Introducing Federation

Wiki is a relentless consensus engine. That’s useful. But here’s the thing. You want the consensus engine, eventually. But you don’t want it at first. It’s funny, I was looking over this keynote last night, and I saw this line and realized — this is the simplest explanation of federated wiki. You want the consensus engine, eventually. But you don’t want it at first. This is a problem. It’s pretty easy to build a system that starts with consensus and then fragments into personal opinion and individual statements. In fact, we build a lot of systems like this accidentally. Entropy can be added rather easily to any system. It’s harder to build something that starts fragmented and personal, but then organically becomes a shared communal space. You’re looking for a system that produces what Polanyi called “spontaneous order”.

So back in February I saw a video of Ward Cunningham, the initial inventor of both wiki and really of wiki culture — he was the first wiki admin as well, and many of the conventions of wiki come out of his unique take on how collaboration fails and how it succeeds. And what he says in this presentation is we end up being pushed to consensus in these systems because we’ve got the technology upside down. Here’s a slide to show what he means by that.

2014-11-06_1659

2014-11-06_1659

In a traditional wiki, you have multiple people sharing a single server, and the server is the ultimate arbiter of what’s on the wiki. In a federated wiki, everyone has their own server which stores the records associated with them. But the meaning is made in your browser. Your browser pulls wiki records from all over the internet, and makes them look like they exist on a single server. If you have a background in network theory, I think you’ll see immediately at how this inversion creates a sort of evolutionary ecosystem where we reach consensus not through arguing who gets to control a page on a specific server, but by seeing which versions of a page spread to where. If you don’t have a background in network theory, I want you to forget this slide, and just watch what I’m going to do next. In my entire time talking about this I’ve found almost no one gets the massiveness of the architectural shift at first — I don’t want you to understand it, I just want you to know it’s there. Ultimately, the theory is that this thing here is the “spontaneous order” engine that can move from the fragmented to the unified. Either that hypothesis is right or wrong. My job here is not to argue a side of that hypothesis — it’s to convince you this is an hypothesis worth testing. Again, if you don’t completely get this idea, don’t worry. The view from the user side is much simpler.