Competitive, Cooperative, and Cohabitive
post by Screwtape · 2023-09-28T23:25:52.723Z · LW · GW · 13 commentsContents
I. Background: The Two Genres II. Something is Missing III. Cohabitive Games: Definition and Examples None 13 comments
Summary: I think there's a third genre of games, which I dub Cohabitive games, where your victory isn't tied to other players winning or losing.
(I’ve been writing this in bits and pieces for a while, and Peacewager [LW · GW] was the impetus I needed to finally stitch it together and post it. Peacewager sounds like a really fun game and an example of the thing I’m talking about, but I do not want this whole genre to get called Peacewager Games when I think I have a better title for the genre.)
I. Background: The Two Genres
I believe there is a missing genre from existing games, and this genre feels large enough that it should contain maybe a third of the games I can imagine existing. More serious game players or game theorists might already have a name for the thing I’m pointing at, though the first three game design majors I asked didn’t know of one.
Let me back up. I’m going to assume for the moment that you’ve played some games. I don’t have a strict definition of “game” I’m working with here, but let's start with definition by example: Chess is a game, Hide and Seek is a game, Pandemic is a game, Apples to Apples is a game, Poker is a game, Magic: The Gathering is a game, Werewolf is a game, Among Us is a game, Hanabi is a game, Baseball is a game, Football (American or European) is a game. I’m not trying to do some narrow technical definition, I’m waving my hand wildly in the direction of a pretty natural category and I’m not planning to do anything weird with the edge cases.
Chess is a competitive game. In chess, you’re loosely simulating a war between two evenly matched factions. When you play chess, there will be one winner and one loser. Sometimes instead there will be a draw. Anything that is good for you when you are playing chess is bad for your opponent and vice versa. You can be mistaken about what is good or bad for you; you can offer trades of pieces to your opponent because you think it is a good trade for you and they can take the trade because they think it is a good trade for them, but this is ultimately what’s called a zero sum game. Your loss is their gain. “Eurogames” where you’re trying to get the highest score are competitive in nature; if you could pay ten points to cost every other player twenty points, you’d do it.
Pandemic is a cooperative game. In Pandemic, you’re loosely simulating a global pandemic and the response of the international medical community. When you plan Pandemic, either all the players win, or all the players lose. Anything that is good for you when you are playing Pandemic is good for your teammates, and anything that is bad for you when you are playing Pandemic is bad for your teammates. You can lose things for yourself; you can spend resources and pay costs and run out of good cards in your hand, but ultimately this is also a loss for your team since they want you to have good stuff.
There is of course the circumstance of competitive team games, like football. If I’m playing Football, I’m trying to help out my team like it’s a cooperative game, and make the other team lose like it’s a competitive game. This adds a little to the picture, but doesn’t change the basic dynamics much. Again, I’m not doing anything weird with edge cases here. There’s also multiplayer games like Risk, where it might make sense to make a temporary alliance to cooperate with another player while still ultimately knowing only one of you can win. Hidden role games like Werewolf or Betrayal At House On The Hill are usually competitive with teams. (A team of one and a team of the rest of the players is basically a team competitive game.)
Picture these as two points on a continuum. You can compete, or you can cooperate. Seems simple enough. You can, if you like, extend this into a metaphor for how humans relate to one another outside of just games.
II. Something is Missing
Except this really isn’t how human beings actually operate in a wide range of circumstances. I’m not competing with my roommates over who does the chores, but I’m also not cooperating in the sense that I cooperate in Hanabi or Pandemic. All else being equal, I’d rather someone else did the dishes! Alternatively, look at the stock market; yes, I’d rather make more money than less money, but my gain is not exactly your loss. Sometimes there are individual trades that leave us both better off, and not because one of us is making a mistake either.
I once ran what was basically a stock market with play money [? · GW], and ran into some weird behavior towards the end of the night. Some players were making very high variance trades, ones that probably were going to cost them play money in expectation. When I asked, they pointed out that since another player was a little bit ahead of them and there wasn’t much time left, it didn’t matter if they lost by a little or a lot but it did matter if they could somehow climb into first place.
This is not really how the stock market works.
In the stock market, I don't particularly care if I make more money than Warren Buffet. I care if I make lots of money. If I could pay ten dollars to cost Warren Buffet a hundred billion dollars and thereby narrow the gap between us, I wouldn't do it, because then I'd be out ten dollars. I'd rather spend ten dollars on a pizza, because at least then I'd get to eat pizza.
III. Cohabitive Games: Definition and Examples
I dub things like the stock market in this way to be "cohabitive," from the word "Cohabitation" which means "the act of living together." Cohabitive is a third thing, roughly the same level of category as "cooperative" and "competitive." You want to do as well as you personally can, but it’s not worth you losing points in order to make other people lose more points.
You’re happy to ally with other players if it gets you points, but you wouldn’t help them if it cost you points even if they would get more points than you’d lose. You aren’t explicitly cooperating with others, you aren’t setting out to compete with them, you’re simply living in the same space at the same time. If it turns into a stag hunt [? · GW], you're not scored on the collective point score or having the highest score but on how many points you earned, without reference to anyone else.
When you start looking for them, you find that cohabitive games do exist! Many MMORPGs have the cohabitive nature, where you don’t mind if someone else gets a rare loot drop but you want as many rare drops as you can for yourself. The scoreboard of a pinball machine may be ranked, but it’s a perfectly normal goal to just want to get as high a personal score as you can and that has the cohabitive nature. Track and Field sports such as the hundred-meter dash declare a victor by comparing the fastest times of different runners, but we still talk about Personal Records as a distinct thing from World Records. We just don’t have a name for this thing, and many otherwise fine cohabitive games have competitive or cooperative win conditions awkwardly grafted onto them.
Cohabitive games can have scores, and the goal is to beat your personal record. Cohabitive games can have victory conditions, where you might win or lose depending on whether you met those conditions independent of whether other players met the conditions. Cohabitive games can allow or even incentivize aggressive behavior, as long as your victory and defeat aren't actually tied to that of the other players.
Let games be cohabitive! Let us play together, not with the goal of sacrificing ourselves for the cooperative victory, not with the goal of crushing our opponents beneath the heel of our competitive dominance, but simply existing together for a time in the same game and seeking our own personal victories.
13 comments
Comments sorted by top scores.
comment by Richard_Kennaway · 2023-09-30T09:05:45.241Z · LW(p) · GW(p)
There is a concept in Japanese, sessa takuma (切磋琢磨), which lives in the same space as these, meaning "joyfully competitive striving for a common purpose", or as Wiktionary puts it:
- applying oneself and studying hard to cultivate oneself
- friendly competition, improving oneself by learning from others
It is often used of the situation of people training at a martial arts dojo, but would also apply to a band, a football team, a cycling club, an MMORPG guild, and so on. It could even apply to a workplace.
Replies from: MakoYass, Screwtape↑ comment by mako yass (MakoYass) · 2023-10-01T22:50:21.781Z · LW(p) · GW(p)
The concept of jita-kyoei, part of the philosophy of judo, may also be relevant.
When we observe at the actual lives of people, it seems that there is a great deal of wasted energy. Even if it appears that people are utilising their energies effectively, it cannot be denied that there is still much room for improvement. We should cease meaningless conflict, and instead abide by the principle of Jita-Kyoei.
comment by mako yass (MakoYass) · 2023-09-29T20:26:27.572Z · LW(p) · GW(p)
Peacewager wasn't intended as the name of a game, it was intended as the genre name. "Peacewager 1"/2 are games, but that definitely wont be their release names.
I'm not sure yet, but I think "cohabitive" is probably a better name for the genre, for reasons I'll have to DM.
I wonder why I didn't go with it. I've considered "coexistence games", but there are many kinds of coexistence, so it didn't really resonate. When I talk about misalignment risk, I tend to describe it as "dehabitation" (It's not that a misaligned optimizer will want to get rid of us, as an end. It will want our space, and our sunlight, and we will be diminished in just the same way humanity has diminished so many wild species.) Maybe I just never got around to adding these words together.
I think one objection is that, "peacewager" implies the prior presence of war, and proposes a kind of militant peace, which puts people into a mindset where they're going to rethink a lot of assumptions they'll have about how peace holds together. And if you don't do that they'll bring the same bag of assumptions to the table as they do in conflicts irl and they might just not learn anything, but they also might be invited to question those assumptions by whatever happens in gameplay, unsure. Maybe that is not the job of the genre name.
Replies from: Screwtape↑ comment by Screwtape · 2023-09-29T21:08:45.665Z · LW(p) · GW(p)
I feel strongly that the name for the genre should be four syllables long and start with "co-" to match the other two. After that I have a bunch of weaker aesthetic feelings on the exact word.
If you try and cohabitate with something much stronger than you and misaligned with you, I think you wind up pushed off the game board unless Ricardo's Law of Comparative Advantage saves you somehow. When playing around with economics games it's fun to see when that does or doesn't happen. (Well, fun if you're me.) Very summarized conclusion so far is that giving all players the same number of turns is enough to save them if everyone is paying attention, since turns are usually valuable.
I notice I don't think of cohabitive games as necessarily peaceful. A game of Risk, but with the victory condition of "At the end of ten rounds, your score is however many armies you control" would count in my head as cohabitive and the process of getting there has a good chance of becoming a bloodbath. A lot of my attempts at the genre come out as trade wars or economics based, but that's reversing the order of things; I kept trying to make games about manufacturing or logistics or economics and kept running into (from my perspective) really dumb behavior in the endgame as people tried to take 1st place with longshot bets.
Replies from: MakoYass↑ comment by mako yass (MakoYass) · 2023-09-30T01:27:31.611Z · LW(p) · GW(p)
It's noteworthy that the law of comparative advantage stops being true exactly at the moment we figure out how to make something that's generally better at the work of humans (strictly, it would have to have better manufacturing and running costs than a human too, but you know it wont be long after AGI before we reach that point)
It would be very easy for a game designer who lived only in the human era to omit this additional factor of violence that we can infer would be present in post-human economies, but I think we should study it.
Oppression might not be very different though.
A game of Risk, but with the victory condition of "At the end of ten rounds, your score is however many armies you control" would count in my head as cohabitive and the process of getting there has a good chance of becoming a bloodbath
Mm, that's neat, right. You only get (immediate) stable pluralism if defense > attack, or if there are diminishing returns to scale. (I think you get a new pluralism later on when the victor develops deep internal specializations but that's less relevant to present day meatbags)
I was going to say, economics (and so, necessarily, industrial growth as well?) is a great subject for games in this genre, because then you can use these games to test economic reforms.
If that's the object of study, then I think it might be fine to just contrive pluralism? Human governments contrive pluralism, because humans/human cultures value survival over the prospect of growth (if we can trade the possibility of attaining global domination for the guarantee of security/survival, we will make that trade, and most of the free world explicitly did).
For my games, the object of study is lower down, cooperative bargaining, the art of cohabitation sole. I want games to at least start without the assumption of the existence of money. I can't really easily explain why we have money in the same way I could explain why we maximize EV, I'm not sure I'll be able to present a clear principled argument for money unless money emerges within the game from pre-money conditions. I'm personally fairly comfortable with the existence of money (though very doubtful of single-global-currency maximalism. I think we'll have complete designs for plural money at some point pretty soon) but I have a lot of sympathy for those who cannot embrace it, and they want cooperative bargaining, and if these games give rise to a principled argument for money, they'll need to play that through themselves and experience it (and so do I).
Replies from: Screwtape↑ comment by Screwtape · 2023-10-01T09:33:50.083Z · LW(p) · GW(p)
I think the law of comparative advantage keeps being true even if the other agent is better at all the things than you are (imagine if I can gather one coconut a day or two gallons of water a day, and you can gather ten coconuts a day or five gallons of water a day; I think it's still worth trading my water for your coconuts?) but stops being true if they're better at all the things and they can freely copy themselves. So yes, in a full AGI foom, humans are kind of useless. From the perspective of trying to make fun games, any exchange rate offered between other resources and extra turns needs to be very finely calculated and I have never managed to tune that in a way that was fun for everyone.
Testing new economic reforms and studying what happens with AGI seems quite in scope for cohabitive games, including the part where sometimes the stable balance point is with some players getting pushed off the board entirely. I've found playing around in this space to be a good intuition pump for how pushing players off the board isn't necessarily the result of malicious intent or a goal to destroy others, but instead an instrumental preference for wanting the resources they're using. For my own designs, one of the things I want is a way to play around with and internalize how complex economies and supply chains work faster than spending ten years working in the manufacturing and logistics field.
I think understanding economics and industrial growth often fails in competitive and cooperative games, or worse, works but gives a needlessly competitive or cooperative view towards economies. It's very relevant to how I negotiate with my boss at a 9-5 standard job that I'm not trying to win by having more money at the end of ten turns, and also that me and the boss don't collectively win by adding our incomes together at the end of the game. That kind of negotiation, where it's positive sum but both of us want to capture as much of the gains as we can, seems pretty niche in non-cohabitive games.
I've mostly tried to avoid having (fake) money as a unit-of-account or medium-of-exchange, mostly because its lack seems to get people thinking about the use of resources like coconuts or water. Once trading is happening, the usefulness of a medium-of-exchange and the limitations imposed by barter seem to become obvious to me, but nobody I've played with spontaneously pointed out a medium-of-exchange would be useful here. (The usefulness I keep seeing is facilitating four way trades with at least four other resources, e.g. coconuts for water for wood planks for cloth.)
Replies from: MakoYass↑ comment by mako yass (MakoYass) · 2023-10-02T20:08:03.047Z · LW(p) · GW(p)
I think the law of comparative advantage keeps being true even if the other agent is better at all the things than you are
Probably, the part that stops being true is the assumption of non-aggression. With humans, damaging a human or even their culture or ego will generally make them far less valuable as a trading partner. But as soon as there are ways for one agent to use the materials of a human to make something more valuable, expect more aggression.
comment by Dagon · 2023-09-29T04:54:27.915Z · LW(p) · GW(p)
I really like this categorization. I'm not sure I fully agree, but it's great to be exploring it.
It would be helpful for me if you could describe what parts of the stock market are positive-sum, and what parts are zero-sum. In one sense, every voluntary trade is (expected to be) beneficial to both - they each think they're better off with the exchanged values. In another sense, every stock trade (with possible exceptions of initial funding or issuances) is zero-sum: the value gained by a buyer of a stock which rises is exactly the value lost by the seller.
I think of such games as "mixed" - there are elements and subgames which are fixed-sum, and others which are variable-sum, and optimizing among them is part of the overall outcome. "Cohabitative", to my ear, is more about shared environment than about the kinds of in-game interaction and optimizations which are being made.
Replies from: Screwtape↑ comment by Screwtape · 2023-09-29T11:30:17.311Z · LW(p) · GW(p)
Example 1. I'm a software developer, and I get paid a lot of money because people want a lot of software. You run a manufacturing company that makes steel, a commodity that's useful whether people want software or not. If I trade you a share of my software company for a share of your steel company, I get something that will be valuable even if later on people decide they don't want software. If you trade me a share of your steel company for a share of my software company, you get something that will be valuable even if later on people decide they don't want steel. Behold, by the power of diversification and the marginal value of money, we are both better off for trading with each other!
Example 2. I would like to become an apple polisher, but sadly I lack both apples and polish. You happen to have capital you would like to turn into even more capital. Thus, initial funding: you buy half the shares of my nascent apple polishing business, I get enough money to start polishing apples, and we both share in the profits. Behold, by the power of the Angel Investment we are both better off for trading with each other!
Example 3. I really, really think Loneay Uskmay companies are really cool and want some so I can talk about them at parties. You have some shares in Loneay Uskmay companies, because you wanted a really diverse pile of shares. While my trading strategy is weird, if we trade I can get to talk about my shares at parties and you can buy two different companies and diversify even further. Behold, by the power of differing preferences we are both better off for trading with each other!
(Warning: I am not a trading professional nor a serious economics teacher, an environment being cohabitive does not preclude the environment taking you for everything you're worth if you let it.)
All of those are, to my view, positive sum.
Example A. We are both high speed traders with a sophisticated idea of value. Bob wanders into the stock market looking to diversify; he's pretty cheerful about paying an extra five cents on the dollar to turn his software shares into steel shares. Me and you are in a zero sum race to get Bob's business. Any profit we seek has to come out of Bob's offered five cents.
Example B. I have realized that the complicated financial instrument on the housing market is, in fact, a terrible instrument and it's value is going to fall from a hundred dollars to one dollar pretty soon, or rather it's value should be one dollar but everyone was wrong about that and had bought it for a hundred. I give you a call and tell you about this wonderful deal I have for you, I'm cutting my own throat really, but since you're an old buddy I'll sell it to you for eighty dollars. That's an intensely zero sum game that's actually trending towards negative sum.
(Econ majors or actual traders, feel free to point out any errors I'm making here.)
Replies from: sinclair-chen↑ comment by Sinclair Chen (sinclair-chen) · 2023-10-19T16:30:20.278Z · LW(p) · GW(p)
Example 4. The management of Generic Corp is missing out on a huge opportunity. So I buy a bunch of stock, use my voting power to advocate for a new set of policies that would churn out nicer widgets at a lower cost. Generic Corp stock goes up, then I sell it at a profit.
comment by Raemon · 2024-12-16T07:18:58.622Z · LW(p) · GW(p)
I appreciated both this and Mako Yass' Cohabitive Games so Far [LW · GW] (I believe Screwtape's post actually introduced the term "cohabitive", which Mako adopted). I think both posts
I have an inkling that cohabitive games may turn out to be important for certain kinds of AI testing and evaluation – can an AI not only win games with rutheless optimization, but also be a semi-collaborative player in an opended context? (This idea is shaped in part by some ideas I got reading about Encultured [LW · GW])
comment by mako yass (MakoYass) · 2023-09-30T08:19:25.541Z · LW(p) · GW(p)
We could also name 'coopetitive': semi-cooperative, but not supportive of negotiation.