The first News Challenge of 2013 will focus on Open Gov and is now open for its Inspiration Phase. To get folks thinking, we asked a handful of people to share their hopes for open government. Below Tim O'Reilly, the founder and CEO of O'Reilly Media, Inc. offers his thoughts.
Edwin Schlossberg once said, "The skill of writing is to create a context in which other people can think." All great movements are wrapped up in big ideas framed in such a way that many people think them at once, and act in concert based on them.
But memes can also mislead. I sometimes wonder if "Open Government" is such a meme. It suggests that the beginning and end of what we need to do is simply to increase the transparency of government and its engagement with citizens, and all will be well. While these are laudable goals, they are far from sufficient to bring government into the 21st century. That's why I've always preferred to frame my government activism around the notion of "Government as a Platform."
Government as a platform provides a new way to think about the great debate of today's politics: is there a way to tackle clear and present problems of society without simply throwing money at them, or having government take them on directly?
Anyone who uses the Internet, or a smartphone, can immediately grasp the power of a platform. Apple didn't write 800,000 iPhone apps, but they did create the opportunity for others to create them. Tim Berners-Lee didn't create Google, Facebook, or Twitter, but the web platform he created made them possible.
So too with government. It creates and maintains roads, but not the destinations to which they go. It ensures that everyone has water, power, communications, education, and increasingly, healthcare, but not what each of us does with those things. It built a space program, launched weather and GPS satellites, but it doesn't do the nightly weather forecast, or give you maps and directions in your phone and in your car.
Open data, as distinct from open government, is a powerful kind of platform. When government opens up data, like weather or GPS data, or the output of the scientific research it funds, society can build on that platform. And it does.
There are also other aspects to the "government as platform" notion that are productive metaphors. The best platforms rely on others for much, but not all, of their innovation. They provide clear rules of the road and protection from bad actors. They provide key system services, so that everyone doesn't have to reduplicate effort to recreate common infrastructure.
But there are many unanswered questions about government as a platform:
- Can interfaces to government be simple, beautiful, and easy to use? 2011 Code for America fellow Scott Silverman asked that question when he applied for the fellowship. If government is to become a 21st century platform, it needs great design. In particular, it needs to focus on user-centered rather than government-centered design.
- How do we measure the result of government programs? Modern platforms are all about feedback loops - measuring what works, and doing more of it; understanding what doesn't work, and doing less of it. As Eric Ries, of Lean Startup fame, notes, a startup is a machine for learning. Can government hold itself to the same standard?
- How is the platform to be regulated? Old-fashioned regulation provides an army of gatekeepers, bureaucrats who give permission and judge quality. Much as in an earlier era, publishers decided which books would see print, and which would remain on the slush pile. But as Clay Shirky presciently observed, the Internet reversed this model, from "curate, then publish," to "publish, then curate." Anyone can publish to the web, but Google's algorithms work to surface the best content; anyone can publish a video on YouTube, but fans decide which ones will be watched. Algorithms that mine collective opinion (pagerank and its equivalents) are the only way to keep up with the flood of content.
- Algorithmic regulation isn't applicable to every situation. I want my school's seismic inspection to be done in advance, not after it falls down in an earthquake. But even in areas like financial fraud, algorithmic regulation is already widespread in the private sector. How else do credit card companies, online advertisers, and ISPs weed out fraud and spam? It certainly should be applicable in less high-risk situations. Can I rely on user ratings to tell me where to stay on AirBnB or which Uber driver to trust? I think so. Bringing regulation into the 21st century is high on my wishlist.
- What is the right balance between citizen participation and thoughtful representative governance? There are legitimate reasons for governance to be done in private. I believe it was James Madison who remarked that the U.S. Constitution could never have been developed out in the open, because of the tradeoffs the representatives of each of the colonies had to make. Internet platforms again give some guidance. They are not an anarchy, they have a carefully designed architecture of participation. How can we redesign representative democracy for the 21st century, so that it is more participatory, without losing the benefits of thoughtful deliberation and forceful leadership?
While "Government as a Platform" might seem to be less focused on the goals of the Knight News Challenge than "Open Government," I believe it provides fertile ground for investigation, thought, and the development of new systems that harness the power of the public to make of our government "a more perfect union" and a more powerful platform for the flourishing of our society.
Related: "News Challenge launches with an OpenIDEO twist" by Chris Barr, "Back to the future of self-governance: the promise of the Open Gov movement" by Mark Meckler and "The end of the beginning, lessons from open government so far" by Anil Dash and “First steps to Open Gov – getting your ducts in a row” by Susan Crawford