Prisoner's Dilemma on Peercoin

Iterated prisoner’s dilemma is an unsolvable game theoretic rule set. Individuals could encode binary values for a large number of moves in the OP_RETURN. The transaction ID can be used to generate a random entrant ID for verifiable randomization of competitions. Winners could get 2 PeerAssets from a central recognized host. One would be a once, single, unspendable token as the ‘badge’ for that win. The second could be given out to all participants as the total end score, unspendable, that tallies their total score over all the games they participate in. The awards are just for bragging rights.

An interesting take on the game theory is in trying to find what games are easy to do this way on Peercoin. To keep down txn costs, we want games held as events where any number of people can play. To do that, we basically have to allow knowledge of previous plays, unless we do something funky with crypto. So the submission window starts, and as people submit answers they are able to see the pool of previously submitted answers. This may impact their decision when creating the output, but because they don’t know how many players there will be by the end, it has limited impact, and could turn against them depending on later submissions. If you submit at the last possible moment, you have the most knowledge but you risk having your submission not make it into the final block. This creates an interesting kind of game of chicken there as well, to prevent waiting too long.

Anyway, this was fun to think about, I’m not necessarily saying we should do it, but it’s a cool idea in my opinion.

1 Like

Role-playing games over P2TH is possible. One thing that makes null-games difficult is the requirement of using OP_RETURN. If we simply send outputs to a few addresses, we can make decisions in-game. Off-chain, say in this very forum, a game master (GM) could announce a game along with a few addresses (or even their private keys) and maybe an intro. Those addresses represent different choices in the game, starting with which character are you, e.g. half-orc barbarian, elven wizard, halfling rogue, or dwarvish cleric. This first round announces you are participating using the first input address to that output.

Next, the GM makes the ‘heading out’ post for the game on the same thread, along with a few more addresses. The dozen or so characters that accepted the call band together in a haphazard unit to fight the cyclops that has been terrorizing a small village, each coming from their own background. They gather information about town but are only able to discern that the monster is to the east, where a rolling portion of foothills curls up against a thick forest. As you stand at the edge of town looking down the forked road to either terrain, the spring wind blows across your face and you consider which way you should go. [Now you submit to an address for forest vrs foothills].

Now here’s the interesting bit: you can use the txid for a random number generator seed to roll dice with, or as a single die roll. You can also have the same addresses mean different things for different characters and track the responses to each question to establish a variety of paths. Much of this can be planned aforehand, but it would be best if some of it could be off-the-cuff, so that the story really responds to the choices and rolls the players make. An experienced GM should already know how to work that though.

To keep games manageable as far as on-the-fly stuff is concerned, an entry amount could be added to the first message without too much added complexity. This can be seen as an extra anti-ddos measure, though the txn fee is already a non-zero cost. Also, given that time needs to be taken to read the latest installation and send in the vote, the frequency of rounds should be limited, say once a day or once a week. If the GM chooses to run an entirely premade session, then the votes could potentially be done more rapidly if they are actively on the forum.

Why do this instead of just responding with your choices on the forum, or countless other ways to do this kind of thing? The answer is that special subset of people you get when you require them to send a crypto transaction to participate. It’s anonymous, but with provable identities. It requires very little but the writer with a story to spin one up. Participation is as easy as sending a txn and you will be in good company enjoying a game. Nothing stops you from responding, commenting, sharing a discord between party members, or anything else, in fact it makes for some really interesting possibilities, like guilds and raids and GMs that play off this concept. And finally, it has the ability to scale-up to large numbers of players, while still maintaining a minimum bar to entry

The barrier to entry as a GM is low, just post a story line with a few addresses you make just for this game that do character creation. Depending on who responds and how many, you can make it up from there. To keep meaningless participation down, you can use a submission filter.

All I’m saying is, if you want to fight the cyclops and you think it’s in the forest, send 1 PPC to:
If you think it’s in the foot hills, send 1 PPC to:

Please don’t submit anything other than 1 PPC outputs to either address, note that I am using the first input address for each transaction as the identity.