Exponential growth devours and corrupts
DHH
2.5K177

How can we turn more of the Twitters and Facebooks and Googles into generics? What shifts in underlying technology and cost do we need to hit to make it feasible to run something like Twitter on Wikipedia’s budget (and fund it by donations rather than ads)? What if the next Big Idea looked more like email and less like the walled gardens of today?

Exactly! As you probably know by now, we are building exactly this at Qbix. An open source social platform that allows anyone to run their own social network, anyone to develop apps for communities to install, same as Wordpress does for blogs.

What we need isn’t a new technology to handle petabytes of data a day in some datacenter which will be described on highscalability.com tomorrow. We need to go back to the basics of how software — yes including email but also IRC, the Web and many others — worked before the “always on broadband” became the norm in rich countries.

I would go one step further in your thesis and say that centralized design is at the root of all the perverse incentives and power imbalances you see. The VC model just leads to it. “Capture the market” says Peter Thiel, and inevitably that meant “make all the roads lead through Rome/Vegas”. Where Facebook offers Internet.org to India to get another billion people on its platform. And then we wonder why we live in a feudal digital society.

Whether it’s Apple’s app stores or Facebook and Twitter’s platform, or Google, all our bits and bytes flow through these providers on their way to the person right in front of you.

I want a future where connections via mesh networking or local wifi are tried first before the global internet and cellphone providers. Where college students in a class can collaborate without Google Docs. Where apps can be installed by communities without permission of the feudal overlords and people in a remote village can plan dinner and book an appointment without seeing an ad from a server in California.

All that needs to happen is for there to be open source platforms for building decentralized, web based social software.

Open source serves the long tail far better than closed source silos. So back to your analogy with drug patents… if we didn’t have these patents, new drugs and technologies could be invented by building top of current knowledge if people could build on each other’s work. Patents hinder the long tail, in drugs as well as in software.

In the current climate, abolishing either software patents or drugs is not very likely. So the next best thing is patent-left (similar to copyleft for free software), or failing that, a compensation model like the one I outlined on qbix.com/blog .

Yes, VCs are one of the problems. But even without VCs, even companies that turn profits from day 1 choose to build centralized network topologies with all roads leading through them. That is the real cause of the feudalism, the perverse power imbalances and the lack of innovation in the long tail.

Show your support

Clapping shows how much you appreciated Gregory Magarshak’s story.