Interview with David Eaves on Open Government

The following are excerpts from an interesting interview with Canadian open government advocate David Eaves, by the Public Policy & Governance Review, (Vol. 1, No. 1, Fall 2009).


The public service is facing a generation shift, and 2.0 technology is bedevilling the government. What do we do? We asked David Eaves, a public policy entrepreneur, open government activist and negotiation expert. David advises the Mayor of Vancouver on open government, works with two spin-offs of the Harvard Negotiation Project and serves as a fellow at the Centre for the Study of Democracy at Queen’s University. He is frequently updating his blog at

Public Policy and Governance Review: You talk about government needing a less hierarchical structure, at least at the lower levels. How do you suggest people start thinking about more networked or less hierarchical modes of government?

David Eaves: This is already all going on. A lot of the way things are changing actually are consistent with the way they’re already behaving. Public servants spend a lot of their time calling around trying to get in touch with the other public servants they need to talk to, and a networked public service is about making that easier. The bigger challenge is around what accountability and authority look like. It’s not that there aren’t models out there. The model I frequently turn to is the Mozilla model, where you have people who own sections of the software code, who own pieces of the product. It’s actually quite an authoritarian model. People think that open source is democratic, it’s actually not, it’s very undemocratic. It’s highly meritocratic. If you own a piece, you really decide what goes into that piece and what doesn’t. There are a lot of people trying to influence you, and the key is, you have to respect people who are influencing you, because the more people who are giving you advice, the better the product’s going to be, the better you look. You can’t just ignore them, you have to constantly be engaging them. There’s an accountability to people who are participating in your community that’s very, very direct.

PPGR: You’ve said there’s a cognitive dissonance in networked government, that while people may be already behaving in this way, they are still connecting it in their minds to the formal hierarchy. How do you address that?

DE: There’s a top-down source of change and a bottom-up source of change. I think the bottom-up change is pushing as hard as it can. There’s a lot of cool stuff happening, but it’s kind of under the radar. The top-down is also necessary and ultimately required to shift. That’s what you see at NRCan, where the deputy minister puts his briefing notes on a wiki and says anybody can edit them. They create the safe space that allows people to come and participate. For a lot of public servants that still feels risky, even though it has the blessing. But it still has an insurgency feel to it.

For me, there are two reasons the insurgency stuff is good. Number one, proof of concept. It shows this stuff works. The second is, it allows an organization that’s pretty averse to risk to take some risk and do new stuff, and it creates demand on the leadership to do something.

PPGR: Everyone seems to be worried about the extra amounts of data created by new technology. Given that government is a slower mechanism, how can it adapt to the social and cultural overload of this data?

DE: David Weinberger talks about how just before the advent of the internet, everyone was like “Oh my god, we’re going to get overloaded with information, we’re going to get swamped and lost in this sea of data, and no one’s ever going to figure out their way and we’re all going to die.” But that didn’t happen. And the reason he says that didn’t happen is because we discovered that the way to deal with lots of information is to create still more information. You just create information about the information. So things like tags, and filters. Google is a massive filter and it filters the information to give you what you think you need. Blogs are great because they filter. People read my blog because it filters information about public policy.

The Government of Canada has too much information. And what it’s doing is actively fighting against the most effective filters for information. It has crappy search engines, it doesn’t allow people to use social media, it doesn’t allow blogging. Those are the filters that people actually need in order to figure out what’s going on. The reason I like wikis and blogs in government is that it can improve people’s capacity to grasp the incredible amount of information that’s being thrown at them.

PPGR: How do you get buy-in for social media, wikis and blogs in government? What conditions do you need to foster these things?

DE: There’s a couple of things. The first thing is, don’t reinvent the wheel. I told the government, “hire Facebook.” They’re used an open source white book kind of networking platform, which I hope will work, it’s certainly a lot cheaper. But the thing I like about the feds is that they’re using open source software so it’s all free, the cost is incredibly low. So let’s say it fails. It was a year of one employee’s time, which is a lot for that person, but for the public service, it’s pretty nil. The cost of failure is low, whereas if you invest a lot of money in a Facebook-like application and it fails, it cost you $30 million, that’s a lot of money. So really, really low costs. It should be cheap, because it’s out there. Failure’s okay, and the reason failure’s okay is because the costs of implementation are low.

The second thing is, you’ve got to design a system that actually meets a need. The Facebook we live and use today was literally created to allow geeky guys at Harvard to connect or identify with who are the hot girls their friends knew. That was the initial use case. It was literally about “Can I see someone’s face, or a photo and get some basic information about them?” And then the other uses were created because of demand. Not because they said, this is what this should be used for. It became about “Can you create this new feature?” And that’s what’s completely lacking in the government. It’s more like scope out the entire project, figure out what the use case is, and then go build it. It’s the opposite of how an emergent system works, which is to find out what a need is, meet is, and then build it to satisfy the emerging needs that arise out of that.

PPGR: In terms of public buy-in, say, to engage people in consultation, what’s to say “if you build it they will come” when it comes to these media?

DE: It depends who you mean by “they.” If you expect a million Canadians to show up, I think you’re asking for failure. The whole point of the long tail is that there’s five people out there that you need to be talking to. [Ed. The long tail is a statistical concept about distribution in a marketplace, such that products that are in low demand or have low sales volume – generally about 20 percent of the distribution – can collectively make up a market share that rivals or exceeds the relatively few current bestsellers and blockbusters, if the store or distribution channel is large enough.] There’s another 100 who are interested, and may have something okay to say, and another 100 who are maybe on the wrong side or talking about the wrong thing. The long tail isn’t about large numbers, it’s about aggregating the right numbers, and finding the five people out there who have something interesting to contribute. Sometimes there are a lot of people who have something to contribute, but you would need a very specific piece of information from them. I find these crowdsourcing applications very cool, where anybody in the city can identify where a pothole is. Anybody can do that, but it’s a very simple ask.

One of the things I’ve been thinking about is, where the app is for doctors, where as they treat patients, every time there’s an adverse reaction to a drug, they open up an iPhone app and say “this drug, this adverse reaction.” And then right away, the federal government has aggregated data from doctors across the country about adverse reactions to drugs. If a new drug got on the market and it has an adverse reaction with another drug – which is very hard to predict – now we have instant data that suggests that it might be a risk. It might not be, but we have data that says we should deploy some resources to investigate that. You have that data instantaneously then. We’re conducting trials three years after the fact or something stupid.

PPGR: There’s the idea that we have a democratic deficit. Do we need to do some repair to our democratic institutions to facilitate buy-in, or are we replacing that with all of this participatory media?

DE: I’m not totally sure I buy into the whole democratic deficit stuff. I think there’s a conversation deficit, and I think newspapers are in many ways to blame. They’ve dumbed the conversation down and they’ve made it simple, so people aren’t interested. I am concerned about the democratic deficit, but not in the way most people are. I’m actually concerned about the role of politicians. We’ve denuded committees, basically your individual MP doesn’t have any authority or any influence. Not that they ever had a ton, but they have no budget to do research, they don’t have money to think independently, they don’t have any committee powers to investigate anything that’s going on. We don’t ask anything of them. All the oversights have been outsourced to the Auditor General, who is now making value judgments. We’ve destroyed any role for the individual MP, and I actually think TV is the cause for this.

I don’t mind if the government is crowdsourcing problems. What I want is a public service and politicians who can set overall strategy and can lay down what the assumptions are and what the priorities are.

Leave A Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.