PoplusCon: Lowering the Tech Barriers for Civic Startups
BY Eilís O'Neill | Friday, May 2 2014
The problems are the same the world over, says Felipe Heusser, executive director of the civic NGO Ciudadano Inteligente, which is based in Chile. Corruption. Inequality. Poverty. But what if—he asks—what if civic coders around the globe worked together to solve those problems? They wouldn’t have to reduplicate efforts; they could save time and money and focus on only the most pressing issues. In the process, more problems would actually get solved.
That’s why Ciudadano Inteligente, together with the U.K.-based NGO mySociety, decided to create a way for civic coders around the world to collaborate and share solutions to common problems. The new program, called Poplus, aims to make it easier, cheaper, and faster for civic activists, even those with limited tech skills, to launch a useful website that monitors politicians and parliaments and sheds more light on those who hold the reins of power.
Heusser explained to techPresident, “[Poplus] is about civic coding. It’s about having a plan of what needs to be done in different countries. What are the shared stories? What are the shared problems?”
To answer those questions, almost 100 civic coders and activists from 27 countries came together from April 29 to 30, in Santiago, Chile. As an unconference, PoplusCon participants proposed, voted on, and ran the sessions themselves. In the process, they strengthened the Poplus community and had discussions about what Poplus components are; what new Poplus components need to be developed; how to improve existing Poplus components; and how to make Poplus components known by, available to, and useful for civic website developers around the world.
Poplus makes it easier for civic activists around the world to harness the power of the Internet to make governments more open, more participatory, and less corrupt. According to Heusser, it’s “all about reducing information asymmetries” between governments and citizens, businesspeople and consumers, wealthy citizens and poor citizens.
Currently, building a civic website—even a very simple one—is expensive. It costs money. It takes time. It requires technical expertise. Especially in much of the developing world, the resources and skills necessary to build an effective civic website simply do not exist.
“What we found out is that open source was not enough for many countries,” Heusser explains. “If you have an organization coming from Bolivia saying, ‘Hey, I also want to build a platform to monitor Congress,’ I can’t just tell them, ‘Yeah, that’s my repository of code; take a look.’ It’s not enough. So Poplus was an idea about how we can get together in a way that we not only share code but we also provide services to the people that want to use these different functions—functions that we need in different countries, functions that are very generic.”
Thus, Poplus members, instead of just open sourcing software designed with one specific social and political context in mind, create generic tools that will function in any context and will be accessible even to those who aren’t particularly tech savvy. These tools are called “components.”
Poplus components are like ingredients, explained Tom Steinberg in his opening speech at PoplusCon. They are pieces of code that solve specific, “boring” problems faced by civic coders around the world. Tony Bowden, a contributor to mySociety who lives in Estonia, said in a conference session, that, though Parliaments differ around the world, all legislatures “probably have people. Those people probably make speeches, probably vote on bills, and probably make changes to bills.” Poplus components model those basic parliamentary functions.
There are currently five Poplus components in existence, though many more are being dreamed up and developed. The component PopIt, which was designed by mySociety in the U.K. but is also currently being used in Kenya, South Africa, and Argentina, makes it easy for a civic website developer to create and maintain a database of politicians. WriteIt, built by Ciudadano Inteligente, allows citizens to email their elected officials. SayIt is a tool for searching, browsing, and sharing parliamentary transcripts, including floor transcripts and meetings minutes. The Parliamentary Monitoring Group in South Africa has used SayIt to build a website that monitors the country’s Parliament. MapIt enables website developers to link politicians to the area they represent, or to communicate information about problematic traffic points or local problems (such as in the U.K. website FixMyStreet). And, finally, BillIt enables website developers to store their documents online in an accessible format, which facilitates tracking bills, patrimony declarations, and government contracts.
Just as a cook might use rice, broccoli, and chicken to make dinner, but doesn’t have a rice paddy, a broccoli patch, and a chicken coop, the developer of a civic website can use PopIt, WriteIt, and SayIt to build a parliamentary monitoring website without having to reinvent the wheel. When website developers no longer have to focus on basic problems, they’re freed up to work on the problems specific to their local context. According to Heusser, this collaboration dramatically reduces the cost of building a civic website.
For example, mySociety’s WriteToThem website, which enables people in the U.K. to figure out who their MP is and send him or her an email message, “took months and months and months” to build, Tom Steinberg remembers. Now, using the Poplus components PopIt, MapIt, and WriteIt, “we could almost certainly collapse months down to weeks and maybe even days.”
According to the participants in the “What makes a Poplus component?” session, a Poplus component has obvious civic or democratic applications; a narrow focus; and a stable, documented API. It also is open source and interoperable with other Poplus components, and it aims to model the political systems of multiple countries. After all, the whole idea is that civic coders from different countries who speak different languages and work in different cultural and political contexts can work together to solve civic problems.
The session participants also batted around the idea that some tools or apps or pieces of code might be useful to people who are developing a civic website but might not, in and of themselves, be Poplus components. They decided these tools could be given the label “Poplus compatible” or perhaps “Poplus endorsed.”
Petrus Janse van Rensburg, from Code 4 South Africa, came to the conference in order to find code that he can use in his projects, which currently include parliamentary monitoring (tracking vote records, meeting minutes, bill versions, etc.) as well as a tool that monitors how much money governments in the Southern African Development Community pay for the medicines they buy. The latter is aimed at assisting governments in their negotiations with drug companies. He’s also involved in Media Monitoring Africa, a tool that scrapes news websites and extracts information to make it easy for reporters or others to get a sense of the media presence of a given politician or organization.
Janse van Rensburg thinks Poplus components might be useful in these or future projects, and, since he’s also a coder himself, he wants to find ways to contribute to existing Poplus components or to build new ones.
“It’s no good sitting at home trying to figure out what the world needs and then trying to build it,” he said to techPresident, “because you’ll probably end up building the wrong thing. So, for me, this conference is about bouncing ideas off of people”—figuring out what code could be relevant in various national contexts, and looking for opportunities to collaborate. “Since none of us have the capacity” of a big company, says Janse van Rensburg, “we’re forced to collaborate early on.” Nonetheless, “that’s a useful constraint because there’s value in collaboration and getting different opinions and doing things together.”
Other conference participants are already working on Poplus projects and came to get feedback and direction for further efforts. For example, James McKinney, from OpenNorth in Canada, is working on a project called Popolo which aims to create international open government data specifications which would enable civic coders to spend less time transforming data and more time using that data to increase government transparency.
“The idea is to come up with data specifications that would work for everybody’s context and that we could all adopt,” McKinney said to techPresident. By standardizing data specifications, Popolo would make it easier for Poplus components to work with each other and across international boundaries.
At the conference, McKinney had the opportunity to talk to other participants about what other things Popolo should do, and how to accomplish those goals.
Unconference sessions ran the gamut. Topics discussed include whether or not a Poplus component that illustrates parliamentary budgets is necessary, what a component tracking politicians’ votes would need to look like in order to be applicable internationally, how to crowdsource drafting laws, and whether Dockers would be the best way to make Poplus components interoperable. (A Docker is essentially an intermodal shipping container for software. Just as a standard shipping container can be filled with almost any consumer good and then moved from trucks to rails to ships, Dockers can be loaded up with code and then run on almost any hardware platform.)
In one session, participants compiled a wish list of Poplus components that should exist—such as a component that would track changes to bills as they move through the legislature, one that would allow users to find everything a politician said on a given topic, one that would look at how government ministries spend their allotted budgets, and one that would track politicians’ declarations of wealth in order to improve citizen oversight and decrease graft. The list goes on and on.
In the session about how to build a useful Poplus component tracking politicians’ votes, participants explained the peculiarities of parliamentary voting in their home countries. In the United States, for example, there are seven different ways to abstain from a vote. In the U.K., politicians can vote both yes and no. In New Zealand, parties, not individuals, cast votes. In Australia, those who seek to monitor Parliament have to infer based on the context provided by parliamentary transcripts what a particular vote was about. In Uruguay, parliamentary voting is recorded by typewriter with only a vote count and no record of who voted which way. Ideally, the Poplus component “VoteIt” (which does not yet exist) will capture the essence of parliamentary voting through a schema that outlines what all, or most, countries have in common, thereby freeing up civic coders to focus on the idiosyncrasies of their local political contexts.
In addition to these sessions, the conference featured so-called “lightning talks”—five-minute speeches on a wide array of topics. In these speeches, participants highlighted the work they’re doing and gave out nuggets of advice.
Alexandre Gomes, the co-founder of SEA Tecnologia in Brasilia, Brazil, shared an app that he made to crowdsource information about Brasilia’s bus routes. Brasilia did not formerly have an online public transportation trip planner, but, thanks to him, it now does. His take-away is that constraints—such as the complete lack of data—can be a catalyst for disruptive solutions.
In another lightning talk, Steven Clift, the founder and executive director of e-democracy.org, discussed the challenge of including poor and minority communities in online civic engagement. He said the central problem is that civic activists often don’t reach out to these communities, and what he has learned in St. Paul, Minnesota, is that it’s important to invite people from these communities to participate.
Akir Uchumura, from Chile, gave a talk about his project Nikkei Youth, which maps disaster relief efforts and needs around the world, such as in Japan after the tsunami or in Valparaíso, Chile, after the recent fire.
And Heusser gave a lightning talk about the importance of not only scraping government websites for data but also of talking to real human beings in government who can give you more information than what’s available online.
In closing speeches, the conference’s organizers stressed everything that had come out of the two days of conversations and work. First, civic coders from around the world had reaffirmed the idea of Poplus components as useful, important, and perhaps necessary. Second, many had expressed interest in using Poplus components in their websites or in contributing new components. And, third, participants had reached a consensus on a working definition of what makes a Poplus component and on what components should be developed in the near future.
Speakers also discussed potential challenges going forward: primarily, the challenge of maintaining the global Poplus community and that of making tools that are useful to those who would most benefit from changes to the way government operates. But they came with solutions in hand, including the formation of a Poplus Google Group and plans to organize component-themed Hangouts as well as lightning talks via webinars.
At the conference’s close, Heusser announced that the first era of Poplus—the initial, experimental phase—had come to an end. Now, the really hard work begins.
Eilís O'Neill is a freelance radio and print journalist who divides her time between the Southern Cone and Seattle. Her work has appeared in the Christian Science Monitor, New Internationalist, and other publications, and her radio pieces have aired on PRI's The World, NPR's Latino USA, Deutsche Welle's English-language radio shows, and other programs. In her free time, she likes to run.
Personal Democracy Media is grateful to the Omidyar Network and the UN Foundation for their generous support of techPresident's WeGov section.