Dec 2, 2025

Nostr as Game

All systems can be approached like a game by "probing" for affordances. This can be a lot of fun, and works well in an adversarial environment.

Open systems are fundamentally different from closed systems. One important way that this is true is in how rules for how the system operates are established and enforced.

In closed systems, rules are set in stone, and can't be changed by anyone except the owners of the system. In open systems, rules are established based exclusively on what Steven Johnson calls the "physics" of the system - the internal organizing principles that distinguish the system from other systems, and which give it meaning and utility.

This latter type of rule also exists in all but the most rigid systems, but are counter-balanced by another type of rule: rules can be changed capriciously, and enforced arbitrarily. When participating in a system run by someone else, it's impossible to predict from one moment to the next what the consequences of your actions are. This is why fiat currency creates so much chaos, and why civil law necessarily suppresses the rights and responsibilities recognized by common law.

But I want to draw a parallel here to another type of system: games.

Games fall into one of our two categories of sytems: games which are a mutual fiction shared between two players (an open system), or a game which requires a platform in order to play (most video games).

An example of the former is that of chess. The rules are well-known and inflexible, leaving very little room for arbitrary enforcement. And yet, there is infinite room for creativity, both within the boundaries of the game, and through the modification of the rules to create something entirely new.

The voluntary modification of the rules of a game is one of the affordances that exists within an open system. Changing the rules changes the game into something somewhat but not entirely different. When is chess no longer chess? When it's 3D? 4D? 5D? Every compounding dimension will still bear some resemblance to the normative version of the game.

This modification of rules by the players of the game is distinct from arbitrary modification of rules by the "owners" of a game. Take Runescape for an example: Jagex is well-known for selective enforcement of additional rules, and modification of game mechanics in order to engineer a more perfect environment for players (and their business model).

And yet no matter how heavy-handed a system administrator's oversight might be, there are always affordances that users can take advantage of.

In "Everything Bad Is Good For You", Steven Johnson identifies the behavior of discovering and exploiting these affordances as "probing":

Probing often takes the form of seeking out the limits of the simulation, the points at which the illusion of reality breaks down, and you can sense that's all just a bunch of algorithms behind the curtain.

As a paradigmatic example, he explains that PacMan

had its rules, which were so simple you could express them in three sentences... But experienced PacMan players soon discovered that the monsters roamed the maze in predictable ways, and if you followed a certain course—literally called a "pattern"—you'd complete the level without losing a man every time you played. Patterns weren't build into the official rules of the game; they were a legacy effect of the limited computational power of those arcade machines, and the predictable way in which the monsters' behavior had been programmed. To detect those limitations, you had to probe the PacMan game by playing it hundreds of times, experimenting with different strategies until one sequence revealed itself.

The abstraction of PacMan is different from its implementation. The idea of the game was simple, but the implementation had enough incidental complexity that was accessible in subtle ways to the user that they were able to exploit it in creative ways.

Another great example of this is the many ways to run DOOM, including on a pregnancy test. The designers of the pregnancy test did not intend for this to be possible, and yet here we are.

Nostr, like all systems, is also a game. It has certain rules that are more or less set in stone (using the secp256k1 curve for example), but which can be remixed to create new, subtly different sytems. Within the boundaries of nostr (however you might choose to define those), the ways its various features can be exploited are endless.

In some ways, this is less fun than attempting to circumvent rules put in place by omnipotent administrators. At the same time, we have endless affordances which are possible by the reconfiguration and remixing of different features.

Want to efficiently sync events between relays? Invent a sync protocol. Want to go back to trusting servers and get deniability? Strip event signatures. Want to manage relays? Add an API. Don't like kind 1 replies? Make kind 1111 "comments". Want to hijack kind 1 for your own purposes? Auto-repost your new events to kind 1. Want to crash clients? Make a relay that serves non-string tags. Want to play chess? Do it.

Every NIP is a possibility for how the basic building blocks of nostr can be reconfigured. And because nostr is an open system, anything goes. Everyone on nostr has voluntarily chosen to enter into an adversarial environment where the only rules that exist are the ones the users bring to the table. This makes all of us "hackers", by virtue of the code we write, or the software we choose to run.

You have the power to create new data types, overload old ones, add an API to your relay that no one else supports (or lobby for adoption), open a PR to your favorite client or fork it, protect your users' privacy or expose it, experiment with new UI patterns or re-implement the old ones, do good or evil, nurture the protocol or trash it.

This is not to say our choices don't have consequences. There are truly bad ideas, and attacks are possible in so far as the utility and logic of the system that has emerged so far can be subverted by users or software implementations. But there's no need to be completely self-serious. Yes, we need some apps that have actual business models, and the discipline to execute on them. But the culture of nostr is first and foremost one of experiementation in a spirit of fun and creativity.

So go play!