Imitation Is the Sincerest Form of Failure

“What did they not do? They didn’t go back and say ‘how does this feature fit in with everything else that we have?’.”

Aral Balkan

Open is a wonderful thing. I think everyone can agree on that. It’s nice to have control over our own data, our own tools. It’s nice to own the things we rely on. But today, most the tools we use aren’t open. We follow people on social networks using internal tools, not the standard created for that precise purpose: RSS or Atom. We use proprietary follow buttons, not my beloved SubToMe.

A lot of open advocates will claim this is because people are greedy, and they want to build tools that will cement their position of power. Like they’re building a pyramid of control to elevate themselves, and each user that uses their proprietary solution is another block in the pyramid, another person that they ultimately have power over.

That may be true in some cases, but today I’m going commit the ultimate blasphemy for an open advocate: I’m going to remind you that proprietary solutions actually have some really great benefits to them. Because recognising the strengths of proprietary solutions is necessary if we want open solutions to succeed.

The Case for Proprietary Solutions

There are objective, reasonable benefits to making solutions proprietary. For example, early technologies are still defining themselves. That stage of product development is better handled by proprietary technologies, which are by nature nimbler and easier to change. More experimentation is allowed while figuring out what the solution should actually look like. A standard is supposed to define something that already exists; it’s not the right tool for the job of figuring out what something is.

But a lot of things that are already defined—blogging/micro-blogging, code storage, social networks—are still created afresh today without standards. The reason for this is, a lot of the time, because it’s easier to build the desired experience that way.

Imagine if each of your friends had to set up their own Facebook instance to have a profile. Or their own Twitter server to be able to write (or, worse, read!) tweets. Do you think we’d see the adoption we do today? Of course not. Even though standards for those things are well-defined, the cost of participation is still too high.

But why can’t they just intoperate? Because keeping the data in one central place has definite advantages. It’s easier for Facebook to show you how many likes a status has. It’s easier for Twitter to surface suggested users. While these things are possible in open solutions, they’re much more involved to implement.

The Open Solutions of the Future

That being said, I do believe open solutions are the future. I do believe that we should have distributed social networking. I do believe we should own our own data.

The distributed social networks we see today, however, are taking the wrong approach. They try to match Facebook or Twitter, feature for feature. They play the game by the rules laid out by centralised social networks, which of course play to the strengths of centralised social networks. But open solutions have strengths, too, and not just for the technically savvy. New categories of tools, new interactions, new possibilities are opened by this different paradigm of computing, and these distributed social networks need to play to these strengths.

A revolutionary social network will not be created by looking at what is already there and trying to reproduce it with different priorities. The distributed social network that finally gains widespread use will look at the problem with fresh eyes, and try to create an experience the centralised social networks cannot possibly match.

I think the decentralised social network of the future will look less like Facebook, a silo everything happens within. It won’t have the equivalent of a “Facebook me”. People will just say “check me out online”. It will be more about supplying your own information, having the ability to post your own statuses and events and such and get replies to them, but it won’t feel like “posting to” a website, it will feel like “broadcasting”. I think the spread of information will draw nearer to its conceptual base, without the intermediary of a specific website’s terms; the conceptual base, after all, is what the standard is derived from. The individual experience is tailored by the tools a user chooses to use. I also see the social network of the future being more all-encompassing; you don’t have Twitter and Tumblr and Facebook and Foursquare, you have the internet, and you broadcast information there.


This kind of open, decentralised social network would fail today.

It requires too much maturity in infrastructure that we simply don’t have. We need the average person to be able to set up and install this software on their own server—and keep it running—just as easily as they set up and install the software on their personal computers.

That is the real power of centralised social networking. It makes things simple for end users, who will trade that convenience for control. And this is precisely what the open community needs to be working on: the infrastructure that brings the level of convenience low enough that non-hobbyists can run it.

We need server architectures that can be administered by end users. My boyfriend’s not going to be on call for his server falling over in the middle of the night, he has a job and a life and things to do. My little brother isn’t going to fire up a terminal to troubleshoot a server. We need to bring our software deployment, ops, and infrastructure management tools up to the level of our desktop software installation tools.

This is hard. But it’s worth doing. There continues to be, even in my own generation, a huge gap between the technically-savvy and those that don’t tell computers what to do for a living. While there will always be a gap—we are experts, and will always be needed—it should be a lot narrower than it actually is. Part of our problem is that we’re acting like impatient teachers; we just take the keyboard from the users, and do it ourselves. Fine, we’ll manage your servers for you. Fine, we’ll ensure your data stays safe.

“Treat your audience like poets and geniuses and they’ll have the chance to become them.”

Del Close said that. If we treat our users like they’re capable of owning and operating their own technology, maybe they’ll finally believe they can.