Debugging Democracy
What is it we’re doing here, people?
Having recently started working as a researcher for mySociety, I’m spending a lot of time thinking about civic tech. My new job has me researching the impact of government-led civic tech projects in US cities. I’m lucky enough to have the chance to choose a set of civic tech tools from across the country and then look at the effects these tools have on users, services, and the cities at large.
With this project, yet again, I find myself asking: So what is “civic tech”?
It’s surprisingly hard to find someone with a quick answer to this question.
Why is that?
Because the field hasn’t yet settled on a canonical definition. There are substantial differences in the way civic tech gets defined.
At one end is the the Knight Foundation, which in an influential report declared that the category included everything “at the nexus of technology, civic innovation, open government and resident engagement.”
Civil society groups tend to define civic tech as technology which raises the power of citizens relative to government. For example, a Code for America brigade “[uses] technology to empower citizens and improve government operations,” and mySociety builds “digital tools that enable citizens to exert power over institutions and decision makers.”
Meanwhile, commercial perspectives often describe civic tech in terms of a service that government provides to citizens. Accela, for example, states that, “Civic tech aims to solve the problems facing citizens. With technology we can help government agencies engage their communities in new ways and modernize the way they serve citizens.”
To some degree, the breadth of current definitions reflect the fact that the sector mixes players with different structures and motivations. There’s just some basic differences between the perspectives of large foundations, idealism-driven volunteers and non-profits, and commercial organizations. With these different perspectives come different definitions. Another reason for the variation is the age of the term. It was not even three years ago (in the pre-Medium age!) that Tom Steinberg, Nathaniel Heller, Tiago Peixoto, and Micah Sifry were actively discussing the “naming deficit/surplus” for the field that would quite soon be known as civic tech.
Now that the “civic tech” label has achieved mass adoption, it’s attracting more theory. Sifry sparked another round of definition-crafting this past summer with the launch of Civicist, and in its first week posted more precise definitions for civic tech from Matt Stempeck, An Xiao Mina, Laurenellen McCann!!, Mark Headd, Hollie Russon Gilman and David Karpf. I’ve seen even more recent definitions from Lawrence Grodeska and Bill Bushey.
This most recent theorizing (and this very piece, in fact) stem from a desire to make the meaning of civic tech less vague. Vagueness isn’t all bad, of course. Vagueness create a big tent that covers a variety of otherwise unaffiliated people. A big tent makes a big light, and attracts more people who might not have found each other otherwise.
However, the vagueness also makes it hard to see if individual civic tech projects — or the sector at large, even — is actually achieving a cohesive set of aims in a meaningful way. This vagueness ultimately leads some actors to question whether “civic tech,” is really useful as an umbrella term. Without a coherent meaning, “civic tech” runs the risk of become an intermediate stage on the way to something that feels more concrete, and more essential.
Happily, in the field we are building a clear definition of civic tech. It has to do with the way we believe we improve democratic self-management. We want to improve everyone’s ability to transmit information about their preferences into government. We want to improve everyone’s ability to get the information they want out of government. And we want to see what happens as information is processed in the middle.
Civic technology seeks to improve government input/output, while opening the process to full view.
Agreeing on the Problems
Democracy, in theory, is a simple function, requiring only adding or averaging. Democracy is a project where for total number of citizens(x), x number of inputs are channeled into a process which outputs a majority result. Representation and delegation are subfunctions, put into place because we’ve decided it’s impractical to get inputs from everyone on everything.
Civic tech is the world where programmers think about democracy and try to debug the problems that cropped up between the simple input-output theory and what we ended up with.
What are the big bugs they’re trying to fix?
In different contexts, I hear the following messages:
Government is slow; I want it to be quick.
Government is captured; I want it to serve people equally.
Government provides services badly; I want it to work effectively.
Government is secretive and complicated; I want to see it clearly.
So what do we expect from democratic government? It should be quickly responsive, serve people equally, use the best-quality information to make the best possible decisions, and make these decisions out in the open.
This dream is familiar — it reflects many previous utopian visions of government. In civic technology’s particular vision, though, digital systems play a key role. The reform goals are achieved through adopting software which perpetually guides users, both inside and outside of government, to the most system-improving information.
Like many other utopian visions — and stemming from its roots in the anarchistic Internet — civic technologists envision a flattening of power across the social organism. It is unique to civic technology’s case, though, that this vision of social flattening is somewhat practicable. With digital technology, transaction costs for transmitting information approach zero. Interactions are no longer bound to face-to-face meetings, or even circulating paper. An individual’s preference can be sent and received instantly, collated automatically. If technology makes it possible to collect and include everyone’s preference, why shouldn’t we do that? Representation, once essential for large-scale democracies, is no longer indisputably necessary.
Civic technologists’ interest in flattened power, plus their recognition of technological possibility, also increases their appreciation for transparency. Public transparency, to civic technologists, is a one-to-many form of communication that speeds both present and future learning. For democratic reformers in general, transparency is important because of the way it provides real-time information to improve real-time input. You must know when the important hearing will occur to get there in time to testify. For civic technologists, however, transparency is also important because of the way it provides a repository of searchable material — for things you might not know you need now, but very well might need later. Like GitHub lets you fork code, or like the internet lets you quickly access specific answers to questions, so does it seem logical that our social, public decisionmaking functions and products be easily available. Do government in the open, so we can do it too.
In a sense, the dream of civic technology envisions a central role for government as an value-neutral aggregator, disseminator and processor of information. Government as I/O.
Agreeing on the Solutions
Although civic technologists are reformers, they are generally positive towards the actual people working in government. They are oriented to changing structures, not people. If our real problem is a lack of appropriate technology, then the problem doesn’t really lie in the bad will — or even incompetence — of individuals. So when I say “government is slow,” I don’t mean that individual workers are slow; instead, I mean that the way that decisions are made and implemented is slow because of existing decision-making systems. When I observe that government is captured by a small group, or government is secretive, these aren’t problems of individuals — they are frequent experiences across governments. They are the product of the way that government works when it doesn’t use technology the way I can imagine it being used.
So how does civic technology help us transform democratic government?
The first problem — government is too slow — is perhaps our least controversial problem, and is therefore the easiest to approach. A number of civic technology initiatives operate on the principle that you can make government work more quickly if you automate more processes. Capture information digitally, store it in structured formats that allow it to be reused and save the effort of entering it multiple times. Provide that information in ways that make it easy for people inside and outside government to notice and consume it. Making this area of civic technology work even more attractive to cities, providing the same service more quickly translates into efficiency; providing more services while holding resource expenditure constant does as well. In other words, the provision of better online forms continues to have a bright future in civic technology.
The next set of problems —improving the equality of representation and service delivery — are more complicated, since this isn’t about “doing more with less” but changing the existing distribution of resources. Traditionally, these distributions are determined by politics. Civic technologists tend to be wary of politics. Unfortunately, if politics is the process by which we decide “who gets what, when, and how,” then it is a political act in itself to use software to change the amount of attention government pays to people who aren’t currently visible. It is a political act to identify ineffective services , because invariably someone out there is benefiting from them. Oxen will be gored, people will object, and so civic technology initiatives in this area only occur where a government’s political goals are in alignment.
Nonetheless, the goal of enlarging the group of people who participate in government decisions remains quite central for civic technology. The civic tech fix associated with this goal is to improve avenues for public input into government.
- In the UK, mySociety’s FixMyStreet is one tool that improves the way we contribute information to government. It’s a platform which allows anyone to lodge an online complaint about a public issue, whether that problem is a broken sidewalk or excessive dog poop. It allows users to track progress on their complaints, receiving updates as government officials decide what to do about them.
- SeeClickFix provides a similar service in the US, increasing the volume and transparency of local service complaints.
In a way, these are easy cases. The input that’s being received through both of these tools is politically uncontroversial. Moreover, studies have shown that the people who are using them, at present, are likely to be people who already have the resources to provide their input to government.
Nonetheless, even among these often-privileged users, using the platform and getting a positive response increases users’ willingness to provide even more input. An easy and effective method of transmitting opinionated information to government creates a virtuous cycle of greater participation. And as the flow of two-way information improves, so will the overall quality of government services.
There’s no real reason why these platforms couldn’t be used to collect issues, concerns or complaints relevant to any other kind of public service, from education to public health to law enforcement. Civic technology aims to transform government through improving methods of getting public input into government, and this is one concrete path.
The flip side of input is output. Civic technology tools which help pull useful information out of government and transmit it effectively to the public play another critical role.
To date, civic technologists have provided platforms which surface government information in two main ways: through public records request portals, which smooth the process of getting public records out of governments, and through open data portals, which provide access to regularly updated information.
Public records requests are an essential method of pulling exactly the information that people want out of government.
- In the UK, mySociety’s WhatDoTheyKnow provides a centralized location that allows any UK resident to request and receive information from any level of British government. Through WhatDoTheyKnow, mySociety aims to “bring the power of the Freedom of Information Act to everybody,” both by promoting the law as a vehicle for obtaining public information and by making it easier to place a request.
- In the US, MuckRock provides a similar service — although one that is considerably complicated by America’s federal structure and its lack of centralized locations for processing requests. (Embarrassingly, there is not even a central portal for making requests to US federal agencies, let alone the full range of American governments — although FOIA.gov does optimistically note that “most federal agencies now accept FOIA requests electronically”!)
Both WhatDoTheyKnow and MuckRock provide public-facing repositories of the requests processed through their sites, which allows other people to see if the information they wanted was already requested and made available.
If public records request portals provide static snapshots of precise requests, open data portals provide regular access to information governments have proactively chosen to make available. These data sources can be presented in ways which help them merge with people’s regular interests and needs, whether that is traffic-watching, or knowing how their government is doing with regard to accountable policing, as is the case with California’s OpenJustice site, or monitoring the status of different public policies as they move through the political system, as is the case with GovTrack in the US and TheyWorkForYou in the UK.
Both public records portals and open data portals help people outside of government better understand what’s happening in government and learn about how it’s currently working. It’s essential information for people to have if they want to make useful contributions to public decisions.
Agreeing to See if it’s Working
Civic technology is still a diffuse sector, without even a concrete definition in common. Nonetheless, it features some regular components. Civic technology aims to help government work more quickly. It aims to help get a wider variety of voices into government, through improving the avenues for input. It aims to provide that input in ways that help governments effectively target their services, make good policy, and use resources as wisely as possible. It aims to bring information out of government so that people can access the information they want, when they want it, in order to help provide the most useful input back into the governmental system.
We see how this dream of quick, responsive, effective government can work — powered by digital tools, clear public requests, engaged governmental response, and transparent tracking — across the small-scale landscape of today’s civic tech tools. If civic technology comes to achieve its goals, the whole of public life will resemble these experiments.
Is this a good thing? I hope so.
And I think it’s a very good thing, at the present moment, to know what it is civic technologists are driving at. We want to use digital technology to change the way that people and governments interact. To do this, we want a wider range of information in, a wider range of information out, and to know what’s happening in the middle.
And with that in mind, let’s see if the tools we’re making are helping us accomplish those goals.
