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 a returning guest. He was most recently on in May 2023 in Currents 094 where we talked about blockchain consensus mechanisms. It was a first-class nerd out. Welcome, Matt.
Matthew: Thanks, Jim. Great to be here.
Jim: Yeah, Matt’s an interesting fellow. He’s a software dude. He’s an entrepreneur. He’s an independent complexity researcher. And my favorite part about him is he’s an all-around smart guy. He runs one of the more interesting Twitter feeds. And I would say three times out of five, it’ll make your head hurt when you read them because they’re so thoughtful and so dense. Make your head hurt in a good way. So today we’re going to have a conversation about a tweet from his tweet stream from 29th of August as always on the episode page at JimRuttShow.com. You can find a link to that tweet and I will read parts of it as we go. But yeah, let’s get ready and hop down into it. So let’s start with a definition. The idea of timed preference and specifically high versus low time preference is fundamental to the tweet. Could you define those terms for the audience?
Matthew: I mean, sure. Time preference in general has to do with discounting uncertainty about the future or discounting one’s relationship to the future in terms of present activities, processes, or transactions. So how much do you value something in light of the fact that it will come to be realized over time as opposed to immediately with 100% certainty so
Jim: The terminology is mildly confusing. High time preference means you have a short term orientation.
Matthew: Exactly, exactly.
Jim: And low time preference means that you have a longer term.
Matthew: Yeah. Low time preference corresponds with long term, high time preference with short term, which causes no shortage of confusion and misinterpretations when trying to talk about these topics for sure.
Jim: And it’s also true. The other model that people may be somewhat more familiar with is discount rates. You know, again, a high discount rate means that you value the present higher relative to the future, essentially. Well, a low discount rate means you value the future, relatively speaking, higher with respect to the present than you would have had a higher discount rate. So yeah, it’s confusing. But let’s make it clear because we’re going to be referring to these terms a lot. So high means short term thinking, low means longer term thinking.
Matthew: Yeah. And we’ll talk about this a lot because it’s quite relevant to exactly the context of the suite, which we might want to read a little bit from. But a lot of this has to do, I think, with the degree to which we are sensitive or are made to be sensitive with respect to the dependencies and uncertainties associated with a future possible event or the realization of a possible event.
Jim: Let’s hold off on that thought for a second, but I am going to start with reading first part of the tweet and then maybe we’ll hop to what you just discussed. A time preference of a culture seeds its orientation towards cooperation. We all know this is rut interjecting. We all know cooperation is actually the human superpower, not the internet or mathematics or the alphabet. It’s cooperation. We’re the most cooperative of the large mammals.
Now back to Matt. High time preferences up regulate the dominance of parasitic relations in which each treats others instrumentally as mere ends to gratifying one’s immediate desires while minimizing the overall need for trust. Lower time preference seeds within us the capacity to see others as complementary aspects of the complex and often fragile processes required to generate artifacts that transcend the capacity of the individual that also inspire the individual to transcend their prior limits. Interesting stuff. Unpack that some for us.
Matthew: This is kind of relating to the interactions between an individual and their context and how that social context changes the parts of the human potential that’s likely to be brought to bear in any given situation. And this is specifically making the claim that to the extent a culture generates a set of norms that are oriented toward, let’s say high time preference first. And remember high time preference means that people have a very high discounting rate. They’re very concerned with what they can get in terms of short term payouts with respect to any interaction. They’re not necessarily concerning themselves.
Therefore, with long term relationships or the long term structures required or processes required to realize long term goals. And so therefore, something that falls out of that, you know, if you have a culture that is obsessed or highly or increasingly focused on the short term, something that falls out of that is an increased tendency toward parasitism. And we might want to also take a slight digression here to talk a little bit about a definition of parasitism.
Jim: Do it.
Matthew: There’s a number out there, a number of definitions out there, but I like two sort of fundamental identifying features of parasitism in my mind. First is asymmetry of energy flows between the agents involved, right? So if you look and you see the actual asymmetry of energy, I say energy meaning like it could be money or it could be attention or it could be anytime you see a strong asymmetry where one agent, has a lot more of that coming from the other agent than they are reciprocating. That’s one of the prerequisites. And then I would say the other prerequisite is something structural about that relationship that actually makes it difficult for the agent that is on the losing end of that asymmetry to change the relationship.
And so if you combine those two prerequisites, that gets you pretty far towards understanding the fundamentals of parasitism. And so, you know, taking this back to time preference and culture, if we focus increasingly on only short term relationships on only short term transactions, first of all, it becomes very difficult to notice when parasitism is occurring and therefore it makes it very easy for parasitic behavior to basically escape notice or punishment. And secondarily, we’re no longer incentivized to track that reciprocal cooperation and play games that allow for us to actually choose our partners for playing games that are in line with cooperative potential and higher payouts over longer periods of time.
And, you know, you kind of end up in a defection loop where everybody’s playing short term games and trying to get what they can from one another over very short time horizons in highly or increasingly parasitic relationships at all scales of the system.
Jim: Let’s contrast parasitism, which is an ecological and biological term, actually, with the other two of the triad, which is mutualism, which is a symbiotic relationship in which both parties benefit. You know, the classic example are those birds that pecked the bugs off the back of rhinos, right, or the fish that clean the debris out from the teeth of sharks. The fish benefit by getting something to eat that somebody else captured for them, and the shark or the rhino benefit by having parasites removed. So that’s true mutualism. And the other is commensalism, where a symbiotic relationship exists in which one species benefits while the other is not too much affected.
It’s not obvious to me what those are, but there must be some. And then there’s symbiosis, which is a close relationship between two species, where at least one benefits. And lichen is a good example, which is a very close relationship between fungus and algae. And the two can’t exist as lichen without each other. So it’s actually a symbiotic system that doesn’t work unless both are there. So parasitism is part of a family of forms of cooperation. It’s just the sucker form of cooperation. If you’re the parasite E, does that make sense to you?
Matthew: Yeah, exactly. And I think, you know, a lot of this falls out of playing with those two knobs that I mentioned a little bit. Like if you think about so commensalism is interesting because I would argue you don’t really have direct energy flows between those organisms. So for example, a lot of those types of situations happens when by a happenstance of nature, there happens to be some sort of ecological niche that’s kind of not really related to the metabolism of the underlying organisms. So like maybe some sort of, like for example, on our skin, you know, we have many living microorganisms and the degree to which those are actually symbiotic or parasitic, there’s a lot less valence differentiation there than with respect to our internal microbiome. Like it’s much more, you know, if you actually go inside the body, it becomes much more consequential in terms of those relationships and the byproducts of, you know, whether a relationship is symbiotic or parasitic and how our immune system will respond to that, for example.
Jim: That’s a good example.
Matthew: On the skin, there’s a lot more leeway for things to make a habitat of that without us being effective or caring much, right? So like commensalism, I would argue kind of violates that first prerequisite in the sense that you don’t have this kind of direct flow of metabolic dependencies between organisms or between the two processes. And then symbiosis, right? You can have short term fluctuations in terms of who might be coming out on top, but as you iterate that game and as you have enough transactions in that game or enough reciprocal interactions in that game, you know, both organisms are integrating the byproducts or the consequences of that reaction into an integral part of their own metabolism. And they’re both actually fundamentally dependent on that.
Jim: Yeah, the famous example of Lycan and algae is one, but another one that’s got a lot of talk recently is fungus and the roots of hardwood trees. They essentially build an ecosystem, neither of which could survive without the other at this point, or at least couldn’t thrive.
Matthew: Now you’re tempting me to like even venture back to some of our previous conversations. I’ll avoid it, but like the metaphors associated with the different kinds of monetary systems we’re experimenting with now and their analogous properties to the trees or the fungi in those.
Jim: Yeah, let’s save that one for another day. We should do one just on monetary system someday.
Matthew: For sure. For sure.
Jim: Now, I’ve thought about this a little bit, and I think one of the subtexts here and probably of your post, but I’ll confirm it with you, is a sense that high discount rates and high time preferences seem to be on the increase in the current epoch. Would you agree with that?
Matthew: I would very much agree with it. And I think a lot of that has also to do with, you know, that second part of the platform or prerequisites I was talking about with the structural features in our environment. So that can lock in certain kinds of asymmetric relations. And I think a lot of the technologies that we’ve brought into our lives over the past two generations, while they’ve had many benefits, they’ve also done a lot to create a structural landscape that favors increasingly short term interactions, short term extraction, one shot interactions, and that don’t really ladder up into much cooperative behavior or cooperative stable relationships over time.
Jim: And that’s what I figured you’d say. And I have the same sense, right? It seems like a whole lot more short term of everything that’s going on. And yet when you stop back and think about a little bit, I said, hmm, that’s kind of odd when I think about it in terms of the longer arc of history. For instance, you know, at least those who are four year college grads in the United States are living longer, right? And that should give you a lower time preference or a longer time horizon. Perhaps paradoxically, I would argue that income inequality actually ought to push planning depth or lower time preference, because there’s actually something to win by being disciplined over a period of time, right? You can go up the curve. You have your shit together for 30 years, you’ll rise further up the curve most of the time. And the payoff for that is fairly significant, as opposed to just being a person with very short term planning horizon doesn’t think across a multi year.
Also, I would argue the world is changing less in some qualitative sense. I use as an analogy my father and myself. My father was born in 1923. So he would have been 100 today. He died in 2003. And when he was born, less than 10% of American households had telephones, right? There were only 60,000 households out of about 30 million that had radio. Prohibition was right in the middle of its season. Women had only gotten the vote three years before. I think about cars being central to the American dream, at least in the 20th century. 1923, it turns out, was the peak year for the manufacture of the Model T. Model T is a pretty damn primitive car, right? Two million of them were made in 23. He lived through as a teenager, the Great Depression, and then as a young man was a Marine in the South Pacific in World War II. And by the time he passed away in 2003, he’d seen most all of the modern world, right? The internet existed, cable TV, satellites, anything you can possibly imagine.
And the world was just radically, qualitatively different. Let’s talk about airplanes. 23 by planes were still the main thing. Monoplanes, planes with just one wing, were a curiosity and thought to be a sure way to die in 23. And certainly by 2003, we had 747s flying all over the world, big jet engines on them, gone to the moon, all that’s left. So in one lifetime, 80 year lifetime, an unbelievable amount of stuff happened. My own lifetime, I was born in 1953, and so if I make it 10 more years, I’ll get to 80. In that time, I would argue much less has actually happened. While the cars of 1953 look a little funky, they’re still very identifiable as the same thing as a modern car. They’re streamlined, they got windows, you can roll up and roll down, they have electric starters, some of them have automatic transmission. So they’re clearly much more like the cars of today than a Model T was, like a 1953 car. Airplanes, at that point, the first of the jet airliners was just coming on stream. Certainly jets had already been used in World War II, rockets had been used in World War II.
Most of the household appliances we know already existed, if in somewhat rudimentary form, etc. So if I shake off the mortal coil in 2033, I will probably have led a life with quite a bit less change in it than my father. And then finally, the peak of interstate household move in the United States was 1975. It’s about half that rate today. When I graduated from college, which was 1975, almost everybody I knew moved to some other location from where they grew up. It was just the thing you did if you were a college grad. Not everybody did, but a large propensity of people to do that. Not only is it my sense, but the statistics bear out that it’s much less common today than it was. So with all these attributes that one would think would produce a more stable, more predictable ecosystem, what’s your thought on why does it somehow feel something we both feel that time preferences are much shorter these days against these huge macro trends?
Matthew: The fact that it seems to some extent as if there was more realized change previously than today, the lens that I tend to look at this through is the lens in which I look at a sort of realization function and also a potential function, right? Where the realization function is apparent when you actually see a new innovation and it becomes very clear that people are adopting it. It becomes very clear that something has manifested in the world when you’re talking about a new kind of airplane or a new kind of invention of the computer to begin with in terms of the computer entering the business domain and then the computer entering the home. Or you’re talking about a change in paradigmatic warfare or there are these fundamental realizations of change in the world that we notice. But then there are also changes to the underlying potential of what could happen.
And I think those are based on concepts like the adjacent possible, which adjacent possibles have we opened at a given time that have not yet been catalyzed or brought into realization by some sort of a calling forth, some sort of an event or set of events that really drives exploration of that niche of potential. And I think, for example, if you look at people like Steven Pinker and some of the arguments that they will make, they will make the argument that the general trend, for example, of people dying in wars is going down on average over time. But then if you look at the actual, not just the average, but the actual unfolding of those sequences of deaths and the scale or magnitude of them, the magnitude when we actually do have large scale wars of the amount of deaths in those wars was increasing, especially in the world wars. And the question is, is there another increasing event in that sequence on the horizon?
Are we increasing the potential of something like that, which relates in complexity terms that relates to the question of like criticality, right? You’re stacking the grains of sand on the pile. At which point do you get an avalanche? And is that avalanche actually a much larger avalanche? We’ve unlocked the genome and are just beginning to understand the side effects of that. We don’t even understand the side effects of globally deploying an mRNA vaccine to the entire world, essentially. In some might consider unwisely rapid form, we don’t understand the effect.
I mean, the birth control pill was its own, almost perhaps one could argue species changing event. And now we’re about to potentially release IVG. We have drones that are enabling an entirely new form of control observation and potentially warfare. We have this entire forefront of AI that is changing our relationship to the real and the virtual is going to very much blur the lines between, you know, our evolved senses and their typical assumption that they are observing the outside world.
And the fact that they are now going to be directed toward observing all sorts of things that are not at all the outside or real world, but are instead the fabrications of other economic entities, other governmental entities, other individual entities. So there’s all these potential functions that also exist right now that I think we’re just starting to see the tip of the iceberg with respect to events that are bringing them into realization. I think a lot of those, it also takes a whole lifetime for us to look back and see, just kind of compare all the realizations of the potential that existed throughout that lifetime. And so we still have some time, especially in the millennial generation, before we can analyze just how much of that potential has been realized and how that compares to past generations.
This is kind of what I would argue with respect to the structural changes on that front that increase time preference that make us much more sensitive to short term time horizons. So much of what has flowed from connecting the 7 billion people on Earth at light speed to one another is actually a dramatic increase in the amount of uncertainty as to what people’s world models are like, what people can actually count on to be stable. How people are going to generate random viral phenomenon that have the kinds of knock on effects that are almost entirely unpredictable in the traditional modeling contexts that we are used to.
And therefore, many of the economic tools, many of the cultural responses, many of people’s personal responses in their own lives is to essentially stop thinking about the future because it’s become just infinitely more unpredictable because of the fact that change, even though change doesn’t seem to be happening yet at the most foundational structural levels, change from the perspective of how people spend their time and actually their daily behavior, I would argue has changed an extreme amount from even 20 years ago from the pre internet culture. The way that we interact with other human beings, the fact that we almost all of our interactions now in terms of time are mediated by screens and mediated through computational systems. The fact that we now tend to perceive, you could even argue that people perceive what’s happening online as more real than what’s happening in their immediate world. The fact that we have more of our social connections at national or global distances via those networks than we do in our local communities. Like there are all these things that are changing that aren’t quite readily visible in the way that previous realizations of change in technology are, but that are happening and I would argue tilt toward higher time preference.
Jim: I think you’re onto something here with, let’s say the fact that maybe many people, only some people understand that we’re near some portals that could take us far away. You mentioned genetics, a specific example is CRISPR, right? What the hell is CRISPR going to cause when you can snip and change? There’s been some little effects so far, but we haven’t been producing centaurs, horses with the heads of humans or something, maybe we could. You know, nanotech, you know, it’s been out there since the late 70s and has delivered some small results, but at some point this could go through some transition point. And of course, we’re on the verge of another portal it feels like to me with the AI revolution.
You know, large language models are going to be somewhat transformational, but nowhere near as transformational as strong AGI would be, which is coming somewhere between tomorrow afternoon and 2080 probably. The liminality might be a factor, but I’m going to then turn that around and say for people like you and me, sure, we’re aware of these things and the people we hang out with are aware of these things. But as the random 16 year old girl who goes to a Taylor Swift concert, aware of CRISPR or nanotech or the AI doom loop scenario, etc. So maybe these things while they could impact people who are cognizant of techno cultural trends, I might argue maybe they don’t have that much effect on the masses.
Matthew: I mean, I think those large portals or those large, let’s say potential singularities and our ability to predict the future. I think those still very much lie in the category of potential functions as opposed to realizations, right, which is kind of what I was trying to get at in terms of the fact that we have very much yet to see what it means when those begin to really affect the public in ways that even that person on TikTok totally enamored with their own immediate sphere of influence would actually notice, right? Like, I think that will happen in our lives. And we don’t know how, but I do think it will occur.
It also forces us to consider consider though, like there have been realizations that we need to talk about with respect to that same girl, right? The fact that her life, her social life, the way that she is receiving signals about what’s valuable, the way that she is receiving signals about what are life goals or what are behavioral models that she should emulate, like all of this has actually changed quite dramatically with respect to the sample set she’s drawing from due to the infrastructure that she’s spending most of her time looking at, for example, right? This applies to all of us to some extent, but it applies to, I would say, someone like that even more because if you look and this is a perfect discussion of this evolution of structural evolution of these particular species of communication social media platforms in particular, toward higher and higher time preferences. If you look at the evolution, like think about the internet first came along, people had web forums, people would write posts that were pages and pages long, right? To each other.
Jim: I was certainly guilty of that. My friends on the well could still make fun of me for my verbosity.
Matthew: You were anchored temporally, culturally and educationally in a culture of books and essays, a culture of much deeper articulation. One need only go and look at books written in the early 20th century on almost any topic to get a sense of the fact that it was an entirely different capacity for writing, articulating, processing and digesting information when people would sit down normatively with a book for hours on end and only use the connection between that text and their mind could be uninterrupted.
That attentional loop could persist uninterrupted for hours, right? That almost never happens these days. I talk with people all the time and it’s not uncommon for me to hear people discuss the fact that they feel they can’t read anymore. They can’t read a book. They can’t sit down with the book and actually parse it anymore because of the fact that their brain has been so conditioned by this need for increasing stimulation, right? And so you look at the evolution of this technological infrastructure and you go from those forums to things like Facebook and then Facebook, you still saw much longer posts and then you get the advent of Twitter and Twitter has had a little bit of a shift there with the release of the limitation of 120, I guess, to 240 and now with paid accounts to some thousands of words. So there’s a little bit of an instability there in that pattern. But then you loop in things like TikTok, right? Where you see and video formats and you start YouTube with longer format.
Jim: Texting is the big one.
Matthew: Texting was a huge paradigm shift as well, for sure.
Jim: That’s very short.
Matthew: All of these are pointing in one direction. Generally, if you had to trace the trends to a convergence, right? Like it’s smaller amounts of information, transacted faster and it’s more like a gossip protocol, right? And if you want to look at people who’ve researched things like this, I’ve been trying to ring this bell as much as I possibly can. But I think if not the one of the most important concepts that Stuart Kaufman ever brought into the world is not the adjacent possible. It’s the complexity catastrophe. It’s when he was studying the side effects in genetic networks of epistatic interaction, meaning the consequences of expression of one gene upon the network of genes in which it was involved and the fact that all the genes were doing that simultaneously.
Meaning like if we each are broadcasting into these networks and our broadcasts are changing everyone else’s behavior in real time while they’re doing the same thing, what are the consequences of that? And his findings with respect to these genetic networks is that the actual ability of those networks to perform the task of adaptation based on information for those networks to actually hold the information required for them to adapt toward improved or higher locations on that fitness landscape was destroyed beyond a certain threshold of connectivity, right? So you turn that connectivity up to high and you entirely destroy the ability of that network to evolve.
Jim: I have a theory very closely related which is super reductive, which is kind of surprising being Mr. Complexity, but I’ve run this by a number of people and it actually sort of fits everything we’ve just talked about. I suppose instead of it being short or being long or being pictures or being video, I suppose it was something else. Suppose there’s a deeper independent variable, which is that humans have some maximum level in which they can be interrupted per day in their signal processing.
And it’s the raw count of inbound messages that we choose to process each day. And it probably the number varies the same way the ability to eat spare ribs vary. I can eat a lot. My wife can eat a little, right? Some people can probably receive 500 messages a day and actually look at them at least for a second and be okay while other people were reduced to quivering, pool, a protoplasm, three molecules deep if they do that. And so having more and more just at least contemplate, am I saying it’s true?
Matthew: Yeah, that hypothesis, I think that is the mathematics, there’s an idea of like a dual, right, which is sort of like if you invert the connections on a geometric form, basically, in so many words, I think what you’re describing is kind of the dual of this complexity catastrophe idea, because the complexity catastrophe idea is much more about like the signaling, like the outbound tolerance threshold.
Jim: But also the important thing about the complexity catastrophe also has a lot to do with the network topology. You can have a lot of signals flowing if they’re routed with care, right? If they’re routed opto.
Matthew: Sure. If you don’t have a ton of scale free connectivity.
Jim: Yes, exactly, right.
Matthew: And you actually constrain the communication between scales to a narrow band of channels 100%. I had a tweet a long time ago that was kind of tongue in cheek, but it’s essentially like, to the effect of we’ve connected all of our people cells overnight, and now we’re having a seizure. And that’s precisely what would happen in the body if you took all of the highly evolved, you know, it’s taken a lot of evolution, a lot of experimentation, a lot of death to figure out the appropriate ways to connect all of our neurological signals. Many ways for those to go wrong. If there weren’t, we wouldn’t have a multi-billion dollar psychology and psychiatry industry
Jim: Or an alcohol business.
Matthew: Precisely. So if all of your brain cells were firing simultaneously and connected to one another simultaneously, all you would have was death, like you would immediately die, right?
Jim: It sees up. I like this. This will take the more nuanced, cough, many in perspective, which included both rate and topology and the rut gross messages received, then generalize it to say that there is a pattern of network statistics that is healthy and there’s a pattern of network statistics, which is unhealthy. And we probably have no fucking idea where we are in that domain.
Matthew: Wouldn’t actually be that hard for our social media or our governments to even look at this, right? To even think about this, except it’s just not in the wheelhouse of the way that they think about these systems and it’s not incentivized. Because for example, if any of these people running these systems ask themselves, oh, how do I measure how my network changes the adaptive capacity of the members of the network? They would very quickly find that answers to this question have been proposed and there’s a rich literature around it, right? But so long as the question is something along the lines of how do we maximize attentional or economic extraction from what we’ve essentially created as a domesticated herd of attentional livestock, right?
If that’s the question, you’re never going to ask why or what we should consider when we’re thinking about how we increase the adaptive capacity of these people in the network or the network as a whole.
Jim: Exactly. That’s the bottom line right there. When you denominate everything in relatively short-term money on money return, there’s no guarantee at all you’ll converge to a healthy statistic and I would suggest that the least anecdotal evidence and statistical evidence is that we have way shot whatever function, let’s put it on a monotonical scale for whatever reason and say there’s some optimal point. It certainly feels like we’re way past it and yet if all they’re responding to is the money on money return signal, what’s my stock price, which is essentially a reification of the money on money return curve, then there’s no reason to expect that we’d be in an optimal range on our network statistics.
Matthew: 100% and this might be a little far afield and it’s a little bit of a logical jump and I could get there step by step, but I’m just going to go straight there, which is I think when you exist for too long outside that distribution of well adapted network statistics or even adaptively capable network statistics, you will begin to see disintegration at increasingly large scales and the response to that is going to be by actors who have disproportionate power over these structures, the response in that case is going to be we need to turn all of this into a large deterministic system. You’re going to get the kind of things that the CCP for example is implementing or has implemented already with respect to network controls, with respect to the extension of governance as a cybernetic system through the entire society, with respect to economic incentives and behavioral curation, let’s say.
Jim: Don’t just point at the CCP, you know, Facebook and Twitter, at least before Musk and the other platforms here are also attempting very hard.
Matthew: Obviously, there’s definitely political collusion there and corruption there. I think a lot of that alignment is emergent and simultaneously there are many politicians who salivate in the West even and perhaps especially right now who salivate at the possible opportunity or at the possibility of getting their hands on the kind of social controls that are being tested and rolled out in China today. That is 100% true just because leaders of Western nations does not mean that they are not highly desirous of that same capacity for power and control because we have to remember that every single one of them, especially in so-called democratic nations, is theoretically vulnerable and what they do in terms of their local incentives has a lot to do with minimizing their vulnerability and if they could permanently remove their vulnerability to being removed for power, perhaps they would do that. So we can roll all this all the way back as well to culture, which is kind of what we were talking about.
Jim: All right. I’ll just add that frankly, a lot of the emerging EU internet policy is kind of CCP soft, right?
Matthew: Yeah, 100%.
Jim: You can actually be thrown in jail in some European countries for claiming that there are two sexes, for instance, right?
Matthew: A quick aside here, like I have a series of essays from years back called Crypto Beyond Capitalism, The Rise of Digital Valorism and like one of the things that I argue there, I think maybe in the fourth essay or so, I talk about emergence and the role of bottom-up versus top-down organization and coherence, right? And like I basically make the argument that to the extent that we as a species or as a community or as a nation, any fractal scale you want to choose in this emergent structure, to the extent that we are incapable of emergent cooperation, you will see the co-ordinated gap or lack made up for with coercion, right? So if we cannot cooperate at the scale of interaction that we have attained, it will manifest as coercion filling that gap, right?
Jim: Yeah, interesting point. I just published last night, my podcast with Neil Howe on his book, The Fourth Turning is Here. He basically predicts authoritarian coercion at the crisis point, says it always happens, right? Because it has to.
Matthew: And the reason why we think, I don’t think it has to, but I think it’s very likely to because of the fact that it’s almost never the case that you have the appropriate emergent prerequisites for capable or competent coordination at that new scale, ready to go when you’re entering that new domain. But I also think that our margin for error in terms of time, how long we have in this particular moment to actually bring that cooperative infrastructure up to speed has never been shorter. Because if you get to the point where the current generation of coercive technologies gets to a lock-in point, it’s never been more difficult to undo, to unring that bell, right so
Jim: Very good point. So let’s get back to the tweet. So the next line of the tweet is, we can then begin to see that the value of trustless infrastructure does not lie with the overall minimization of trust. Say more about that.
Matthew: Yeah. So that comes from a lot of the discourse that has occurred in the crypto world around trustless infrastructure, around the need to get rid of trust in our systems. And this was another point that I actually made in that same essay series, which is that there’s no way to get rid of trust. Trust will always move to the edge of a network. Like you can create something that is fundamentally predictable in its mechanics between two parties. But at the edge of that transaction, you can always have corruption. Like you can always have, even if you have the most secure transactions available, if you have a hostage on the other side with a gun to their head, there’s influence in that system, right?
So the best I think we can do, and that I would argue that what we actually are doing by creating trustless infrastructure, and I should be clear that what I’m saying is not that we shouldn’t try to create this thing we call trustless infrastructure. It’s just that what we’re actually creating is actually something much more akin to, and some people will be familiar with this term, some people will not. So I’ll explain it, but something called cognitive chunking, right? Which is actually, if you think about like, look at the way people learn to play chess, different levels of chess players. When you first learn, you actually have to pay a lot of attention to the individual moves, how the individual pieces are allowed to move on the board, the local context.
But then as you get better and better, by the time you’re a master, you’re seeing large abstract gestalt patterns, you can see into the future in a way that doesn’t even necessarily involve directly looking at any given piece, right? Because of the fact that all those previous learnings have been encoded and have become automated and have become therefore the sort of process foundation or the foundation on which these new emergent processes from which they can arise and upon which they can rest. And I would argue that the kind of infrastructure that we’re building that we call trustless infrastructure is doing something similar where we’re basically trying to take a lot of the previous types of interactions that would have required trust. And we’re trying to put them into a domain where we can no longer, we don’t have to worry about whether they’re trustworthy or not.
And we can move to a higher level of the game of cooperation, where we can coordinate at a higher level, while depending on that previous infrastructure for stability and to not require trust. Because fundamentally trust, you know, this definition of trust that I like to give typically is the ability to, for people to come together with their own internal map of past, present and future, and use those maps, combine those maps with one another to plan for present and future actions, right? And future actions, the limitation on which you can plan with someone else toward the future has a lot to do with this concept of trust. Because fundamentally, it’s all about the extent to which those two parties after having made a plan, first of all, we have to understand that we have to first avoid those parties lying to each other, creating false maps and false worlds to take advantage of one another’s behavior. So if you were and I were to talk about planning a trip together, and you were to tell me that I was supposed to meet up with you in some location and that we would go have a vacation from there and what you really did was plan for someone to go there and rob me, right? That would be a violation of trust.
I would therefore obviously think you’re if I survive, I would think you’re an untrustworthy fellow, and that would destroy our possibility for future cooperation, perhaps might even put us into a feuding state of defection. So first, we have to avoid that. But second, the only way we can plan in the future and bring these maps together is if these maps, a, are of fairly high quality, but also we can count on one another to not defect or give into other temptations along the way, other distractions along the way, if we made a joint plan to realize some future event or future outcome together. Like if we plan to meet in that location, let’s say you did plan to meet up with me for that vacation, but along the way, you decided to stop by a casino and boo all your money, right? And then you just didn’t show up and I was sitting there, sitting there alone and we couldn’t realize that future that we had planned together.
I would also discount your trust. So what I’m getting at here is that we fundamentally require infrastructure that allows us to bring out the best angels of our nature to plan with one another in ways that are resilient, that allow us to overcome our tendencies toward distraction and our tendencies towards temptation. And actually, that’s a prerequisite of doing anything over longer time horizons with lower time preferences and that infrastructure that allows us to move certain things that required some of that trust energy into an automated space, freeze up energy in that space of scarce energy for us to invest more in low time preference cooperation.
Jim: Could you give a tangible solid example of where a piece of trustless infrastructure would free up, say a lot of frictional costs in which that we’re using today to police soft trust or defective or corrupt systems that we could use then that energy to actually do positive cooperation? Give me an example of something that if it existed with the-
Matthew: Sure. I mean, like typically a go to of mine is a Bitcoin example involving the Fed and a lot of the entailments of that. But we could also talk about, for example, like zero knowledge proofs. Those are a fun one to talk about with respect to voting.
Jim: Yeah, let’s talk about that one. They are a good one.
Matthew: Voting infrastructure, right? Like right now, like how much energy are we spending as society and how much chaos is flowing from the fact that we have a voting system that requires us to invest heavily in policing and enforcement. And there are many middlemen and there are many places and points of failure in which that system can give rise to situations where there’s ambiguity. And that ambiguity can lead to chaos and it will lead to interested actors who wish to leverage that ambiguity to increase their own status, power, leverage, et cetera, which is exactly what we’re in the middle of this, especially as Americans right now with the various election issues that we’ve had over the past couple generations really. It’s obviously been increasing and increasing in terms of the skepticism of the public toward elections.
And as you increase the skepticism of the public toward elections, you increase the efficacy of telling stories about bad elections for political actors. And we see actors on both sides doing this. Like every election, there will be people who are calling into question the veracity of the win of their opponents. And so much attention and so much of our social energy goes into this, especially now that our energy as individuals is so readily pulled up and our attention as individuals is so readily focused on what’s happening at these national scales, which is also something new.
Jim: So why’d you explain to people how ZKP zero knowledge proofs might make that a less fucked up and hazy process?
Matthew: That was just the setup of the problem, right? That was just kind of painting the fucked up landscape that we exist in today. And there’s reasons why we exist in it. I’m not saying like we could have necessarily even done better, but we are in a place where it’s falling apart in many ways. And so what do you do about it? Well, you know, zero knowledge proofs without getting too deeply into the details give us a tool by which we can prove that something has happened without revealing the details about the parties that have done that thing. You can maintain anonymity and you can maintain security while also having verified outcomes about what’s going on. And so, people are using this to do many different, it’s a very general tool, but people are using this to try to attack a number of problems, one of which is voting. And so, voting in that way, you can very readily, again, none of this is perfect, right? Because you could still imagine a world in which people go around with guns to people’s heads and force them to use their identifying keys to actually vote in a way that they wouldn’t have otherwise voted.
So if you use enough force and enough coercion and enough violence, you can break most systems. But the question is whether or not that is game theoretically feasible, whether or not someone will actually be able to do that. In any case, these zero knowledge proofs would enable us to the extent that people actually felt confident and had decent demonstrations that these work at lower levels, which is why I don’t think you can go to the national scale with something like this immediately, because people are going to be like, oh, it’s just a computer, it’s just an algorithm, why should I trust that? Or we’re at such a low trust point already that without seeing this kind of stuff work at more local or tangible scales, can’t just roll it out at a national scale and expect people to actually believe it. But it would, in fact, give you computationally verifiable security over signaling mechanisms across the society. And you could imagine different kinds of voting protocols, you could implement liquid democracy protocols, there’s a bunch of things you could do with it. But fundamentally, the nice part about it is that you can maintain privacy. So it satisfies a lot of the concerns that people typically, let’s say, on the left have with respect to the vulnerabilities of some people signaling their political preferences and how they might be coerced if they’re seen to be supporting certain causes.
But that concern is universal, it applies to both left and right. It helps to satisfy that. And it also helps to satisfy the need for actual security. And it does so in a way that also cuts out many of those previously mentioned layers of human counting, of corruptibility, of people dropping off voting records in the middle of the night, or machine politics, running certain voting locations and getting up to all sorts of shenanigans that can theory happen and do happen. These happen in every election. It’s nothing new. There’s some level of corruption always. It’s never perfectly incorrupt because it’s not incorruptible. But these kind of new technologies that we have allow us to have greater levels of incorruptibility and therefore refocus attention, hopefully, on more essential questions, or even perhaps free us from having to worry quite as much about a lot of these national issues. And also the stories that are told at that level that capture us, that capture our attention, that capture our emotions, and therefore distract us from actual cooperation and building things of value.
Jim: Yeah, think about it. We’re arguing about, as you point out, every election has corrupt a bit. Just the placement of where the voting districts are, right? It’s a huge level for corruption. What kind of voting machines do you have? Do you have the paper ones? I actually testified once that the right answer is the scannable paper ones, so that there’s always a paper trail. But if the ones that still have the purely electronic or the electromechanical, all kinds of opportunities for badness there. But anyway
Matthew: But the nice thing about electronic issues are issues. But like, if you have mathematical proofs, if you lift that up into the realm of like platonic mathematics, right?
Jim: I understand that. You can go one step further, but you’d have to prove it that you had it. It looks like it doesn’t exist.
Matthew: Well, and then therefore, people also, this doesn’t get us around fundamentally the need for trust again. Because remember, the need for trust therefore just goes to the question of whether or not the public will accept the system. And you need to build the trust required to actually roll a system like that out, meaning that you’d have to demonstrate its incorruptibility or its dramatic improvement at lower levels of the system. And that gets us back to the problem of the fact that the people who tend to have power over their decisions like a bit of that corruptibility baked into the system.
Jim: Oh, absolutely. All the states, gerrymanders, are like motherfuckers, right? They decide how many voting machines have at what precinct to disadvantage their opponents, right? They don’t want to lose that power. And here’s a very important point. I came to me in a flash recently. I’m working with my co-author on the Game B book to explain it all. It’ll be out one of these days, right? And one of the patterns I just extracted, I said, every system needs an immune system because the blight will always be there. And I have the expression, fight the blight. The blight being the perfectly reasonable game theoretical tendency for people to cheat if they can get away with it, right?
Everything from nepotism to financial embezzlement to coercion of sexual favors from employees. Every system that has humans in it, the blight is looking for a way to get to work. And if you don’t design to fight the blight, you’re going to end up infected and collapsing or at least be grossly suboptimized. So don’t ever think that things will ever be uniformly good because the blight will ratchet up. But it is our job to make the system more and more blight resistant. And of course, also to take the blighters out back and put a nine millimeter round behind their ears. I think there’s an awful lot too less of that going on.
Matthew: Yeah. I mean, out back or even perhaps more in public, I think understanding consequences through that means a whole different conversation.
Jim: Object lessons.
Matthew: I would argue that the deterrent theory of justice has something to do with social immune system. But yeah, fundamentally, I think that that’s actually what you’re describing there is one of the criteria and I always use when I’m trying to understand whether what someone is saying falls into the category of naive utopians.
Jim: Exactly. Excatly
Matthew: Or are they proposing a system where they’re telling me that they can avoid all the costs of an immune system, where they can essentially create a world where you don’t need to invest in an immune system? I think that’s a pretty good litmus test for whether someone’s bullshitting you or offering you some sort of story about an actually unrealizable utopia.
Jim: I find they usually bullshitting themselves first because most of them actually believe they’re bullshit. So, you know, they’re smart, but
Matthew: Depends whether you’re talking to the useless idiot or the cult leader.
Jim: Exactly. Exactly. That’s a good distinction. All right, let’s move on to the next step. Actually, you sort of covered this, but I’m going to give you a chance to bring it into tighter focus. The next sentence is rather it lies with the introduction of foundational infrastructure that is in and of itself resistant to high time preference parasitization at its most centralized nodes while enabling higher levels of trust to reemerge within the peripheral relations of the network’s terminal units. There’s a lot in that sentence.
Matthew: There is. There is.
Jim: I want you to unpack that one for us a little bit.
Matthew: So, I mean, I guess I will bring up the Bitcoin example here just because I do think it’s edifying in this example. I’ll make the argument you may not agree, but at least structurally, I think it shows the point here, which is that let’s say we have something called the Federal Reserve, right? And if you look at the Federal Reserve’s current behavior, you look at the kind of loop of incentives that it gets itself stuck within, we have this news cycle, we have this tension. They’re tasked with essentially maintaining a particular interest rate and trying to responsibly manage the levels of currency in circulation to help mitigate inflation. Also, you have a lot of local incentives that are operating on these people in terms of their own reputations, in terms of what they’re going to do after they leave the Federal Reserve, in terms of the network that they are enmeshed with with respect to banks and the fact that these banks have interests as well. So, you have all of these lower time horizon, therefore, higher time preference actors who have pull on the Federal Reserve when the Federal Reserve, even given that it’s a highly flawed mechanism for trying to operate a complex system or parameterize or steward the behavior of a complex system, you can’t operate such a complex system with just a few levers.
But fundamentally, even if you could, the fact that you have this highly centralized node that is interwoven with actors who have incentives that are on different time horizons than the national behavior, if you were to think of the nation having a heartbeat, the heartbeat of the nation is a longer heartbeat than the heartbeat of the banks that are pulling on the Federal Reserve to do whatever they are pulling on it to do at a given time. And so, you have this kind of a gravity toward a race to the bottom where you have actors with the power to make decisions that are increasing the time preference of its decisions out of step with the actual system they are tasked with stewarding. And so, basically, one way of looking at what I was saying is saying a system like Bitcoin basically says, we’re taking this out of human hands. This is a mechanism. The mechanism has a behavior. No humans can affect that fundamental behavior. Humans have to comport themselves around that behavior. And therefore, no matter what, you can depend on the heartbeat of that behavior being stable.
And then, obviously, it’s a complex system, so many complex behaviors can occur around that heartbeat, but at least you can count on that heartbeat being stable. And then the second part of that claim when I’m talking about the fact that that enables higher levels of trust to reemerge within the peripherals of the network’s terminal units, what I’m saying there is essentially that by having a stable heartbeat that people further out in the network. So, if you go beyond the bankers and you go beyond the central banks and the local banks and out to the actual people trying to create businesses or take out loans, if they have something stable to look at and count on, if they can know the behavior of that heartbeat, they themselves can start to coordinate with one another without being fundamentally dependent upon all the effects that are created by people closer to that heartbeat manipulating it. Right? So, they can actually create locally stable and locally valuable economic relationships without any dependencies and all the corruptible relations and incentives higher up that stack or closer to that central point of management. That’s what I’m getting at there.
Jim: I think maybe we have a wonderfully simple and vivid example, which is if you have a Fed, which can turn the spickets of money supply open whenever they feel like it, it becomes possible for politicians to spend money now for people to pay off in the future. Right? A giant deficit is a classic example of a high time preference. Right? You’re basically saying, oh, that’s spent $2 trillion more this year than we take it in taxes and we’ll push that out to our children and grandchildren.
Matthew: That’s a great example.
Jim: That is the example, right? And it ties to your case because, truthfully, that would not be practical on the gold standard. You know, look at the size of the deficits during the gold standard era. Other than in world wars, they were, you know, a few percent of what they are today because, you know, post 1971, when funny money could just be grown at will. And then, of course, the politicians didn’t really have the nerve to do it until Ronald Reagan in the late 80s, where they opened the spickets up like crazy and then little Bush went even further and Obama further and Trump even further and Obama’s has thrown any possible restraint out the window. But none of that’s possible without the ability to massively inflate the money supply, essentially.
Matthew: We can use this as a way to come back or like circle back into conversations around parasitic behavior as well because of the fact that when we look at fiat currency and we look at the ability to print money in the way that, like you just said, that encourages the inflation of debt, we can ask ourselves, what is a fundamental pattern that is enabling this? And, you know, whenever you create an asset, you have a commensurate liability that you’re producing as well, right? And if you have an asymmetry between that asset and liability time preference, you are producing the perfect conditions for parasitism because the asset can be used now. There’s no uncertainty or very little uncertainty about what you can do with it in the current economy. But the liability, whether or not it will ever actually come back into symmetry or come back into relation with that asset is put into the future of uncertainty, right? So you basically throw the responsibility for actually backing that asset you’ve created into a totally uncertain space of the future where it’s always someone else’s problem. And if you can do that, you’ve created the perfect ecosystem for a parasite.
Remember, we already, that was the structural prerequisite that we just discussed. And then you, all you have to do is insert someone willing to basically take asymmetric advantage of that energy flow and insert themselves right there at that point where they can take all of the advantage and give very little back. And then that basically takes the form of skimming that asset from the system that’s supposed to be allocating that asset into an investment infrastructure, right? The capital that’s, or the assets produced by the Federal Reserve are supposed to be flowed out through these banks into the economy, into economic actors who are going to make responsible use of that capital in theory. But if you actually allow for a siphoning of that up the chain into the hands of personal individuals before it ever even moves itself through that economic cycle, then that is, you know, definitional parasitism, right?
Jim: Yeah, huge parasitism. Most Americans don’t realize they’ve been parasitized. The major corporations and banks who issued bodacious amounts of long-term debt when interest rates were… artificially low, right?
Matthew: Yeah, I mean, how many people outside the Bitcoin or crypto community understand or are aware of the idea of the Cantillon effect, right?
Jim: I’m not familiar with that.
Matthew: So the Cantillon effect, it literally is this effect of essentially the sort of gradient of monetary value and the gradient, the highest point of that gradient, the highest gradient of that function being at the Fed, right? So like the highest dollar value of a newly printed dollar is at the point of minting. And then the closer you are to that minting source, the closer you are to that spigot, the more purchasing power you have, right? And that is the Cantillon effect. So if you can position yourself to parasitize as closely as possible to the spigot, then you can take maximal parasitic advantage of that Cantillon effect of that higher purchasing power of these new dollars that exists before those dollars depreciate themselves via inflation.
Jim: And if you’re a bank who has access to the Fed window or a major corporation who can issue bonds in cooperation with people who have access to the Fed window, then you are at most two steps removed from the issuance. So yeah, I can see that. That makes a lot of sense.
Matthew: Precisely.
Jim: All right, now we’re getting kind of a little bit long here on time, but that’s all right. At the end of your tweet, you point at a hopeful direction. So I’m going to read a fairly long couple of senses. And this is how maybe there’s an opportunity to produce local or at least minority networks that reverse this pattern. In other words, it allows us to reestablish a low time preference culture of trust among those who apply their energy and insight towards generative as opposed to parasitic endeavors. It rewards those who can and even especially amidst our contemporary world’s attentional race to the bottom, establish and effectively coordinate action toward and most importantly, build trust around a vision of the future that transcends the immediate gratification and self indulgence so prevalent in today’s widespread cult of parenthesis. I think I know what you mean there, which is it’s possible to build networks in the midst of this shit that nonetheless are not tuned the way the shit is tuned. Is that what you’re trying to say there? If so, put a little bit more meat on that bone.
Matthew: Yeah, I mean, I think it’s helpful. I’ll go back to a visual metaphor that I used in a, I think in essay we’ve discussed previously on hyperconnectivity, but I use this metaphor of like ants have this response where if you actually certain ants, you put them in water, they can actually link up with one another locally to form rafts. And those rafts can actually float in a way that is far more likely to create or improve the survival odds of any given one of those ants, right? So like if they find themselves, let’s say in a massive rainstorm, they can form these kinds of local groups that allow them to weather the chaos that was induced around them. And so it’s a way of thinking about how we might use some of these trustless pieces of infrastructure or how the signaling mechanisms like Bitcoin with its stable coordinate heartbeat can allow us to directly cooperate with one another to create local networks that have a different time preference than the average network does.
And this is why I say it rewards those who can even and especially amidst our contemporary world’s attentional race to the bottom, establish effectively coordinate action toward and most importantly, build trust around that vision of the future, etc. So what I’m saying there is precisely that if there is actually in the midst of a culture that can’t plan ahead, that can no longer think about the future, if you can leverage these technologies of higher coordinate capacity of lower time preference of more stable foundational behavior to coordinate and cooperate and build things that do have that future orientation, you can dramatically outcompete those who are not actually taking advantage of this over the long term, right?
You can actually weather the storm and end up in a position that is far more advantageous than you otherwise would have been in after the storm. And I do think we are seeing this, I think we are seeing micro communities emerging with an ability to directly coordinate around systems that are ostensibly trustless. I think this is more effective in communities where you actually have stable, trustless infrastructure like Bitcoin, because I think as I’ve argued in our previous conversations, I do think that is fundamentally a binding mechanism to a more stable energy gradient. But if you are able to directly connect to that and bypass all the corruption and not pay those costs in your micro community, in your project, in your company, if you’re able to have a more stable, fundamental foundation of that coordinate of infrastructure, in the long term, you can outcompete those who don’t. And the issue is it takes a long term orientation, it takes a low time preference, and therefore you actually have to have trust. Those communities actually have to be communities of trust.
So you have to build those relationships, you have to actually understand that the people that you’re working with can be counted on over time to resist temptations and to actually fulfill the obligations that they’ve made to one another. And those communities, I think if you have those two sides of the coin, if you have those two pillars in your community, you orient around the infrastructure that enables us to bypass the corrupt previous infrastructure that is parasitically captured, and you have built communities of trust, then you can actually focus your eyes on the horizon together and build things that we will, that will fundamentally save us from ourselves, at least maybe for another century or two, right, give us some time. So that’s kind of what I’m getting out there.
Jim: Yeah, I would add to that, by the way, this is where I think we disagree that I think that the trust components and such and firm foundations and less noise in our projections of the future are all good. But there’s also engineering questions, right? That the signaling modalities have to actually work for cooperation. And for instance, I continue to believe Bitcoin is just badly designed because it’s inherently deflationary. And you’re not going to get the proper signaling from a deflationary currency just can’t happen. We could talk for five hours one day about that we probably But
Matthew: I would go back to my similar arguments that I don’t think I’m not someone who says Bitcoin only, but I do say Bitcoin foundationally, I think that other things emerge in that constellation, that ecosystem, that scratch the itches you’re talking about. But I think that a fundamentally deterministic mechanism to tap into the energy gradient of the universe. And using that as a stable heartbeat at the center of it all, which is why I call Bitcoin a metacentralizing attractor, not a decentralizing force. I think that is our best way to at least currently outcompete the corrupt infrastructure on hand.
Jim: Part of the design parameters for game B, which we have not actually done finished a little working on it, is that game B should be able to outcompete game A, at least in narrow wedges initially, which will rise over time. And so it is indeed the game B hypothesis that what you’re pointing to is possible and can be done. But the more I’ve dug into it, more other people have dug into it, it appears that it’s a very complicated, complex problem, not that different from Stuart Kaufman’s autocatalytic networks.
And I think it also explains why most trials so far have failed, because they’re not sufficiently big to have autocatalysis, right? Why did the hippie communes of 1972, 73, 74, 95% of failed within five years, a few of them are still around, but they’re weird little isolates, right? They did not have an r greater than one, they did not grow and take over the world, because they lacked a whole series of components that made them a new form of life in contrast to the old form of life. And so I think the problem is that most people underestimate how many different elements have to break simultaneously brought together to produce what’s the equivalent of a new origin of life event. And if you’re missing even one of the critical components, it won’t fly in the same way.
If you knock enough genes out of the smallest autotrophic life forms, they’ll die, right? There’s a minimum amount. And and every attempt so far has been way under the minimum amount. And that Bitcoin might or might not be even a part. But let’s suppose it is by itself, it’s nowhere near enough to produce this autocatalytic network that can thereafter grow forever and event with an r greater than one and eventually out compete game a because it is the interesting thing about life is it happened once as far as we know. And every single one of us is a direct descendant of Luca, the last universal common ancestor, because it never lost its r greater than one ever on a statistical basis over a period of time. And that’s what the real job is. We both see it that there’s an alternative engine. But the question is, what does that engine have to be?
Matthew: Yep, I mean, I enthusiastically agree with that. This is why I personally spend my life right now, my professional life, what I’m building is is precisely the kind of modeling toolkit that is able to understand the causal closures within networks, and the degree to which those causal closures allow that network to use information to improve its model of the world, and then also to extend those causal closures and understand how its externalities might become the inputs or become an affordance for new members of that network of that new closure. Like so so I am literally I mean, this is why I make the case for Bitcoin as this centralizing metacentralizing attractor. But I am fully aware and 100% believe that we need other tools to help us understand and navigate the kinds of more complex auto catalytic questions that you are bringing up there. And I’m spending my professional time doing exactly that so.
Jim: Very cool. We had another great conversation with Matt Percowski. Look forward to further conversations, I get a little further along with our map of how we actually sort of think the critical mass of origin of life for Game B occurs. I’d love to run it by you and get your thoughts.
Matthew: That would be wonderful. I mean, I think our set of conversations is also a wonderful and exemplary embodiment of this process of creating a reciprocal and trusting relation over time that improves its quality and improve its capacity to communicate, and also therefore improves its effect on the world over time as well. So I’ve loved this series of conversations and would love to see it continue.
Jim: We’ll certainly do it.