, I know this is true with Ethereum, there are certain elements of I guess what was called the Beacon Network or is called Dot Network that rely on computational execution that is still fundamentally an energetic requirement even if it’s not necessarily shown to be that because they rely on injection of randomness, all right? They rely on a certain amount of stochasticity.
And so, at least back when I was looking into this, I kind of looked at that as a bit of a backdoor way of sneaking in the concept, the actual fundamental function of proof of work without calling it that. Because if you have to create randomness in your network, if you have to use computation to do that and you have to use energy to perform that computation, then sneaking in that randomness to generate or to secure your network is another form of proof of work. And then it has to be compared apples to apples with the mechanisms used by something like Bitcoin or the computational process used by something like Bitcoin. But then you also pay the complexity price of having all of the staked users in the system and all of the incentive alignment issues that arise in trying to herd those cats, let’s say.
Jim: Yeah. Also, I think one of the issues with proof of stake is for them to work in a practical sense. Network latencies have to be pretty low, unlike proof of work. In theory one could have all the Bitcoin miners come together once a day. Now you get some problems of double mining, et cetera, but it doesn’t really require a fast dance of communications back and forth across lots of people, while proof of stake, to be practical, really does.
Matthew: Yeah, definitely. I mean this relates specifically to, it might be a decent transition to move into my frame of reference for this, and one of the reasons why I think we can talk about the specific details of the infinitude… Not infinitude, but a very large amount of realizations or attempts at different computational mechanisms used within these two families of securing a protocol. But I think I typically try to look for frames of reference that are more deeply rooted in evolutionary biology, that are more rooted in the concepts of emergence first principles that can be used as a lens to separate conceptually what these systems are actually trying to achieve. And I think that this is a good place to begin doing so because when we talk about the requirement of a system to be in rapid communication as a property of proof of stake, or in low latency communication as a property of proof of stake, my perspective would be this actually falls out of the reality of the kind of system that proof of stake is in relation to the kind of system that proof of work is.
And the kind of system that proof of stake is, insofar as it relies on its own internal auto catalytic properties to maintain its coherence, those auto catalytic properties have to be in very highly regulated and rapid communication with one another because they’re not relying on any extrinsic cadence or clock, let’s say, regular pattern or regular gradient to help maintain that for the network. And so to the extent that any instability emerges within that kind of an auto catalytic network that is not nested within a larger gradient that provides a stability for it, that instability can spread much more rapidly. I think this is a property of many systems across biology and emergent evolution. It’s also a property that we see of the kind of thing that proof of stake is because of the fact that proof of stake makes the choice to sever itself or attempt to sever itself from reference to an external gradient that would establish a kind of contextual stability for it.
Jim: I like that. I actually never quite thought of it that way. That’s an interesting perspective. And if we do try to build a biological analogy, one that’s fairly obvious is that a, let’s say, multicellular organism is involved in a very complicated and complex dance of signaling in many modalities from oxygen, or to nutrients to removing toxins, to coordinating cadences of heartbeats, et cetera. And guess what, sometimes those complicated and complex networks break down and what happens? The organism dies.
Matthew: Yes, precisely. And then this is exactly how we can relate it to proof of work as well because then we could say something like the ability of that organism to manifest successfully. And this also pulls an active inference. The ability of that auto catalytic network to manifest its own tendencies successfully in relation to its environment enables it to procure energy. But the frequency at which it procures energy, aka it eats, let’s say, is A, far less frequent than all of the complex synchrony of internal activities that have to go near perfectly for that organism to function. It can’t have a heart attack while doing so, but it can miss its lunch and have lunch at a delayed time. Now there might be some consequences to that, but those consequences are not necessarily critical. This is because the entire organism and every single emergent layer of stability that regulates those auto catalytic relations has emerged in relation to that external relationship to a reference frame of an energy gradient.
And that predictability enables the entire system to build up resilience to variance in that relationship to available energy, available free energy across time. So the fundamental claim is something along the lines of, and I think this is where people have a hard time, we’re taught at some level it has to be the case that we observe systems from the outside in. We therefore think that what we’re looking at is a specific example of what’s in front of us as opposed to the end result of emergent systems that may share their properties with many other emergent systems that don’t look like what we’re looking at. So for example, you could say an eyeball. An eyeball is an eyeball, or you could look at eyeballs from the perspective of apertures and the aperture in the iris, or the pupil shares a bunch of functional properties or functional traits with, for example, the aperture at the mouth of a bay of water in terms of the way that the waves are changed by moving through that aperture.
And so typically people would say, “Is an eyeball like a bay of water?” No, those are totally different things from the top down analytic perspective, but if you look from the bottom up immersion perspective, they actually share a fundamental property and that will impose certain constraints and therefore induce certain similarities in their function. And when we’re talking about proof of work and proof of stake, I think you can make an analogous comparison to certain kinds of ways that closures are formed in emergent systems where proof of work is the kind of closure that you’re very familiar with, the adjacent possible. But the adjacent possible is not equiprobable at every point along the boundary of the existing rail. You could analogize of something like there might be portals at certain locations that actually open up much more, let’s say, profitable or much more patterns of exploration that are actually gateways to entirely new ways of being.
But that portal has to be held open between the previous space of stable interaction and the new space of exploration and the kind of closures that open those portals and hold them open I would analogize to proof of work and I can give a number of examples in emergent nature if we want to go through those. And then the degree to which that new space is explored by auto catalytic or possible auto catalytic relations in terms of their self-consistency and that self-consistency’s ability to then potentially link up and form even further explorations of the adjacent possible. From there, those exploratory forms of emergent coherence are analogous to what we call proof of stake. So I know that’s a very abstract frame of reference, but if we had the time or if we wanted to go through it, we could basically tie that into a number of examples that make that more clear and show that this manifestation of proof of work, proof of stake in the crypto world are just point manifestations of lower level properties that are realizing themselves through us at this point in time.
Jim: Yeah, let’s do that. That would be great. But before we do that, I should have started with this. My bad. We should also, for the audience, put a couple of practical issues on the table. One is that proof of work and proof of stake are essentially means of securing a consensus process, in the cases we’ve been talking about at least so far, in the crypto token cryptocurrency world. So I’d love if you could give us a little introduction to what the concept of the consensus is and why it needs to be protected in a cryptologically secure fashion. And then the second is the very practical issue that at least in the implementations we know of today, the amount of energy, essentially open system energy, that has to enter the system to do proof of work and proof of stake are actually quite different. So let’s start with what is a consensus process and why does it need some form of cryptological security?
Matthew: Yeah, for sure. So I think the easiest way to think about this is to ask the question, what is a clock and why is a clock useful? If everyone had their own sense of time, if we all defined our own time, there would be no utility in wearing a watch or understanding the time. The utility of the devices or the manifestations of that system are only realized through the consensual reality around the nature of time so that we can use that to co-operate or cooperate or coordinate, which is literally to simultaneously order or simultaneously enumerate. So fundamentally, time as a concept is only useful to us as a coordination mechanism, and it gives us the ability to use regular processes to coordinate our action.
In a similar way, Bitcoin or blockchain, these mechanisms allow us to generate a record of information and that record of information is theoretically protected and guaranteed to be integrous or be an accurate point of coordination or accurate piece of information about a sequence of actions through time around which behavior can coordinate. And if it can’t actually prove that history is real, that history is stable, that history is immutable, then the utility of having that cooperative artifact or that informational artifact doesn’t exist. And so the question of proof of work or proof of stake, at least in so far as it’s typically had, is about which mechanisms are able to guarantee the integrity of that representation of information across time, that sequential set of transactions or encodings across time.
Jim: And let me add again to bring this down for the audience that may not be quite as familiar with the details as you and I are, is that in the case of a cryptocurrency and related class of applications, the semantics is essentially a public ledger plus a series of transactions on that ledger. Which could either be relatively simple in the case of Bitcoin or couldn’t be arbitrarily complex in the case of things like Ethereum or Cardano. And it’s the integrity around the ledger and the transactions on the ledger which need to be consensus secured so that we can all agree that this is a ledger that we can trust for doing business on. Is that fair enough?
Matthew: Yeah, it’s just a set of linked records that make a plausible claim that some things happened after other things.
Jim: And again, to compare and contrast it to the world prior to Satoshi is that the representation is this is done without any centralized authority vetting it. For instance, when you look at your bank account, your bank is the centralized authority which claims that the ledger is accurate and that can do transactions with other ledgers in a very complicated fashion using third parties, et cetera. But the individual banks make the claim that they are responsible for their part of this complicated set of transactions and there are other third parties. But in the blockchain world, the claim is that the protocols itself take care of assuring the integrity enough for practical use of the ledger.
Matthew: And I think going back to the clock idea, it makes it easy to understand why the question of who controls the coordinative map, let’s say, is so important because to the extent that everyone starts behaving or coordinating their action using a particular tool, whether that’s the clock tower in the middle of the town or the Bitcoin blockchain or another blockchain, or let’s say a currency mediated by a political body. Whatever we’re using for that cooperative co-ordination, to the extent that it can be manipulated, to the extent that people can observe how people are behaving using that tool, whether we’re using the time or we’re using the money or we’re using the blockchain, you can start to potentially try to manipulate that to change people’s or shape people’s behavior such that it might be more beneficial to you.
And so to the extent people have those incentives, if you believe that incentives will tend to shape behavior, the easier it is for a given actor to alter history or alter the coordinative mechanism, whether that’s a record, whether that’s the time or whether that’s the amount of money in circulation, the more that actor will do so to bend reality, let’s say. Meaning, when I say reality, I’m talking about the coordinated human behavior that is using that tool for coordination or cooperation. They will use the manipulation of that tool to bend that, quote unquote, reality in their favor or to stack the deck or to put their foot on or put their finger on the scale. All of these ideas that we have encoded in sort of aphoristic language,
Jim: And even if they’re not self-serving, they could be ideological or political in the most obvious sense is that the US dollar and other nation state currencies these days are subject to changes in their rates of issuance, their sinks, et cetera, that fundamentally change the nature of our economy, at least that the margin. And those are under the control of actual humans who are turning these knobs, I would argue often not understanding what the hell they’re doing and maybe for good reason, maybe not for a good reason, but most importantly in a complex system that they don’t truly understand. While these other systems are more principled in the control of their parameters, and in the cases of Bitcoin in particular, very rigid, basically no ability to change the nature of the system itself, and which is both a plus and a minus in the Bitcoin world.
Matthew: And I would make precisely the argument that if you look at the emergent stack of complexity in nature, you will find certain kinds of mechanisms that come into being only because they can take advantage of a highly regular property of some extant aspect of the world. So if you look at mitochondria for example, and you look at the proton pumping capacity through the transport chain proteins in the actual membrane embedded on the mitochondrion. You can see that, based on the properties of electrons you can actually, after a lot of evolutionary experimentation, and it’s unclear in the field of the origin of life exactly how this came into being, that there’s some leading and interesting hypotheses. But what you can do is create a pump that generates a gradient in a membrane that actually can act as simultaneously a battery and a signaling mechanism within cells.
And so you get these emergent properties, and I’ve skipped over a bunch of complexity there, but you can get these emergent properties, but you can only get them because a mechanism emerged that created a binding with a set of highly regular properties of the electron and other contextual elements of physics in that domain. And so that’s this sort of binding closure that I’m claiming that a proof of work mechanism is an emergent example of, as opposed to other kinds of activities that are more internally exploratory or exploratory insofar as they form closures within the emergent level or emergent space of adaptive exploration that they exist within. But they don’t actually themselves open up a novel space of exploration, which I would argue is more like the proof of stake. And so the proof of stake is much more, it’s required for adaptation because it can actually, because it’s freer of the constraints, it doesn’t have as tight of a coupling with an external frame of reference.
So it can cover more adaptive ground. You’ll also see a much higher failure rate or death rate in terms of these experiments that it’s essentially trying to run. Because if you’re not constrained by some external frame of reference, you can basically posit all sorts of configurations that may or may not work. And most of them won’t, and I think we actually do see that kind of an asymmetry between proof of sake and proof of work mechanisms in the world. And I’m not trying to dis on, I actually think that there’s a necessity for that proof of stake, low binding experimentation, but it also has a high death rate. So I don’t use the word shitcoin, but that is exactly why that word emerges because people observe a high death rate or a high empiricism rate, which does also exist there. And they draw the conclusion that there’s no evolutionary utility there.
But that’s actually also incorrect because if you look at evolution itself, almost everything dies. The vast majority of new combinations, new attempts are failures, but those failures, there’s always the question of what is the necessary failure rate to actually for the entire life process to continue. And it seems to be pretty high.
Jim: Yep, exactly. So I’m very excited to go back to this, but before we go there, I will wrap up this introductory section. Remind me-
Matthew: Sorry, I keep pulling this away. I apologize.
Jim: Yeah. I want to remind you, essentially there is a practical issue here, which is proof of work, at least in the current manifestations. And again, keep in mind these are frozen accidents of history in a bunch of curious and strange ways that for instance, Bitcoin, while the percentage of its total market cap that’s spent in mining is going down for reasons that have to do with the design of Bitcoin, it’s still about $10 billion a year of electricity that’s being consumed to secure the network, which is a lot, right? It’s the amount of electricity of a medium size country. I think last I saw it was something like the Czech Republic, something like that. So it’s environmentally costly. It’s energetically costly, and as my friend Ben Goertzel likes to say, “It’s accelerating the heat death of the universe.” Being a little over-
Matthew: But I mean, any negentropic process is doing that. So I mean life itself, if you’re going to make that argument, you have to make an anti-life argument from first principles because you then have to argue that all of life is not worth having because it’s negentropic and therefore accelerating entropy.
Jim: Yeah, again, and you made the case that there is a coupling with randomness, stochasticity with the proof of stake. So there is still a coupling there, but it’s way smaller. Again, these are not matters of principle, they’re matters of practice at this stage. Think about it as at least two orders of magnitude difference to secure a similar size, similar value networks using proof of stake then.
Matthew: So the fundamental aspect of my claim is that these are non-fungible aspects of the adaptive process. These are not interchangeable. They’re not actually for the same evolutionary purpose, even though, through our narrow human perspective. I mean, let’s be honest, through most people’s total lack of understanding of almost any of these aspects of evolutionary biology, emergence or systems dynamics, complexity, it’s still a very new field. The level of understanding and comprehension there is still quite low. And especially in the world where people who are attracted to a system are very interested in using the mechanical implementation in the world of that system to generate some sort of local benefit, aka profit for themselves. And it’s not that there’s no utility in that, and I get the purpose of that, but I am making the claim that proof of work and proof of stake are symbiotic, emergent properties of general systems.
So when we’re talking about the energy utility, the utility of energy usage in a proof of work system, whether we’re talking about, and I have a number of other examples, but I’ll go back to the proton pumping example. There’s a massive amount of evolutionary experimentation that has to occur to get past the break even threshold such that you can actually create a potential, an energy potential across that cell membrane. Actually maintaining that potential is an extremely energetically intensive process. It’s non-trivial and it’s only occurred once that we know of.
But what matters, what enabled life to come out of that is the fact that we did generate the gradient, that we generated a stable gradient, and that gradient of that potential across the membrane then enabled a whole new adjacent possible that we now call multicellular life. And you can make this claim as well. Now I’ll take it way, way higher up the stack. Now let’s think about something like fusion energy. Fusion energy, we’re playing out the same game in the realm of fusion energy where we’re trying to hit break even there. We’re trying to use, and there’s a bunch of steps in between in terms of the introduction of consciousness to the process and reflexive understanding of the world and use of symbols and all of these other steps. But I’m jumping to the end here and saying like we are now consciously grappling with the same problem where we do realize that if we put in the energy, time and investment into something like fusion, but you actually crack that gradient and establish a stable gradient with something like fusion energy, that opens up a whole new adjacent possible.
And that could be analogous in its complexity and scope to the opening up of the adjacent possible of multicellular life itself. And I would say that kind of a mechanism, if we do stabilize and get sufficiently past break even with fusion, that is itself also a form of binding closure in any case, or at least it shares those properties.
Jim: All right.
Matthew: Yeah.
Jim: So, I’m just going to make an aside here. The other example, I think at least as applicable to mitochondria in trying to think about this metaphorically is photosynthesis, which also-
Matthew: That’s another one of mine.
Jim: … synthesis, which also-
Matthew: That’s another one of mine.
Jim: Which probably, current thinking is, evolved once at the most base level. Only once, right near the origin of life, within a hundred million years of the origin of life. And then has developed various versions, c4, c3, et cetera, subsequently. But the actual fundamental… And it was hard. There’s lots of ways to attempt to capture a photon that don’t quite pay off, right? And the evolutionary cycles that happened very quickly to get to the point where there was payoff, it’s the same story essentially, just another [inaudible 00:30:37].
Matthew: Yeah, and let’s go into it. Let’s unpack that because that’s another one that I’d really love to talk about because-
Jim: Let’s do it. Go for it.
Matthew: If you’re talking about photosynthesis, we’re talking about, again, we’re tapping… The proof of work analog or the binding closure analog is this process of tapping into certain properties of photons and actually then encoding or generating a process that then produces a kind of finite unit of stored energy. A sugar, right? That sugar, now we get into this interesting question, because the reason I love this example so much is it shows the symbiotic complementarity that emerges from something like that happening.
So now we have the entire world of mycorrhizal fungi that can also tap into this, that can actually perform an exchange. And this is exactly what happens in the root systems. We see an exchange between a mycorrhizal fungi network that is far more dynamic and actually makes its terminal units, like the ways that it makes contact with physical reality, are at a far higher resolution than the tree roots. There’s a bunch of minerals and smaller molecules that the tree actually needs to grow and to live oftentimes, but it can’t actually provision those efficiently for itself via its root system. It requires on this massive extended phenotype, let’s say, of the mycorrhizal system to actually create an exchange network between that sugar and then all of these other much wider dimensionality spectrum of properties associated with the molecules that the mycorrhizal network is providing to the tree or the plant.
And so what you see there, the mycorrhizal network… And this is actually, when you think of a lot of the philosophy, for example, Deleuze was very big on talking about arborescence, these hierarchical arborescent structures versus rhizomatic structures. And in a lot of ways what he was talking about, though he might not have known it or thought about it this way, was that dichotomy between the kinds of systems like a fungal network that are fundamentally these catalytic networks whose point is much more transient. It’s actually exploring and constantly in a decentralized fashion evolving across time in relation to, in many ways, its own internal signaling. And then it will eventually root into or connect up with something that has a more stable, less ephemeral structure, like a tree.
And so in a lot of ways the proof of work mechanism or the binding mechanism that binds to some deep physical regular process or set of properties acts as a coordinating central point of reference for these much more evolutionarily expansive systems like the fungi that can cover a lot more ground in terms of their ability to interact with adaptive space, and also literally physical space in the case of mycorrhizal fungi, that the tree simply can’t do because the tree is a fixed point of reference and that fixed point of reference is actually induced upon the plant by its nature as a photosynthesizer in a lot of ways. So in any case-
Jim: Yeah, that’s very interesting. And it pops another question into my head, and I hope this doesn’t take us too far off field, but I suspect it won’t. And that is the distinction which is true about all autocatalytic systems, is that they’re open and that they are energetically open in particular. And so of course if we talk about photosynthesis, the openness of that system, there are many open points around gases and water, et cetera, but the most specific one is sunlight. That the complexity that has evolved over time has been driven by the non-equilibrium energy flux provided by the photons from the sun. How do you take that concept of emergent complexity within systems that are autocatalytic, but the fact that those systems, in nature at least, all have to be open? How do you tie that back to reasoning about signaling modalities such as cryptocurrencies?
Matthew: Yeah, I mean, this is perfect because the next pivot in this conversation should be about this question of, what is the relationship between energy and information in this equation? And let’s talk about it… First, let’s address the open closed question. Because just like in physics and thermodynamics, many of the equations in thermodynamics are only relevant once you’ve drawn a systemic boundary, you have to define your system. But the thing about that is that at the edge of this, at the edge of this thing that we know as life, let’s say, there is no absolute boundary that we know of that we can prove. And so it is at the very outermost level an open system. The interesting thing though is that within that process, there are many intermediary boundaries in relation to which processes have their own boundaries defined. So for example, when you have something like a closure, like a cell, a membrane that actually forms a closure, that actually defines the system.
And so you actually, yes, you have a gradient. And interestingly enough, this is exactly, when you see things chemotaxis, you sort of see a coordination between external gradients and the internal ability to process information about those external gradients. So chemotaxis is like a eukaryotic cell detecting a chemical gradient, and if that gradient is food, it will move up the gradient to higher sources of food. If that gradient is not food, or let’s say it’s a toxin, it can try to move away from those. But that had to be developed and evolved in coordination with the internal capacities of the cell itself.
So any system is typically existing within some kind of set of gradients. When we’re talking about many of the systems we’re invoking today, those gradients are energy gradients. When we’re talking about something like the ability to… And here’s where the subtle part is introduced, because when we’re talking about monetary networks, we have to start introducing the way in which representation changes this game. Because for most of evolutionary history, most organisms were not aware of their own… They were not simulating themselves, let’s say, right? There was no sort of reflexive simulation of the game. And so they were essentially, either the system worked or it didn’t. Either they were in harmony or in synchrony with their extrinsic context, their adaptive context, evolutionary environment of adaptiveness, or they weren’t.
But when you start bootstrapping the symbolic world in which we exist, there’s simultaneously the physical world, and then there’s this whole network of symbols that we share with each other that help us to generate a distributed picture of the rest of world that we otherwise wouldn’t have access to. You can tell me about your trip that you took to China let’s say, if you went there, without me going there. And if I trust you, if I believe that you’re portraying that information accurately, you are extending my map of reality via symbolic representation. The interesting thing about that property is it can also be used for manipulation. You can also generate false realities if you want to. And if I trust you, you can take advantage of that trust to generate a false reality in my mind. Therefore, I’m operating off a false map and then you might be able to manipulate me that way.
So the reason I’m bringing this up is because when we get into the way in which we use symbols, there is an interesting property that we have to start considering, which is how constrained or how bounded or how connected to what are the actual checks on inauthenticity or checks on mutability for point purposes of abuse that exist in the symbolic realm, whether that’s language or money. And when we’re talking about something like fiat versus Bitcoin versus other proof of stake mechanisms, we have to look at that and evaluate through this lens of how easily or what kind of properties emerge with respect to the fidelity between reality and representation and what the consequences of that are. So I’ll pause there and, yeah, we can go from there.
Jim: Yeah, I think that’s a very nice pivot. So why don’t you take that on the two tracks. I’ve talked just about that, on to what degree to what you see as differences between proof of work and proof of stake and similarities with respect to their ability to be resistant to various forms of manipulation.
Matthew: Yeah, 100%. So I think what people should hold in their minds as we go down this path is maybe the difference between something like mathematics and narrative fiction. So in the realm of language we can create, for example, a set of very well-defined rules to which one must conform, otherwise one is not doing mathematics, let’s say. When we’re talking about something like fiction, oftentimes the entire point is to generate imaginary landscapes that are convincing to us, but that are not necessarily an analytic proof of anything in the world. It’s more of either they evoke a kind of resonance within our experiential domain, or they don’t. And so the interesting thing about that is it exhibits the full spectrum of how we can use symbols to go all the way from a hyper-rigid constraint based system, to something whose entire point is exploration of realms that don’t even exist.
And so when we’re talking about proof of work and proof of stake or when we’re talking about in these other terms binding closures versus exploratory closures, we’re talking about systems in the proof of workspace that are very concerned with maintaining a highly enforceable binding to an external frame of reference, fundamentally, that imposes a strict set of a strict set of behavioral constraints over the system. And that means that it’s high integrity, but it also means that it’s a little bit more adaptively limited. It’s the reason why it took, for example, Bitcoin as long as it took to do anything like NFTs.
Because fundamentally working with much tighter constraints means that if you want to have a facsimile or something that’s similar to that behavior, it’s going to actually be a much harder thing to do than if you can fabricate from whole cloth out of the world of all possible patterns in the proof of stake space, a set of behaviors that are new. It’s much easier to explore, to essentially write fiction or have imagination, let’s say, in that less constrained space, which goes back to this fundamental tension between the kind of systems that enable strong binding across layers of emergence versus the types of systems that explore possibility within a layer of emergent potential. And so that’s kind of what I would claim is the big difference there.
Jim: Okay, maybe I’m missing something, I’ll just drill in on this. I’m sure you haven’t missed it, but I may have, which is I certainly agree with you that the set of systems that are physically grounded like Bitcoin, or mathematically grounded, but then physically grounded through the energy expended to do the math, are less flexible and adaptive than essentially arbitrary protocol closures like proof of stake. However, a given proof of stake system, once it has been completed, let’s say Cardano, which is a quite clean and fresh implementation of the Ouroboros protocol. Once it’s been completed, doesn’t it have essentially the same kind of grounding that something like Bitcoin has? And that it is now algorithmically fully defined and complete and its definition of its consensus process constrains Cardano as much at the level of consensus process at least, as Bitcoin’s consensus process constrains it. Now the difference is that Cardano built much richer syntax and semantics inside its membrane before it closed it. What am I missing here?
Matthew: Well, in a word, drift, right? So there’s a big difference between a system that is able to establish closure. So it can be internally consistent, which is what these proof of stake, the entire goal of this is to generate essentially worlds of constraint based relations unto themselves that can kind of generate a free floating set of interactions that generate their own universe, generate their own universe of protocol based interaction. And in that universe, imagine it like a small fractal version of a cell that emerges inside of a cell. So it’s a set of internally consistent behaviors that can potentially start to do something useful, right? If a cell emerges inside of a cell, inside the space and outside the nucleus, inside the membrane. If some new combination of things forms a closure and that has utility, it can then start linking up with other elements or other systems outside of itself.
But the interesting thing is, as soon as it actually does that, as soon as it actually has to make a stable set of relationships outside of itself, it starts becoming more subject to the constraints that drive the evolution of proof of work systems because fundamentally what it’s trying to do is establish a stable set of reciprocal relations with some other system. And if it does so, that potentially opens up a new space of adjacent possible. So you can start… This is why my whole argument was the point of proof of stake systems is actually exploring adaptive space such that you can generate new proof of work systems essentially.
Jim: So maybe this will be the point where it comes together. So would you say that once a closed proof of stake system, let’s take Cardano as an example, has been defined and is in operation, it is now congruent with a proof of work system, an instantiation of a specific proof of stake system? As opposed to the system of proof of stake systems, which I think we both agree is a highly exploratory design space. But once you’ve actually defined one of these proof of stake systems, it’s not entirely obvious to me. In fact, it seems like it perhaps is the case that we can say that they’re both congruent with respect to the degree that they are constrained by their designs.
Matthew: So it’d be nice to be able to draw this out. So the way that I’m trying to communicate this and the way that I can visualize this is once that system, let’s say Cardano, once it becomes closed and once it, let’s say it bootstraps itself to the point where it has some sort of internal set of functions that are useful in relation to the world beyond itself, that world beyond itself is still an enabling context that is itself writing atop of some other set of proof of work based systems or some sort of energetic binding closures. This goes back to that question of inside and outside of closed systems. There’s a bunch of nested systems in which Cardano emerged as a possible candidate for itself creating a new binding to something else in its exploratory possibility space. So it’s a new possibility space. It bootstraps itself into something closed internally, but it’s not analogous to a proof of work system yet because it’s divorced still from its context. If it then-
Jim: [inaudible 00:48:10].
Matthew: If it then establishes a regular behavioral reciprocal relation with something else, another system or set of systems in that possibility space, those together can become a new kind of binding mechanism from that space into a new adaptive or adjacent possible.
Jim: Okay.
Matthew: In and of itself it’s kind of solipsistic, right? Solipsism is kind of the way of thinking about all this.
Jim: Well, I think that’s true for any signaling modality, right? I mean, I think I’m still not quite clear what you’re pushing at here because I think of let’s say Bitcoin and Cardano both as signaling modalities that the external world, the world of grounded reality and actual energy transformation and physical transformation, have chosen to use because they have utility with respect to signaling. And the fact that one is a closed system with a defined set of proof of work principles that users of the signaling modality believe are secure enough to be worthwhile to trust the signaling for coordination is not different in any obvious way in how it couples to the external world than let’s say Bitcoin is that happens to use a more physical grounding for the basis of its protection of its semantics. Well, what am I missing?
Matthew: So the drift idea, if you begin with the idea that you are going to generate a system that is immutable and non-capturable and game theoretically stable, and one of the core ways you’re going to do that is to have an extrinsic reference frame that is one of the most fundamental gradients we understand as humans, which is energy, and you begin there, you are beginning with that tether to the outside world and everything must conform to that strong set of constraints before you even start to think about internal consistency, coherence, the internal properties of that system and what you can in theory later build with that system. Everything must follow from that initial conformation to the extrinsic reference frame.
Jim: [inaudible 00:50:30].
Matthew: All I’m just saying, proof of work inverts the experimental or the developmental timeline, let’s say. It starts with the idea that you can generate a protocol that is self-consistent, yet not attached at that moment behaviorally to an extrinsic reference frame. And then based on what you build within it, to the extent that that protocol enables some sort of utility outside of itself in relation to the world beyond Cardano, it may or may not establish adaptive relationships or reciprocal autocatalytic relations with non-Cardano entities. And my claim is that only in developing some sort of stable set of autocatalytic relations beyond itself, can it become analogous to something like the proof of work or the binding mechanisms upon which it already exists or within which it already exists. Because the ability of any adaptive system to explore space that is unbound from those constraints that is imaginative relies on a certain level of stability.
And we’re actually culturally dealing with this right now as well, which is a little bit of a digression, but linguistically and culturally, I mean, postmodernity is highly relevant to this. The idea that we can take language and arbitrarily generate sets of symbols that don’t necessarily need to ground themselves strongly into reality. This is a similar idea, this idea of if you can bootstrap something, or even utopianism, you can generate a series of phrases that could be a theoretical reality and then that becomes an attractor potentially for behavior. But only if that attractor generates stable relationships with other entities in the world can that attractor manifest itself as a thing that is likely to be conserved evolutionarily or process that is likely to become part of the emergent stack that we build atop later. Every mechanism that I’ve listed, so you can talk about proton pumps and mitochondria, we can talk about the ribosome as another really interesting example of a binding closure. We can talk about this idea of photosynthesis.
Every one of these layers exhibits this ability to maintain strong symmetric constraints between separate layers of emergent complexity, and in so doing enables new degrees or domains of exploration. Now, in those new domains of exploration, the first thing that starts happening is if it’s stable enough, exploration free from or removed from those strong constraints because fundamentally that’s how you actually get novel combinations. You can even look at something like a mutation genetically as something free from its inertial constraints, of its… In theory, if you perfectly reproduced the DNA sequence at every time, you actually have far less adaptive experimentation. You could look at mutation as almost like evolutionary imagination, right?
Jim: Well, in fact, we now believe that evolution has evolved its own evolvability, that-
Matthew: Yeah, exactly.
Jim: That evolution could have created considerably better error checking than it did.
Matthew: But it was not adaptive.
Jim: Yeah, it was not adaptive, so therefore those branches that did produce lower error rates actually did not evolve as fast and got left behind. And the mechanisms are not yet proven, but it’s, I would say strongly believed by evolutionary theorists to be the case.
Matthew: You don’t want an error catastrophe, but you also don’t want to waste unnecessary energy in minimizing errors, especially if some amount of errors allows you to let’s say move around the adaptive landscape conceptually, right?
Jim: Yeah. Again, it always comes back to, always comes back to exploration versus exploitation.
Matthew: Precisely.
Jim: It depends on the nature of the dynamics of the fitness landscape, what those tunings ought to be. So there’s also a meta evolvability, which is to what degree do your evolvability coefficients evolve themselves in response to the various rates of change, co-evolutionary fitness landscapes? I mean, so-
Matthew: Yeah. And so put proof of work on the side of exploitation. And I know that word has gotten a negative connotation, but evolutionarily exploitation tends to occur when you have some sort of predictable element of the environment. And then something novel can come into relation with that predictable aspect of the environment and that relationship gives rise to something new. Now, in evolution, the environment can change and therefore adaptation has to also adjust to that. But then exploration, you can put proof of stake on the exploration side of the equation, the explore side as opposed to the exploit site. And the thing about those, the thing that really bugs me in the crypto community is that people are like team explore or team exploit in that way. And the fundamental nature of evolutionary and emergent reality is that exploration exploitation exists in a mutual relationship and the balance of exploration versus exploitation changes over time.
And that’s actually a fundamental aspect of, I don’t want to digress, but that is a fundamental part of the equations of active inference, especially with respect to being able to calculate the expected free energy in the future. And understanding a part of that computation is very specifically mapped onto the idea of explore versus exploit. So I mean, these are the deepest constraints and patterns of evolutionary systems. And therefore, at least to my mind, it is no surprise that we see that same emergent tension at the level of these new systems we’re experimenting with when it comes to how we encode and constrain and explore using informational networks in the era where it became so commonplace and essential to use information sharing networks at the speed of light, but we hadn’t yet developed any understanding really of the necessary adaptive constraints for that information sharing. And I think that’s what we’re grappling with now and that’s what we’re trying to figure out, and we’re not doing so consciously necessarily, but evolution rarely works consciously.
Jim: Yeah, certainly it does not work consciously, not non-human at least.
Matthew: Well, except as it works through us, supposedly.
Jim: And certainly in the social realm where you’re manipulating it consciously. But again, I’m still missing something. Okay. So I’ve said this a few times, but somehow we’re talking past each other. The ensemble of proof of stake possible signaling modalities is more easily explored than doing the same via proof of work, I think we both agree with that. But once a specific proof of stake closure is created, isn’t it just as constrained as one that is physically grounded? So we’re not-
Matthew: So not onto itself. So this is the thing, if you were to put it in a vacuum, in theory with a proof of stake system, you could then transplant it. But it has some sort of… And this is also what gives lie to the idea that these things are fundamentally divorced from outside energetic or informational constraints, because you could pause it, putting one of these networks in a vacuum and then saying, okay, well the entire economy of this digital reality exists within inside the Cardano network. But in reality, that network has to relate to entities and processes outside of itself. And those are something that come with respect to proof of stake, with respect to these kind of essentially speaking protocols into existence and then seeing if they stabilize.
Jim: If they couple to something useful, right?
Matthew: So exactly, you can look at the Bitcoin protocol and the Cardano protocol, each of these, if you’re defining it in terms of the fact that they are self constrained in relation to their own protocol definitions, yes, that’s equivalent. But then you add on the proof of work side another additional constraint, which is the need to be in stable relation to an external reference frame. And Bitcoin chooses energy fundamentally. And a proof of stake, in theory… And this is why I kind of said there’s an implicit reliance on proof of work based energy whenever you’re dealing with computation anyways. So all of these computational systems, to the extent that they are computational systems, and especially to the extent that they rely on algorithms that use energy to generate random seeds, they are also fundamentally writing on an energetic gradient. Now, they’re doing so less consciously or they’re trying to say that they’re not perhaps, but they are. And also-
Matthew: Not perhaps, but they are. And also by not being explicit about that, A, it kind of makes that domain fuzzier. But B, it also means that you’re going to then have to rely on some other coordinating external reference frame to eventually, hopefully, create some sort of larger emergent utility that then gets baked in.
The reason I say that is because you’re probably familiar with the idea of adaptive drift, right?
Jim: Of course.
Matthew: Why does adaptive drift occur?
Jim: Because, well, in terms of Darwinian evolution, where you typically talk about it, because there is especially smaller, the more likely you are to get concentrated results in a direction which aren’t fitness correlated.
Matthew: Exactly right? And then you could ask the question of, well, in a non-conscious system with no ability, because we’re talking about evolution, we’re making the assumption that it’s not directing its movement on that adaptive landscape, and therefore it might have a certain set of profit properties that tends to drift away from the environment of adaptedness and therefore introduces existential threat. So it drifts away from the environment that it was adapted to. There’s a divergence and that divergence introduces existential threat to the species or the entity and the species through perhaps the degree that entity’s genes spread in the short term. Yeah.
Jim: Well, actually I could put a little sharper example on it. My home field is actually evolutionary computation. And what you’ll find is what you do find is that kind of, again, optimal locations in evolutionary space, again, let’s assume relatively static co-evolutionary fitness landscapes is towards the center of the neutral network. If you drift, you can still be on the neutral network where your phenotypical fitness is the same, or not measurably different, but you can drift to a point where random moves are on average off the network are worse. If you’re in the center of the neutral network, a random move is likely to also land on the neutral network. If you’re at the edge, let’s say the outer edge of the neutral network and you have a random mutation, you’re more likely to fall off the neutral network and have a mutation, most of which are deleterious.
On the other hand, you’ve also enhanced, so in terms of a large enough end species, you want people on all parts of the neutral network. For survivability, you want a big concentration near the center. For evolvability, you want a section that’s out on the edge. So it’s when it takes a single mutation step away from the neutral network where the phenotypes are all more or less the same, you get an effect. But knowing that most of those will be deleterious. So it’s very, very similar to how you manage your bank at playing poker, actually.
Matthew: So yeah, so 100%. And I’m hoping now that after, you laid all of that out, we can use that exact example to hopefully cut this Gordian knot we’ve been having of perhaps misunderstanding or just sort of communicative dissonance, let’s say. Yeah.
Jim: We probably have different models that we’re using that they haven’t yet quite connected.
Matthew: Hopefully we can come into resonance here because basically, think of it this way, using exactly what you just said, think of a system that sets out from the get-go to establish a predictable relationship with an extrinsic energy gradient as anchoring itself to the center and its development to the center of the neutral space you were just talking about. Whereas the possibility space of exploration, of proof of stake, because of the fact that you can bootstrap the protocol into existence, the initial point of development of those networks might be much further from that neutral space. And so therefore, when they come into existence, if they establish a closure, just like you’re saying, they might be closer to that edge boundary where their movement in any given direction or the relationships they establish once they become coherent, could lead to death and often likely lead to death because they are around that periphery.
And then the question you’re asking earlier is, well, once they come into existence and they actually do start acting as agents, coherent agents on the evolutionary landscape, why is their set of constraints any different than the entity that is anchored from first principles to what it has considered the middle of that space? Because the center of that space, in theory, the hypothesis for something like a proof of work or a Bitcoin network, is that the center of that space is actually a connection point to an extrinsic reference frame that we can depend on and that we-
Jim: Yeah. Okay. So this is where I would say I don’t agree at all. If I’m hearing you correctly, I see no reason to assume that the anchor point of, let’s say Bitcoin is in the center of a neutral space for cooperative quasi financial currency type signaling. It’s just a space, it’s arbitrary. I would say there’s no reason to believe that Satoshi’s Bitcoin design, as brilliant as it is, and I remember when I read this thing, myself in 2008, I slapped myself in my forehead and go, “This is brilliant. And it’s not even that hard. Why didn’t I think of that?” Just as Thomas Huxley supposedly said the same thing about Darwin, but there’s no reason to believe that that particular design is in the center of any particular co-evolutionary fitness neutral network. It’s just a spot-
Matthew: But it’s a stable spot and it’s like it’s oscillation. So think about why. So first we start, let’s go back to time for a second. We created clocks, right? Clocks is coordinated mechanism. But then over time, what have we made clocks? What are our time standards based off of now?
Jim: Oscillating crystals or speed of [inaudible 01:06:40]. I think today they’re officially speed of light across the width of a particle or an atom, not quite sure which, but I think it’s now actually speed of light across a a fixed distance at the subatomic scale.
Matthew: Yeah. And I’m just like you, I’m not exactly sure where we’re at right now cause we have been pushing it. But the point is the trajectory and the direction we’ve been pushing it towards more and more predictable, stable phenomenon to which we can anchor our coordinative capacity based on this idea of time that we have induced upon our worldview. And we want the most stable possible oscillating or measurable, observable upon which to build that entire edifice of temporal cor coordination. And what I’m getting at is that that’s not an absolute center perhaps of anything.
But if you’re going to choose a foundation upon which to build infrastructure, then having a stable foundation is highly useful. And so in the same way that it’s not inherently the case that evolution discovered a fundamental center in the electron when it created the proton pump. But because essentially the electron or proton basically has extremely stable properties across time, we exist in a world that emerged out of that arbitrary closure, maybe not arbitrary, but whatever closure emerged in the membrane boundary of the primordial ancestor of the mitochondria, we now exist in its causal aperture window, so to speak. Causal in terms-
Jim: Yeah, yeah. It became a frozen accident and then pruning rule for what came next, right?
Matthew: Yeah. But if it’s behavior was arbitrary, whatever was built upon it would be far more ephemeral. The reason why life itself is not as ephemeral as in theory could have been, and why we still have this whole unbroken lineage of reproduction is because the properties upon which the proton pump landed were so highly stable.
Jim: Well, not only stable, but produced the surplus. If they didn’t produce… And they were no doubt earlier versions that did not produce a surplus. So it produced enough surplus to exist against stochasticity essentially. And so it was a good spot in design space to build from. So then you turn that back around and say is let’s say bitcoin a good place in design space to build from. And I’d actually say for the purposes of currencies, it’s not, it’s actually a terrible design point in design.
Matthew: But regardless of what someone thinks about Bitcoin specifically, I think the question is more along the lines of is energy a deep enough concept upon which to build a system of coherent symbolic representation that we can rely on? And I mean, I think some of the clues to the answer being yes are, for example, the equations of information theory and thermodynamics having such tight relationship, for example.
Pairing our representational or anchoring our representational capacity in energy regardless of how one chooses to do so in a point specific manner. Maybe it’s not Bitcoin, but Bitcoin is an interesting initial example. I think that energy seems to be, at least in my mind, the leading candidate. So I have a hard time, perhaps it’s possible to do so otherwise, but I think that if you try to argue that proof of work and proof of stake insofar as crypto goes are fungible, and you can replace proof of work with proof of stake, what you’re really trying to make the claim, you’re trying to say that you don’t actually need to tether into energy. But then what I’m saying when I say that you can’t really do that is that, well, you’re saying you’re relying on people and their behavior or their computers, and fundamentally that ties into energy anyways. You’re just sort of skirting the point. You’re not fundamentally allowing yourself to make the claim. And I think Bitcoin at least allows itself to make the claim and says “Energy is fundamental whether we like it or not.”
Jim: See, my read of Bitcoin ain’t nearly that strong, that essentially it is that Satoshi found a very, very clever way to bootstrap a radically trustless network at the cost of energy, no more, no less. Not that energy per se was the a metaphysical correct way to anchor such a system, but that it was possible to create such a thing. The first one, right, first one. But
Matthew: It’s not metaphysical in the sense that we exist in, so our earth exists in an energy gradient. Life exists because the earth exists inside the energy gradient. All of the complexity exists for that same reason. And so the question, if all of life to this point has fundamentally acknowledged the reality, even if not consciously, fundamentally acknowledged the tropic reality of that energy gradient, then it at least to me seems to be a safe bet that we should make sure that our communication and representation and historical, historical recording systems of that information also fundamentally tie into and respect these energy gradients.
Jim: Yeah. See now that I see where we’re at, I can see why I disagree. I had an intuitive understanding, but I didn’t quite ground it because I would suggest that if I’m correct, that what Satoshi did was he found something that worked and that Cardano also found something that worked, which is able to bootstrap radical trustlessness from something. In this case, it is a set of moving parts that the experts at least claim is secure. And I would say we have empirical evidence of it. What’s the Cardano market cap these days? $30 billion or something. If it didn’t work, somebody would’ve stolen it. Or if it was easy to break, at least somebody would’ve stolen it by now. And so that the fact that one magic spell to produce trustlessness depends on energy and another depends on a set of relationships between an internal moving parts is immaterial. I don’t see the material, I don’t see the materiality of the nature by which radical trustlessness is achieved. It’s the achievement of it that is the significant line that the two cross happen to cross in different ways.
Matthew: I agree that you have to achieve that as a fundamental constraint to maintain yourself as such a system across time. There’s question of, so you can certainly create a self referentially consistent system that is more difficult to destroy than the majority of representational or computational systems that we’ve created. That’s definitely true. And Cardano seems to be one of those. I still also don’t think though that it exists in non relation, for example, to external constraints. So one of those is Bitcoin and others of those are actual computational requirements and energy usage. Others are certain sorts of interesting psychological human factors that are required to maintain their stability across time for people to keep using this network. And so again, I think that it is maybe they have found a adaptive niche that is at a really within a goldilock zone between that neutral center or the sort of ostensible neutral center that Bitcoin claims to be and some other adaptive periphery that is totally of no utility.
And perhaps they have found that. But my entire point is if they have, what you would expect to see is first of all, it would not replace Bitcoin, it would potentially still float or whatever the fundamental proof of work system would be. It would still have to anchor itself into something outside of itself, which would most likely be Bitcoin. So then you start looking at Bitcoin as a fundamental encoding layer of this entire emergent stack of protocols, which is how I tend to see most of these emergent protocols in the proof of stake space anyways. And the information that is of highest utility in a network like the Cardano network will not stay in the Cardano network. It will leak out into or be encoded in whatever that network deems to be the most stable internet work coordinating layer. And therefore something like Bitcoin becomes the internet work coordinating layer between all the proof of state networks.
Jim: Yeah, I don’t buy it, I don’t see it, I don’t see the Bitcoin as being… It’s just an arbitrary spot in design space. In fact, I would argue a pretty poor one, at least in the currency use case. And it happened to be the one Satoshi thought about and has no particular priority of wonderfulness for that reason. If Bitcoin disappeared tomorrow, let’s say, I don’t know how it disappeared. Let’s just say it disappeared.
Matthew: That whole space will collapse when Bitcoin collapse.
Jim: I don’t think so. I don’t think, let’s say Cardano is an internally consistent system and to the degree that it can couple to the outside world and produce utility, it’s like the dollar. The dollar is, as we know, a very, very creaky foundation based on some crazy shit. But so long as it is actually useful to people, it’ll continue to persevere. Cardano’s vastly better designed than our central banker managed fractional reserve banking system and should be to the degree that it were able to replace, let’s say the dollar and it’s very, very creaky architecture to be used as a signaling modality for coordination of production, distribution, consumption, savings, investment, and innovation. It would be just fine. It doesn’t need Bitcoin at all. It’s a signaling modality that couples to the real world that people find utility from, or at least in principle could.
Matthew: So my prediction would be that if we see something like the fundamental proof of work network collapse for whatever reason or break its tether to extrinsic reality, the next largest or the next most stable entity in that chain would likely we would fall back upon exactly what we have now, which is the most stable mode of enforcement being the capacity to hold physical territory and impose the regularity of behavioral constraints through force, laws, enforcement, things of that nature central coordination. I think that if you try to imagine what it would look like to run a large scale set of coordinated human activity through a system that is only self-referential, what you’re going to end up getting is the same kind of political games and the same kind of self or reflexively driven behavior that becomes parasitic eventually that we see within the West Australian state system.
Jim: But I would say that would be a question on whether the proof of stake system was mutable with respect to its parameters or not. If it was designed such that its parameters were not subject to arbitrary political capture, then it would be just as robust as Bitcoin.
Matthew: So if you’re going to try to freeze it in design space, then we’re also talking again about the adaptive to the extent that it can no longer adapt in design space, it will be more likely to suffer a death from adaptive drift for the same reasons you discussed earlier.
Jim: Now let’s go up one level. So let’s take Cardano as a point in design space and proof of steak space. But we have the wider game of steak systems, so we have Cardano, we have a bunch of other ones and they will compete with each other and there will, and let’s even assume that they’re all immutable with respect to their parameters, which some are and some aren’t. And there will be evolution, some will live, some will die. And let’s call it the signaling frequency through those nodes. Let’s call each currency a node will change over time as better design currencies are used by people for purposes, for different purposes.
And over time the fitness, literal fitness of the node of the currency to the coordination task that humanity has be a better and better fit because the transaction will move to those currencies that do a better, less expensive job of doing the signaling. And so you still have the evolvability advantage of the broader space of proof of stake systems while each one of them, each instantiation, each object brought into reality in the design space could be immutable. So you have the best of both worlds.
Matthew: So my [inaudible 01:21:04] would be like let’s imagine a world in which all of the explicitly proof of work systems are subtracted from that process, that set of networks, processes. And all we are working with is the emergent adaptive landscape at the order of the proof of stake experiment dynamics and adaptation. The claim that is forced upon me by my logic here is that what you would observe is to the extent that those systems must reference extrinsic reality, which they have to do, even if implicitly, you will begin to see whichever system is able to prove itself most effective in its relationship and managing its relationship with extrinsic reality, perhaps that’s running its network efficiently and with stability in whichever way it achieves, that could become the central node in that network.
Then you will see a, again centrality likely, it’s sort of meta centrality in this adaptive landscape. That meta centrality will direct attention increasingly and incentives increasingly towards the efforts to corrupt either intrinsically the protocol or capture the relationship between the protocol and extrinsic reality. Perhaps at that point that is whoever is running servers or whatever energy relation it might have. And at that point you fall back on the question at the game theory of how resilient is the network to capture at the level of the actual physical implementation of the computers it’s running on, which is this question of it becomes a question of game theory, geopolitics, et cetera. And I still think that fundamentally the game theory in relationship to the fundamental dynamics of energy is the most stable solution when you fall all the way back to that failure mode. So that, that’s the simulation as far as I simulated in my head. Perhaps I’m wrong. That’s my hypothesis. Those are the inferences based on my assumptions generally.
Jim: I think I now understand what you’re saying and I would say I disagree. That I have to think about it more. And I would say slightly stronger than just disagree, is that because Bitcoin is such a piss poor place in design space, that the fact that let’s say radical trustless achieved with a self-referential system, that parameters of operation are incorruptible or at least practically incorruptible via let’s say political capture, the equivalent, would lead me to believe that over time these proof of stake systems will out compete Bitcoin in particular.
Now of course it’s possible that there could be other proof of work systems that are closer to a good place in design space. Bitcoin’s almost useless as a currency, you do not want a currency that’s deflationary for a bunch of reasons. The biggest reason of which is a deflationary currency encourages people to use the currency as a store of value. And the last thing you want is your currency to be used as a store of value. That’s a sign you have a mal design currency because wealth stored in currencies is sterile, it’s not invested in productive assets, it’s not invested in innovation, it’s not invested in anything. It’s literally the same as being $100 bills stuck on your mattress.
Matthew: Well, but again, this goes back to the point that typically whenever, that’s fundamentally a store of energy for future dispensation or utility. And most successful systems that we have examples of historically seem to suffer at their own hand when they succeed in the way that we see, for example, with global reserve currencies. Because the global reserve currency attracts precisely the dynamic that we’re talking about here, which is the desire of other nations to store wealth in that currency.
Jim: And I agree, by the way, the US by way better off in the long term, if we got the dollar out of being a reserve currency, it allows us to cheat too easily, which we are doing, which makes us weak in our moral fiber, in our productive capacity where essentially rent extractors around the dollar being a reserve currency. And that’s a bug not a feature.
Matthew: Yeah. But the same social patterns seem to crop up in any civilization that tends or has become a global reserve at that point in time. Which is why I think that any system that is fundamentally to the extent that we can put the dynamics of that capacity for storing value outside the hands of human control, and it can just be something that is fundamentally stable and runs on autopilot and removes… Because for example, you don’t have to, my vision of these cryptocurrencies, I don’t see any of them being totalizing. I’m not someone who thinks that even something like Bitcoin, because my role, like I was saying, I think there’s that exploration and exploitation balance that is always [inaudible 01:26:30] .
I think that you can have something like a very strong link to energy that can act as something like a very difficult to corrupt sore value that is a meta coordinating layer that doesn’t necessarily have to be the daily currency that everybody uses, but that also makes it much harder for any nation state or any group like you’re saying, to leverage the position of having their currency that they control and manipulate arbitrarily oftentimes be the manner in which everyone else tries to store whatever excess value that their energy expenditure has created in the world for future use.
I think that that utility, even if that utility alone is what ends up being offered by whatever system mediates between energy and our information representations for monetary uses, even if it’s just that, I think that that’s extremely adaptive compared to where we’re at today. So that’s my perspective.
Jim: Yeah. And I do agree with you that getting our coordination modalities out of the hands of self-serving politicians and bureaucrats would be a hugely good thing. But I also would suggest that the design matters and in particular, and maybe this is an aesthetic, but I think it’s deeper than aesthetic, but at least put it out as an aesthetic. I think that the coordinating singling modality should absolutely minimize the store of value attribute. That value should be stored in productive assets beyond the minimum necessary to adapt to short term fluctuations. And that money in a mattress is literally an anti-social act. And hence investing in Bitcoin, at least holding it raw and not using it in a way that is an investment or like a lend is a fundamentally anti-social thing to do. And that I would suggest a good system would not be deflationary because deflationary, just think about this, if our dollars were deflationary, it would literally make sense for us to take $100 bills and stick them into mattress and more and more of the capacity, the surplus.
Because the most important thing as the cycles turn year to year is one, how much surplus is graded? Then the second, how intelligently do we apply the surplus? Why did the West beat the Soviet Union? They actually had a higher savings rate than we did most of the time, but because of the fact that their pricing signals and their whole method of organizing production was so piss poor their return on the investment of their surplus was grossly worse than the West’s return on its surplus because we used almost completely decentralized individual investment schemes to not reach optimality, but reach a fairly good return on investment of our surplus. And that is the macro end result we want from our economic system and our currencies are merely a signaling modality to achieve that across the chain of production, distribution, consumption, savings, and investment. And hence the design of the institution matters-
Jim: And hence, the design of the institution matters greatly. And Bitcoin is a very badly designed institution for the purpose of putting our surpluses to good use.
Matthew: Yeah. So this is actually somewhere where I will significantly diverge from anyone you might consider someone like a Bitcoin Maximalist or whatnot, which I’ve never considered myself anyway, so not odd that I diverge from that perspective. But just to give you a frame of reference, oftentimes it is proposed in that crowd that purely deflationary concurrency is a sustainable path. I think because from my first principles’ perspective, when I try to run that through, when I try to simulate that, it seems to be the case that like you’re talking about, money is in theory, we’re trying to do multiple things, as multiple purposes. One of those purposes as a coordinative set of informational symbols that allow us to further generate non-zero sum or positive sum outcomes, to the extent that we believe in the fact that positive sum interaction can exist, we are implicitly making the claim that you can actually create… Bring more of potential into reality.
And so you actually have more stable functions or processes or things that we can actually work with. And so if we have a finite amount of symbols to represent the things, to some extent, you start getting this asymmetry and you start being able to cover less of exploration space or adaptation space of that positive sum creation process with your monetary symbols, which is exactly… That’s the abstract way of talking about what you’re talking about with respect to not being able to realize the many possible potentials in the world if all the money is sitting in mattresses, right?
Jim: Correct.
Matthew: And so in theory, the Keynesian perspective or the way we went about trying to solve this is like, okay, well we need to try to understand when there’s real underlying growth and therefore more stuff and therefore create more money to be mapped onto that more stuff, so that we don’t suffer from those collapse cycles associated with there not being enough tokens to put together the pieces of new potential that we need to put together to keep the growth going.
So I think that… I don’t see it as either… Again, I see it as a multi tractor system. I think that we don’t have… I like the idea of having Bitcoin as a deflationary tractor intention with the naturally inflationary tractors that will always exist, attendant, any real growth. Because regardless of what Bitcoin does, if there’s not enough money in this circulation and we’re actually having real growth and we actually have deflationary crashes, people would use anything as money. People will start creating and fabricating… This is what we’ve done in the past. People will start generating new money, new tokens, mapping that to value so that they can actually represent what they’re trying to do in the world to create novel, either to meet their own base level needs or to create novel value that others might recognize as worthy of those value symbols.
So I guess the way that I see it is that I think Bitcoin, I don’t think that it’ll be the only system. I think that it’ll likely exist in juxtaposition with an inflationary attractor. I would like that inflationary attractor to also have better properties than our current state-based human, highly malleable, highly corrupt modes of mediating inflation through extremely reductive and simplistic levers that we pull, often too late and in ways that make the situation worth worse rather than better. So yeah, I mean that’s kind of where I’m at there. I don’t see it as an either or. I see having a stable in the same way that having a stable ability to understand time doesn’t necessarily mean that people can’t choose how to use their time. Having a deflationary attractor doesn’t, in the system, doesn’t mean that everyone has to be orbiting the deflationary tractor all the time, but they might store a lot of their wealth there.
Jim: Yeah. See, I think it’s a horrible thing for people to store their wealth in Bitcoin and is literally an antisocial act for the reasons I laid out. Well, on the other hand, I agree with you that our current systems are stupid, easily abused, send very confusing signals that are often out of sync of what we really want, et cetera.
Matthew: Can I ask you a question then?
Jim: Yeah.
Matthew: How much shielding from entropy should the value we create or the representations of the value we’ve created with our time and energy and labor have? Because fundamentally storing value is like we’re shielding that from entropy such that it can be used at later points in time and not subject to deterioration or destruction or co-option by some other actor. So do you think that there’s an appropriate level for the degree to which the value we accumulate should be shielded from entropy?
Jim: Yes. Those are two different questions here, which unfortunately gets confused when people start thinking about currency and wealth. Keep in mind, currency is not wealth. Currency is at most a pointer to wealth. Wealth is land, wealth is factories, wealth is human capacity, wealth is a truck, wealth is a road. These currencies are merely pointers. They’re not wealth itself, and having a big pile of Bitcoin is not wealth at all. At most, it’s the equivalent of a capacitor on electrical circuit.
Matthew: When you think of that as realized energy with utility, which is what you’re talking about as wealth. You take potential energy can convert it into a road, and now it has road utility and you can move from stock to a flow and then you can basically extract value if you wanted to. Or this is what people do with toll roads is the stock flow becomes a flow of capital a top the…
Jim: Hell, even your dining room chair is providing value. It holds your ass off the ground every day for 200 years. So it’s producing value.
Matthew: Although the value in that recognized or the positioning of that chair in your home and as a commonplace item that’s relatively easy to replicate as a chair, means that not many people are going to come to pay you to sit in that chair.
Jim: But they’ll pay you for the chair.
Matthew: It’s some marginal transactional value.
Jim: Probably worth a hundred bucks. So that’s a hundred bucks worth of utility, literal utility used.
Matthew: But what I’m getting at is that the store of abstract wealth, let’s say, is very analogous to a store of potential energy that can be fungibly transformed in theory into these realized forms of wealth.
Jim: Okay, perfect. Let me drill that.
Matthew: So should we be able to have personal stores of abstract potential energy? Is that not something that…
Jim: Well, from a systems design perspective, we want the correct amount of capacitance in the system. Just think about how our electrical grid works. We have almost no storage today, probably too little. We want some storage on our grid so that we can take better use of intermittent non-carbon sources like solar and wind and such. But we don’t want a system where people hoard energy and batteries for… Because somehow the pricing of electricity has become deflationary and eventually the whole country’s taken up with batteries and everybody’s saving electricity for no good reason other than a flawed institutional design. So there’s an optimal amount of capacitance, literal capacitance in the electrical sense that one wants in a monetary system, so that for the same reason we have capacitors in our electrical circuits so they can handle fluctuations. They handle small scale voltage fluctuations, for instance.
Matthew: I mean, we can say that that optimality exists, but it’s likely not computationally reducible. We have to actually compute it as we go by all of the processes, distributed processes and interactions that we have along the way. And so it’s like… I think that at least from my perspective, I value having a system in place. I think that fundamentally it is a desirable property with respect to future possible emergence and adaptive behavior to have a system in place that instead of trying to compute the global optimum, identifies a stable phenomenon and binds to that stable phenomenon as a point of potential coordination.
It doesn’t mandate coordination around itself, but at least it acts as a… I mean, this is why the memes exist that portray Bitcoin as a stable beacon. That is just pulsing with a cadence that is just regular amidst the sea of chaos around it. I think that kind of potential, that kind of an attractor within the system that we have today is something that has a great deal of utility. And I think maybe some people do better or worse at trying to compute those global systemic optima. But that’s a really hard problem. And I think the failure rate on that is really high.
Jim: Yeah, I would say I started fooling with Bitcoin when it was worth the 10th of a cent. I still have little fractions of Bitcoins back in the day when if you sent an email to an address, they’d send you back a millionth of a Bitcoin or something, right?
Matthew: Yip.
Jim: And I’ve got them on a chain on a computer someplace, up in my dead computer warehouse. And the only reason Bitcoin isn’t worth a 10th of a cent today is the collective hallucination aspects that have occurred, which could easily go away. It’s not useful actually. How many people actually bought something with a Bitcoin? A few people probably have. I never have. I don’t know anybody that has, but I’m sure there are a few. But it’s, again, it’s mal design. The transaction costs are too high. It’s a fucked up currency, right? And that the only reason it works is a store of values if there’s greater… An increasing number of people who believe that it will be a store of value. So it’s a shelling point actually, and it’s only a shelling point, so long as everybody decides to go to that club every Thursday night. The moment they stop going to that club every Thursday night, it’s back to a 10th of a cent. It’s a curiosity item.
Matthew: But so is our concept of time and so is our concept of calendars. And so all of these things are fundamentally shelling points that we bootstrap around some regular phenomenon.
Jim: On the other hand, land is not. So if you wanted to store your economic surplus in a relatively fluid medium, buy good quality farmland. Go up and down in price to some degree, but you can always sell good quality farmland. And it is actually a productive good, you have not put your money in the mattress. And of course in today’s financial industry, you can buy shares in mutual funds that own high quality farmland so that if you can’t afford to buy a farm yourself, as most people can’t, you can.
Matthew: Sure, I mean its high scarcity, but very low fungibility. So it’s very hard to have a farmland exchange program that can be used to generate the capital flows necessary for trying to create new things in the world. It’s a harder problem. It could in theory be done, but you know…
Jim: If it’s done right now, there are mutual funds in farmland. So let’s suppose I had $300,000, which I didn’t need right now. I could invest it in the farmland mutual fund. I needed to pull $300,000 out to start my next company. I have 24-hour liquidity on the mutual fund. Take my money. The guys that are managing that group of farms, they have their problems of dealing with the time things. And so he is still able to act in a liquid, relatively liquid fashion a day, maybe it’s a week, but in a short period of time to be able to get my assets back into a truly liquid fund for a short period of time. I say sell my interests in farm fund and I put it in my bank account, or if you want USDC, I don’t really give a shit. And my aim is as it is in my own personal way, I manage my own portfolio, is to minimize the amount of cash I hold, right?
Because USD, our state’s dollars is an depreciating asset. You definitely don’t want to hold much of that motherfucker. And so I want to get it off the books as quick as possible. So I might hold it for two or three days and then I won’t liquidate my farm mutual funds until I know I’m about to invest $300,000 in a startup and maybe it sits in my bank account for three days, which is slightly suboptimal, but not worth me worrying about. And so that is actually a monetary system, oddly enough, despite its piece of shit nature that causes the correct behavior, I would argue, which is to minimize the amount of asset held in money. It’s not to zero by any means because there is an amount that’s needed, but to minimize the amount of money, and specifically in quantitative terms to maximize the velocity of money. When V goes down, there’s something wrong with the monetary system.
And if you look at V until very recently, V has been going down for 30 years, it used to be two, it got down almost to one, and now it’s creeping back up again as the money supply kind of surprisingly is shrinking again, fairly rapidly. And that was going to have some weird effects too. But if you look at the V of Bitcoin, the V of Bitcoin is 0.001 or less. It’s hardly used at all in actual real non-financial transactions. And so by that measure alone, it’s just a completely crappy currency for encouraging the behavior of keeping as little as possible in dead sterile stores, in as much as possible invested in different classes of assets with different degrees of liquidity such that one can make the moves one wants in the time one wants to.
Matthew: Yeah, I mean we’re unpacking, we’re continuously expanding the territory. I mean the externalities associated with monetizing land is a fascinating potential conversation. I think also the velocity question is a question of… If you’re talking about, again, there’s this question of are you trying to have a entire global perspective on optimal V, or are you okay with having something like a multi-tiered system that has different strata at which different V have different relative functions in relation to other possible, let’s say variables that aren’t just the velocity of how transactions are moving in the system?
So I think at these deep levels, again, I keep wanting to come back to time because the ability of stability or the ability of predictability of systems is a property that’s not necessarily the velocity of transactions that they enable, but is a property that matters. I mean, again, I think that a lot of the corruptibility of these higher order, high velocity systems comes from the fact that they do not have tethers to an extrinsic frame of reference that are beyond the ability of individual humans or coordinated networks to manipulate or corrupt those systems.
And so regardless of the degree to which V is low, I think that the ability of something like a deflationary system that is highly predictable to establish a baseline that may be considered like a rising tide above which human coordination has to continuously maintain a certain degree of fidelity or connection to reality. So to the extent that you are less… Let’s say you’re making the argument against the efficiency of the Bitcoin network, well, at the very least that gives us an interesting lower bound with respect to the utility of energy usage. If you’re less efficient than the Bitcoin network, that’s a pretty strong signal then that perhaps that energy is being wasted and it should just go to that actual network to preserve that potential energy associated with that value source. Even if you know what you’re saying, and I do believe what you’re saying has truth in it, in the sense that we don’t want a huge proportion of our abstract representations of value just sitting there doing nothing because we fundamentally will just do less at that point in time.
And maybe that’s so much less that it becomes maladaptive. But again, I don’t know. I can’t predict where it’s going. Computing that function is extremely complex and difficult. But the first principles that I at least think are worth adhering to, demand that we have at least some way of computationally respecting some fundamental constraints within which we are embedded. And it doesn’t have to be Bitcoin, but at least as when I look around the landscape and I ask what systems out there are actually trying to establish and maintain a coherent extrinsic frame of reference to something as fundamental as energy, I just don’t see many systems. So that’s kind of where… That’s my perspective.
Jim: I hear you. But I don’t see the significance of it. I would say that the only significance of Bitcoin is that Satoshi found a way to use energy to achieve radical [inaudible 01:48:01] and end of discussion. And the fact that it was energy and not by praying or something else is immaterial. That it’s the achievement. Actually two parts, and again, we’ll have to sort of wrap up here soon. One is radical [inaudible 01:48:19] and then theoretical immutability of the parameters. And yet we know that isn’t quite true. I don’t know how many times they’ve had to fix the Bitcoin blockchain when it’s drifted apart, when coins were minted in very short time, synchronous identical coins minted a few milliseconds apart, and there has to be a manual restitching the system. In the early days, it happened at least four times. And so the alleged immutability of Bitcoin isn’t quite what it’s claimed it to be. There is a…
Matthew: Well, I mean it’s also proportional to how early it was in the system’s history, right?
Jim: Yeah, so there’s a social component.
Matthew: Like an act like that… If you were to actually do that today, you could be an existential threat. If programmers were to attempt to do that… But then also the likelihood is that the system, the system’s evolutionary trajectory would simply move away from that branch.
Jim: So people would just fork it. And then of course Bitcoin is forked, I don’t know, 13 times, something like that. And if you add up all the forks, I think they add up to about a 10th the value of Bitcoin itself, which is pretty amazing. So it’s a very weird thing.
Matthew: Yeah. I mean, they are residual evolutionary branches and I think that’s a natural phenomenon. Again, I think that as you were saying, I don’t think that it should be frozen. The parameters can’t be frozen. Just like you can’t freeze a genetic sequence and expect that species to continue to make hay, so to speak, evolutionarily and adapt. Profitably, you can’t expect any of these networks to, if they were completely frozen, continue to adapt either. So yeah, I think these are all… I mean, they’re interesting parameters to play with. My fundamental… I mean, I’m glad that we were able to explore this duality that would be this possible duality in the way that these kinds of systems map to explore versus exploit tensions in emergent systems. I don’t have a strong felt desire for any one system to be the system. I just am trying to surf the wave and spot the patterns and think through it in terms of the weird first principles that I hold due to my sort of [inaudible 01:50:32] into emergence and complexity and my own personal…
Jim: Yeah, it’s been a wonderful conversation. It’s caused me to think about things that I had never thought about before. I just love, love, love your pretty close analogy between these systems and mitochondria and photosynthesis. I think that is very fruitful. It’s going to give me some nightmares and some thoughts for a while, I’m sure. So I very much appreciate you doing this.
Matthew: The one that we didn’t do that’s also really interesting is the ribosome protein layer. And it might even be the most relevant, but maybe for another day.
Jim: And unfortunately it’s what I know not enough about. I know a fair bit about photosynthesis and a fair bit about mitochondria, and I know what a ribosome does, but I don’t know about its internal mechanisms at all. So I’d have to study up on it for a day or two probably, if we’re going to go down that road. But yeah, I think this is very interesting. And I will say at the bottom line, I am not… Even though I think Bitcoin is anti-social in principle, I think it should be allowed, right? I think there should not be a crackdown against Bitcoin. On the other hand, they should make sure the motherfuckers pay their taxes, which most of them don’t and such. And that as a shelling point, if people believe that they want to put their money in a imaginary collective hallucination built around Bitcoin and all the peculiarities it has, great, people do it.
People buy lottery tickets, people go to casinos, people do all kinds of weird ass shit, right? There are even people that hold lots of USD. One of the stupidest things I could possibly imagine doing. But I would not outlaw Bitcoin at all, but I would bully Pulpit that it’s both antisocial and probably way less secure than people think. Because I didn’t even get to the two killers, which is… We both know Bitcoin has two brittle failure modes. One is the 51% capture and the second is cracking of various crypto algorithms using quantum computing. Either of those two things happen and a fork will be forced to deal with them.
And probably some parameter changes, certainly some parameter changes in the quantum case. In the 51% dominance case, it may just be they’ll have to continue to fork to keep ahead of the swarm and perhaps change the algorithms and make people’s hardware obsolete, something like that. So Bitcoin ain’t as Bitcoin as the Bitcoiners think it is, and for it to exist as a selling point for suckers to go and coordinate and fool each other, great. But I think it’s a bad idea and I encourage people not to engage with it.
Matthew: Well, I think that’s another one of the beautiful things about evolution is that it conserves a very wide variety of perspectives. And ours obviously diverges here. And I mean, I think place to some extent the utility of a deflationary tractor is advanced through or to the degree that they corrupt in our current reality, especially in our [inaudible 01:53:48]… Around our [inaudible 01:53:49] tractor. We’ll try to kill it. I think it’s hard for me to imagine anything more antisocial than the kind of [inaudible 01:53:58] effects that emerge from the ability to have a highly fungible mechanism by which money is brought into existence that makes its way through the same channels that have the ability to change its course of printing creation.
So I mean, think we probably agree there. I think that with respect to where the structure, emergent structure, of the eventual communication infrastructure with respect to the way in which value is stored and communicated or transacted, seems like we have disagreements, but it’s not surprising to me given the fact that this is a novel evolutionary frontier and it’s always hard to understand what it’ll eventually look like. I really enjoyed the conversation.
Jim: I did too. It was very worthwhile conversation. I’m very glad we had it. I said, it’ll make me stick to new things, and I will remind folks, regular listeners know this, but I’ve put out my own monetary system out into the world back in 2012 actually, designed it. It can be found on a video called Dividend Money on YouTube. Jim Rutt, Dividend Money. You’ll find it at… For better or for worse, there’s a Ruttian monetary system that’s been defined, that people look at it and go, not bad.
Bottom line where it fits in conceptual spaces fairly close to Irving Fisher and Milton Friedman who suggested that yes, indeed, the politically corruptible [inaudible 01:55:26] money system is immoral and a disaster, but so is deflationary money in both their senses. They were thinking of gold as the villain rather than Bitcoin. But Bitcoin and gold are very, very similar in a whole lot of design aspects. And what Irving Fisher and Milton Friedman both independently derived… I think Friedman was influenced by Fisher, is that what you really want is a signaling modality that’s slightly inflationary, like 1 or 2% a year. And I go into great detail on why, and I’ll leave that to the reader who might be interested. So Matt, I want to thank you again for a wonderful conversation, and I’m sure we’ll talk again in the future.
Matthew: Always a pleasure, Jim. Look forward to it.