thejaymo

---

The Rise of Co-Play: The Future of Multiplayer Software

A quick note: Issue 6 of my zine, start select reset went out to paid supporters of the show earlier this week. Get yours at thejaymo.net/support. Anyways, on with the show.

Matt Webb recently posted an update about what’s been keeping him busy. His long time interest in what he calls ‘multiplayer software’ has led him to becoming ‘inventor in residence’ at PartyKit.io. An open source software tool that makes real-time collaborative software applications easy. Congratulations Matt!

The slow roll out of multiplayer software over the last decade or so has been unevenly distributed. The onscreen cursors in Figma or Google Docs are only now just catching up to Engelbart’s vision for personal computing back in 1968.

Real-time multiplayer software is a techno-social system, and therefore a world with its own digital physics. Who doesn’t think Figma has a kind of game feel?

All worlds too develop cultural mores about how its inhabitants should behave in shared code space. For example: one team I often work with park their cursors in a little box when someone is presenting in Figma.

The metaphor of multiplayer software allows for easy deployment of a related term:

Game modes.

To change the mode of the game is to apply a distinct configuration that varies gameplay and affects how game mechanics behave. The two most common game modes being: cooperative and competitive. 

I would like to propose that effort be put toward the exploration of a third kind of game mode: Coordinative.

Or as I call it: Co-Play

In the spring of 2021 The Moving Castles team, best known now for their development work on Autonomous Worlds, were prototyping a livestreamer VTuber rig. When done, they then did something amazing. They hooked up Twitch chat directly into Unreal Engine via API. Experiments followed. By typing a command in chat one could appear as a butterfly floating around the world of the Avatar. Another command could change the camera to follow my butterfly labelled with my username around. In another experiment, a chat participant would be granted permission to spin a wheel of fortune by typing SPIN WHEEL. Soon collective experiments emerged. Self selecting on to teams and spawning objects in the game world via frenetic chat spam.

The boundary between game world, player and observer had been breached.

So what is Co-Play? 

Imagine for a moment a near future where most games have modes built for livestreamers. I don’t think this is a particularly hard thing to imagine. People spent 1.7 billion hours watching other people play games on Twitch last month alone. 

So we have a livestreamer playing a game in real-time. Talking to and responding to chat which is wired directly into the game world. Chat is playing the game collectively with the streamer. Think Twitch Plays Pokemon With … thejaymo. Chat commands, emoji sentiment, voting and other mechanics yet to be developed are all used to control elements in the game world. Acting cooperatively with, or competitively against, the streamer. 

I call this ‘coordinative play’ as the participants in chat need to coordinate in real time in order to influence the game world. Twitch Plays Pokemon way back when had two coordination mechanisms: Anarchy and Democracy. With the inclusion of a human in the loop, the role of the player may become more like a channeller or conduit for the asymmetric intelligence they have access to.

There is so much to be explored here. What coordinative decision making mechanisms for swarms can we prototype, experiment and develop? Because I regret to inform you, for better or worse, the future of societal coordination and maybe even politics will emerge from gamer culture.

To explain why, let us for a moment turn our attention to the real time chat UI paradigm. It’s one we find everywhere in our daily lives. Family & friends group chats on Whatsapp, Community Discord channels to Slack or Microsoft Teams. DAOs – basically group chats with bank accounts – need better coordination tools not better governance too. 

The question posed by the potential in coordinative play is a simple one. How do you get a chat room containing 10,000 people to come to a decision or reach consensus in real time?

What sort of real time UX can we create? Does everyone even need to have a vote? Is voting even the right way of thinking about swarm alignment? Can we do real time liquid democracy in a chat room? How about teams or clans? Does everyone in the chat need to agree and play by the same rules? Can you have multiple co-play modes to reach consensus within the same channel?

I have many questions and no answers. 

Think about the mechanics we currently have at our disposal across our day to day virtual lives. Maybe a poll, emoji voting, maybe a doodle calendar? These are fairly blunt instruments and also require voting periods. They don’t dynamically capture or represent group alignment. 

The first step is to prototype all this stuff in video games. New rules, mechanisms and logic’s will be battle tested in virtual worlds without consequence for serious use in the real world.

When these new  UX mechanisms are ready, they will flood out from the gaming world, and into the workplace, into our daily lives.

Watch