On Hygge, ergodicity economics and Risk Bushido

Good question, I did not think to ask. I think it was @Joriam who came up with that name… what say you, Joriam?

Pete - sure! Ergodocity is not directly relevant to Hygge, but here are my go-to sources:

If ergodicity is related to Risk Bushido, then it is related in some sense, at some time, to Hygge. I’m starting to think I’m starting to get it. Ergodicity is the reason why most people who invest in stocks now put their money into huge low-fee funds that track market indices. Because without fees and by cutting out much volatility you end up with low but steady growth. Right? So Hygge is that group of people who have decided they can’t beat the market. But you see, in the real (US) market, the person who has won the most over the long term is Warren Buffet – he’s the bland gray guy who invests in value stocks for the long term. But since the financial crash Mr Buffet has not done well, because we are in a time of transition from industrial stocks to tech stocks, and so the firm foundation that he thought would exist forever is now blown to smithereens. Does this make sense? So this is the influence of Goro – the long term small-scale tick-tick of evolutionary change eventually leading to a paradigm shift. The master who can ride that paradigm change becomes the next “most influential citizen”. So long-term Ergodicity is to be found in Goro’s actions and leads to evolution in Hygge.

Reading some of the conversation between Yudhanjaya and Joram and sometimes Alberto:

I like Joram’s approach of “faux-market”. Have you both read the story in Asimov’s “I, Robot” called “The Inevitable Conflict”? The reason why I ask is because the Machine which controls the economy of the Earth in that story seems to have significant parallels to Sata (actually to the combined State Machine). But this story emphasizes the “inevitable conflict” between the vision for humanity which the Machine is following (which is never revealed in the story) and the human desire for individual agency and free will. The Machine in that story is suppressing anti-Machine groups in society. In the Witness world there is far more instability. Hygge, as Joram remarks, is the remnant of the central group of supporters of a conformist, AI-controlled society, but other Distrikts are much more favorable toward human-sourced ideosyncrasies.

So the nature of the State Machine(s) is critical. Are they “mere” creatures of programming, or do they have some form of independent decision-making capacity based on some sort of foundational operating system? Asimov had the Three Laws of Robotics, which ultimately guided the Machine to work toward the good of humanity, whatever that is – this is the reason Asimov gave for trusting his robots, and the same applied to the Machine. It seems to me that giving Kiri and Sata and Goro some independent agency would be good for the plot, but not of the “robot-takes-over-the-world” type. One of the best things about Asimov’s system was his ability to reason an outcome of a robotic choice from first principles of the Three Laws.

Denton, as the creator of Hygge, seems to me to be the best source of both the foundational totalitarian impulse of a nanny state built around AI, but also of the schizophrenia of three split personalities of the former State Machine. So even though he is off the scene, he would be source of the slightly cracked controlling intelligences that set things up and allocate resources. Another reason why I think the AIs need individual agency is because otherwise they would end up controlled completely by programmers and those who feed them information. In that case the story of Hygge (and to some extent the entire world) would be the battle over control of programming of Sata, Kiri and Goro.

From the economics side I think Alberto’s idea of a social democratic economy would make sense – there is no reason why an AI could not set this up. Besides, from Denton’s point of view, supposing he was interested in human well-being, allowing people to have productive jobs makes sense. He would have to have been VERY cracked after 1991 not to have noticed the faults of Gosplan. He would also have noticed that such a system, although it benefited in the short run from being the successor to a hugely mismanaged and inefficient economy, ultimately disillusioned its citizens so much more than modern Germany that they primarily considered their jobs from the point of view of how much they could steal… And social democracy also introduces conflicts which could lead to stories, such as the one Alberto mentions between haves and have-nots. So Sata would not be running a UBI, but would be managing a REAL but constrained functioning market economy built around exchange value (not labor value, Alberto), and complemented by social support for such basic needs as education, support for those who cannot or can no longer work, and health care, managed by the AI in such a way as to ensure that prosperity is maintained.

Yes, that makes sense. My modifications to Goro took it from being an aesthetic part of Hygge to being a critical element that induces randomness to perform some sort of social experiment that lets it create, observe and trigger paradigm shifts.

The State Machine is a social democratic system where the ‘democratic’ is increasingly turning to performance theater because of the sheer complexity of the system: however, it ensures free education, healthcare, and a certain Keynesian Welfare State approach. It had individual agency. @Joriam, what are Kiri and Sata optimizing for, and what principles are they running on?

I could certainly see a political system where politician-programmers constantly propose new updates to Kiri and Sata, and the public votes on them: and this is a very highly contested arena. And furthermore, systems as complex as Kiri and Sata, no matter how deterministic the proposed updates, would exhibit chaos. More so with Goro in the mix. And thus there will be interesting histories of simple/good ideas that led to complex, disastrous emergent effects.

Re: Denton being the origin of this schizophrenic system. I am still not entirely sold on why it exists in the first place, but this conflicts with @Joriam’s canon of an intentional split. I would like to see some rationale for this - I can picture Denton splitting the State Machine up to give it less power over his increasingly curmudgeonly life, but we will need some consensus on this.

@petussing @joriam what do you think? Also, best if we move this discussion to Hygge. Easier to keep track of, and more relevant there.

Not exactly. The best way to understand the Peters paper, for me is to write some code and run it a few times. I have seen the “oh s##t” look on the face of at least one mathematician. My take (and some code) are here.

But yes, Risk Bushido is related to ergodicity economics in the following way: if you accept that our world is not ergodic, you will make choices that will look like risk-averse to someone maximizing expected payoff. This is Bushido: live to fight another day. It’s finance as Sun Tzu would do it: “never attack if you are not sure of winning”. Commanders that risk losing the army are bound, in the long run, to lose it… and then there is no coming back.

Moving this discussion to its own topic.

Alberto – I’ll give it a try. There is a lot to learn. I am working on figuring out Hygge first. OK, First question: What? Hygge? It’s a Danish word that denotes coziness and conviviality that engenders a feeling of contentment. Like Gemutlichkeit in German. Why?

I have plowed through Ole Peter’s Ergodicity article. I can’t say that I understand. I know about ideal gas laws, but the connection to economics is fuzzy to me. Could Yudhanjaya refer me to a less mathematically heavy explanation?

I like Kiri, Sato & Goro: Of course Goro is particularly good – I think Asimov would have been horrified, but Lem would have been happy – I’m on Lem’s side in this one…

1 Like

Hey, there’s some strange formatting going on and I feel the first post I can see here is not really the original thread (??) but let’s brush that and lemme tell you about my choice with the name Hygge.

Yes, Hygge is that Danish (and also Norwegian, it shows up in a couple of Nordic cultures) word that means cozy, comfortable, but in a particular… homely way.

“Hygge” was a rebranding choice.

The Distrikt was trying to get away from the forceful external politics of Denton (after his retirement) and starting a new era of diplomacy and compromise. The name was still up for grabs and it was a good opportunity to send a message.

Hygge also comes in direct relation to “Viking” (as in Project Viking), as a way to both recognize its origins but also propose something completely different — I was trying to be clever working with the violent image of the ‘old vikings’ and the peaceful image of the ‘Scandinavian countries’. But maybe I was just confusing.

I wouldn’t mind changing the name, but I do think it was well chosen for the context.

That is an excellent question. I built this whole system under an impression that those technologies would be far too advanced to need the crude goal inputs we need right now. I was imagining something like an AI that can check over time what is the average will of a population and adjust accordingly — but the idea of a pre-established optimization direction also sounds interesting.

The way I constructed Hygge, those optimizations would need to go to the direction of:

  • The least amount of conflict
  • The least amount of drama
  • The least amount of misery
    Always focusing on minimizing, never maximizing.

In the canon I created, the system is implemented post-Denton. It’s true that the division of the system is a direct response (and adaptation to) the Denton pro-party, but not the man himself.

Yes please! Idk why I’m only seeing a chunk of this conversation and not the whole thing!

Working on getting it, because it seems to me that Risk Bushido is critical to this World, and that ergodicity is a key concept, and a key way in which Goro in particular works. It’s not clear to me that Peters is using the concept strictly – his conclusions in some ways seem to be extrapolations of the mathematics in ways that cannot be mathematically demonstrated. But yes, I see what you are indicating. Again, we get back to the fundamental rule of gambling: to win, you must take small risks that you know by the odds will increase your income; people who make large bets lose in the long run. So Warren Buffett wins. Unless the paradigm shifts…

Sorry man, this was in the “introducing worldbuilders” topic, and it was taking a life of its own, so I split it to its own topic.

Pete - in narrative sense, Goro and many of the events of Hygge would have been before Risk Bushido was invented by Kintoki: so we can iron out one without necessarily effecting the other. Risk Bushido was written in the market-led Libria, so it can be spreading from there without necessarily needing to be the backbone of the world here.

Indeed, we’re running into some narrative conflict here by forcing one on the other. So, to respond:

“Hygge” was a rebranding choice.

Okay, now I see. Joriam, you really should explain this in the main text itself, instead of just here in the comments. Framed this way, both your choice of Hygge and Nadia’s aversion to it make sense: it implies a certain “we know what’s best for you” aesthetic that perfectly fits into the paternalism we’ve already established. It is indeed well-chosen, but please show your hand next time.

Re: the rest:
We’re having a devil of a time justifying some of this initial canon. I’m going to propose a controversial change: we get rid of Kiri and Sata, because they’re essentially doing exactly what the State Machine is supposed to be already doing in this world - providing a kind of maternal state that reduces the pain on the bottom at the expense of the risk: reward system that, say, Libria would provide, plus a sort of limited direct democracy: everyone votes, but the choices you get to vote on are limited, because nothing that truly rocks the boat will be presented.

The least amount of conflict
The least amount of drama
The least amount of misery
Always focusing on minimizing, never maximizing.

Basically the ultimate welfare state, operated by an AI that gathers as much behavorial data as possible - from household power consumption to message logs to marketplace purchasing decisions.

And we let Goro be the product of another agent within Hygge - perhaps a rogue CIVICSMOD creation - someone trying to counter this ‘no-risk’ attitude towards life by injecting some randomness.

As a result of this:

  1. we minimize the number of Deus Ex Machinas which break world consistency and internal logic

  2. Risk-Bushido becomes less important to Hygge, @petussing - because, as Alberto pointed out, Risk Bushido is about making rational decisions in a life of risk and uncertainty. Hygge is literally the counter: a world where life is so well managed almost everything is certain. Risk-Bushido canonically became popular in Libria for this reason. Canonically, Hygge-Bushido is a variant about political marketing, much like Machiavelli’s Prince.

  3. Hygge, as a conclave of hardcore Denton supporters, better fits into the top-down, data-heavy “the state knows what’s good for you, child” system that countries like Singapore are pioneering. This gives it its own economic flavor that contrasts nicely with the rest of the Distrikts but has a cohesive system of economics behind it.

In a way, I am acting like Hygge itself - optimizing for the least amount of conflict so that we have a coherent world canon, that has explainable roots, that other people can build on without having to ask “why is this here? What was the logic”?

@alberto, could you design an economy that optimizes for @Joriam’s statement here, as you did for the Covenant?

The least amount of conflict
The least amount of drama
The least amount of misery
Always focusing on minimizing, never maximizing.

I think I did something bad with Hygge, it seems to me like the scrutiny I see applied to my Distrikt is on a completely different level than anything else in this project. There’s very little ‘yes and’ energy and that shows something (particularly in this very accepting team).

I’m sorry, Yudhan. When you gave me that task I felt I was ready for it, but I clearly wasn’t. I should have understood my limitations and kept to what I know best.

I’m not gonna die on this hill, everybody has issues with Hygge. Not even the name passed without a fight. Besides, my capacity these days has been limited and I don’t think it’s productive to come here and ‘defend’ something, which is how I’ve been spending my Economics Lab energy instead of creating stuff.

To @yudhanjaya’s proposal of:

I raise I counter-proposal: let’s delete Hygge altogether.

We keep adjusting details, rewriting paragraphs just to rewrite them again. It’s a waste of energy.

I made it, I can burn it.

I think I was a poor choice to build a Distrikt from scratch. I still fail to see what the problem is with many of the things pointed as hard-to-buy, it just proves my naivety. I’ll instead focus on particular aspects of other Distrikts that I feel I have more knowledge about, and also doing some community networking.

I’m very sorry.

Well, if Hygge is a ship on fire, I still see things to salvage before it vanishes. I like Goro, and the panopticon-like structure, and I think those play well with the guiding principles “the least amount of misery” - one to add a bit of spice to life, the other to enforce the kind of surveillance required by a state that would eternally seek to minimize risks for its citizens. And the name-as-rebranding, now that you explain it, makes perfect sense.

There are things that work. And our arguments here inspire a political history of for-and-against that, if storified, add depth and weight to the world.

I have many promises to keep, and proverbial miles to go before I sleep today, but if you don’t mind: let Hygge stand for now, and see what you can add to other Distrikts. Almost all of them, for example, are missing aspects of the arts and entertainment, and we’re even a little thing on the ground in terms of visual landmarks. I’ll take a crack over the weekend at Hygge-II and set up some scaffolding that can take a Red Plenty:

"Strange as it may seem, the gray, oppressive USSR was founded on a fairy tale. It was built on the twentieth-century magic called “the planned economy,” which was going to gush forth an abundance of good things that the lands of capitalism could never match. And just for a little while, in the heady years of the late 1950s, the magic seemed to be working. Red Plenty is about that moment in history, and how it came, and how it went away; about the brief era when, under the rash leadership of Khrushchev, the Soviet Union looked forward to a future of rich communists and envious capitalists, when Moscow would out-glitter Manhattan and every Lada would be better engineered than a Porsche. It’s about the scientists who did their genuinely brilliant best to make the dream come true, to give the tyranny its happy ending. "


No, you did nothing wrong… even though the result is still incoherent. Part of the responsibility is mine: I basically careened in, noticed your emphasis on UBI, thought “OK, this is a redistribution measure, I need to imagine a corresponding organization of production. What kind of economy would focus so much on equality? Well, a Nordic-style socialdemocracy on steorids, for example. OK, so a market economy with heavy redistributive corrections, and low Gini coefficient as a serious priority for economic policy, rather than an afterthought”. So I sketched that in.

But in doing this, I made a crucial mistake: I overlooked the other bit you had written, the allocation-making AI. If it’s a market economy, the AI is left with not much to allocate. So I tried dropping the market, and following the AI idea (hence the Red Plenty reference), but then I needed a plausible way that allocation-by-computing would not sink on the politics, in the way that it did in the historical Soviet Union. And there we got stuck.

We should probably do two things.

The first is keep developing a hyper-social-democratic Distrikt. Reason: it is a good way to experiment with a variant of a familiar (and, for many, very seductive) real-world model, economists might find added value for thought experiments. The name Hygge is actually on point there. We could even make it explicitly neo-Scandinavian, and talk about Folkhemmet and what not.

The second is dream up a Distrikt that runs a (conceptually much harder) planned economy, AI powered. Making this feel real is now beyond my power, but the wiki will grow, and smarter people will help out, for this is an interesting problem. For example, @amelia and her fellow scholars of smart cities might have some insights of what makes algorithmic decision acceptable when it does not favor you, and what a political debate around the algo’s objective function would be.


1 Like

My vote if I get one is to continue Hygge. It makes sense as the Distrikt that preserves Denton’s vision and much of the machinery he created.

I have been a team member in group projects before, and it is hard to accept suggestions from others – I think it is great self-training, not that I’m always good at it, even now. At best, it forces us to focus on the rationally best choice, and stop thinking about “this person suggested this, and I respect them, so I need to accept it” – the preparation has to be a form of Bushido – “this is valid because it follows a logical valid path” – it is not about technical preparation, which comes with practice, and this is the very beginning. Reminds me of Robert Browning:

“Grow old along with me! The best is yet to be, the last of life, for which the first was made. Our times are in his hand who saith, 'A whole I planned, youth shows but half; Trust God: See all, nor be afraid!”


Nice one, @petussing! Going to run with this. :slight_smile:

Lot of sense being made here. So let’s try this two ways:

I feel that @joriam’s framing:

The least amount of conflict
The least amount of drama
The least amount of misery

Is important. It goes well with the name; it is an interesting counterpart to the rest of the economies we’ve set up: and crucially, it is a good backbone for both superScandanavia and Red Plenty-style economics to come about. I am reminded of Otto von Bismarck’s 1884 speech:

“The real grievance of the worker is the insecurity of his existence; he is not sure that he will always have work, he is not sure that if he will always be healthy, and he foresees that he will one day be old and unfit to work. If he falls into poverty, even if only through a prolonged illness, he is then completely helpless, left to his own devices, and society does not currently recognize any real obligation toward him beyond the usual help for the poor, even if he has been working all the time ever so faithfully and diligently. The usual help for the poor, however, leaves a lot to be desired, especially in large cities, where it is very much worse than in the country.”

“Call it socialism or whatever you like,” Bismarck said during the 1881 Reichstag public policy and budget debates. “It is the same to me.”

Always focusing on minimizing, never maximizing.

Speaks to the inherent decision biases that any such effort might take, because now we hit on the question of: what is fairness? If one group wants to maximize gain, and another wants to minimize risk, what is seen as the fairer decision?

Fairness is a concept that constantly mutates bases on space, time, social constructs and practicality. Kleinberg et al. [*] posited that, assuming a population divided into groups, the output can be shaped by one of three mathematical notions of fairness:

  • Calibration within groups – for each group and each bin the expected number of members with a positive outcome should be proportional to the score assigned to that bin.
  • Balance for the positive class – the average score of members with a positive outcome should be the same for each group.
  • Balance for the negative class – the average score of members with a negative outcome should be the same for each group.

They posited that that no method can satisfy all three notions of fairness simultaneously, with the exception of highly constrained special cases:

  • Perfect prediction – for each feature, we know for certain what the outcome is.
  • Equal base rates – the two groups have the same fraction of members that have a positive outcome.

I feel like this split required will naturally lead to two different implementations of Hygge competing for the goal of a fair society. One, superScandanavia, trying to optimize for equal base rates. The other, Red Plenty, relying on perfect prediction.

@Joriam, you have not done bad, but given us an interesting intellectual challenge. Like I said earlier, I see things to salvage here.

This requires some more thinking, but let me return over the weekend that gives us not one but two Distrikts to play with - possibly a ‘Big Hygge’ and a ‘Little Hygge’ (names we can change later). Let’s not delete this. Give me a few days to think and write something that then can be shaped by @alberto and @petussing - I think I have a fair idea of your style of plugging in now, enough to accommodate.

[*] Kleinberg, J., Mullainathan, S., & Raghavan, M. (2016). Inherent trade-offs in the fair determination of risk scores. arXiv preprint arXiv:1609.05807.

1 Like

Woot! I think it would be a pity to scrap it - the discussions raised by it are already interesting at this stage. And we have barely begun! Maybe we need some nice ground rules/social norms around how we propose/develop stuff in general. You know like the whole occupy hand signalling but adapted to the medium: Agree :+1:t5: - Agree and :writing_hand:t5: - Let’s fork :vulcan_salute:t5: .

just made that up with little thought but you get the idea.

Folks, of course you can salvage any bit of Hygge you find useful: the framing, Goro, BUI or whatever else. Please feel free! But I guess the word salvage says everything.

My thinking was that rewriting Hygge was a bit like a dog chasing its own tail, as every change asks us to change everything else. Changing Denton’s death changed a bunch of stuff and a lot had to be rewritten, for example. Yudhan’s suggestion to remove Sata and Kiri would also requires rewriting just about every section of the wiki.

My grand take is: rewriting hasn’t been efficient, maybe somebody should write the whole thing from the ground up in a way that makes sense within itself — and that means having the freedom to change everything if that makes more sense. It would take less time than adjusting over and over.

I’m well aware of the ‘somebody should rule’, but in this particular case I’ve already done poorly in my first attempt. So here’s me asking for help and wondering if I could pass the baton. Being Hygge’s advocate hasn’t been productive for me and also I haven’t been fixing Hygge quickly enough.

I guess using Nadia’s signaling (which is a good idea), I’m here proposing a fork :vulcan_salute: and seeing if somebody would take it.


Oh! See? I clicked my notifs in order and just now I’m seeing @yudhanjaya already went ahead and did the fork — brilliant!