Due to an editing error the first paragraph of this story was initially omitted.
More than 4 million people joined together online in December 2011 to express outrage over the Stop Online Piracy Act, a bill Congress was considering that would have made content-sharing websties legally responsible for their users' copyright violations, with punishments including prison time.
The proposed law would have drastically changed if not destroyed the business models of YouTube, Wikipedia and other popular websites whose content is contributed by the public. People who frequent those sites vented their ire in online petitions, blog posts and Web comments. They emailed their representatives in Congress and bombed the lawmakers’ Twitter feeds and Facebook pages with angry comments.
As 2011 drew to a close, politicians got the message. One by one they announced their opposition to SOPA or to its Senate counterpart, the Protect Intellectual Property Act. President Obama said he opposed the legislation too, and in January 2012 both houses shelved the bills before they ever faced a floor vote.
Experts called the campaign a victory for digital democracy: The people had spoken— the ones who don’t have lobbyists or make large campaign donations. And just as important, their representatives had listened.
There was a problem, though. Through social media, ordinary citizens told Congress and the president what they didn’t want. But the filmmakers, recording artists and others concerned about protecting intellectual property rights, many of whom supported SOPA, had a legitimate beef. And there was no good way to gauge what measures the public would support to address that.
A handful of staffers in the office of Rep. Darrell Issa, R-Calif., thought they might have a solution. As the debate over SOPA rose to a boil, they launched the Madison Project, an online forum where users could comment on proposed legislation, suggest alternative text and vote those suggestions up or down. It was a cross between Microsoft Word’s track changes function and crowdsourced book reviews on Amazon.
SOPA was Madison’s first guinea pig.
“On the first day we had 150,000 unique visitors and were getting ridiculously awesome content, with ‘awesome’ defined as useful,” says Seamus Kraft, then an Issa staffer who helped launch the project.
Comments came from Web-savvy citizens as well as from a handful of advocacy organizations such as the Electronic Frontier Foundation.
Since that first posting in 2011, Madison has hosted other bills aimed at preserving Internet freedom, making federal spending more transparent and increasing cyber protections for government computer networks, most of which have been introduced in Congress.
Issa introduced the Online Protection and Enforcement of Digital Trade Act, a SOPA alternative developed largely through Madison edits, in January 2012. Government Executive Media Group’s technology website Nextgov is hosting the Madison version of a bill introduced by Issa and Rep. Gerry Connolly, D-Va., to reform federal information technology purchasing.
Management of Madison has also spun off into the nonprofit and officially nonpartisan OpenGov Foundation with Kraft, Issa and California attorney James Lacy listed as co-founders. Issa, who made millions in the automotive security field before running for Congress, provided the foundation’s seed money. It has received additional funding from the Knight Foundation, which supports media and open government initiatives.
The OpenGov Foundation has also partnered with the state of Maryland to release a machine-readable version of the state’s code of laws and plans to release all bills introduced in the Maryland legislature in a social forum during the next legislative session, Kraft says.
“Do I think we’re ever going to develop to the point where everything is in an accessible interface like Madison and citizens are helping write every regulation or piece of legislation from scratch?” Kraft asks. “No. But can that happen frequently? Yeah. And the data will be there for [lawmakers and others] to do whatever they want with it.”
Madison is just one of a dozen or more initiatives that are transforming the way the government engages citizens online. They range from gathering feedback on Philadelphia’s transportation reforms to making computers more accessible to the blind. While the initiatives have different objectives, they have three things in common: First, they target more specialized audiences than the typical government Facebook post or Twitter blast. Second, they tend to live on platforms tailored to their audiences. Finally, they are interactive, asking more from participants than a simple like or retweet.
Social media took hold slowly in the federal government. In 2003, during the George W. Bush administration, White House Chief of Staff Andrew Card became the first administration official to answer citizen questions online, but it wasn’t until 2009 that most agencies began launching Facebook, Twitter and YouTube profiles. The impetus to use social media in government has gained urgency, though, spurred largely by agencies’ desire to remain relevant to increasingly tech-savvy citizens for whom checking for updates on an agency’s blog or home-page seems hopelessly inefficient.
The earliest and easiest way for government to use social media is as a broadcast platform for policy statements and press releases.
The second phase is interactive, which typically has meant responding to citizen questions. The Education Department’s Federal Student Aid office holds regular Twitter office hours, for example, to answer questions about the labyrinthine process for submitting the Free Application for Federal Student Aid. The White House website We the People offers official government responses to petitions that receive more than 100,000 signatures.
The third phase of social interaction goes beyond responding to questions to involve expert citizens on a particular issue in the policymaking process. Citizen edits to legislation on Madison, for example, have been incorporated into the official bills.
In other cases, the document itself might be open to public edits. When an interagency team began work on a governmentwide open data policy that was released in May, the group decided to do the majority of its editing inside Github, a popular open source development site that stores a unique URL for each edit to computer code or a text document. Once the policy was released, the team opened the documents up to anyone on GitHub who wanted to weigh in.
Within days, more than three dozen citizen edits had come in, ranging from correcting typos to suggesting larger policy changes. Federal Chief Information Officer Steven VanRoekel personally managed and integrated many of the citizen updates, known as “pull requests.”
Ben Balter worked on the open data policy as a Presidential Innovation Fellow and now works at GitHub as a liaison to the federal government. He envisions a world in which similar policy documents will be open to the public before they are officially released so nongovernment experts can contribute at every stage of the process.
“The government, in its public-facing efforts, has the tendency to say ‘we’re the experts here and if you want to give us some free labor that’s fine, but we know what’s best,’ ” Balter says. “But that’s not how the Internet works. If the government publishes something on a social network, that means the world’s foremost experts can comment on it, which can be very helpful. The government has to shift from worrying about managing that community to worrying about growing it and making sure it has the resources it needs.”
The developer set calls this a “Web-based” model, meaning disparate parts of the machine are all cranking at once and fully aware of one another’s work. They contrast that with a “hub and spoke” model in which a central authority, say a federal agency, manages the entire development process and integrates feedback and advice from all the other players one at a time.
As agencies struggle with tighter budgets, relying on volunteered expertise from outsiders can be a money saver and a force multiplier, Balter says. The culture shift is that government will have to be freer with the staff memos, budget requests and other information that’s typically guarded from public view.
“In the intelligence community, yes, things need to be centralized,” Balter says. “But a lot of stuff the government is working on, if it’s data that eventually will be or could be public, there’s no reason they can’t open it up earlier [and] that they can’t blur the distinction between government and public.”
Kraft sees similar potential when uninvited experts comment on legislation through Madison.
“A lot of people know a hell of a lot more than you do about any issue you’re working on, and you should have the ability to put that to work,” he says.
Not all examples of this new breed of interactive social media happen at the macro level of legislation and presidential directives. Agencies across government have been turning to the platform IdeaScale, for instance, to gather feedback on more granular policy questions.
Once an agency poses a question on IdeaScale, anyone can offer a response or suggestion and other discussion participants can vote those suggestions up or down. That typically means the wisdom of the masses will drive the best ideas from the most qualified participants to the top of the queue without officials having to sift through every suggestion.
The Labor Department’s Office of Disability Employment Policy has run numerous IdeaScale campaigns aimed at addressing accessibility for the disabled, such as improving transportation options for veterans and making it easier for students to enter the workforce.
Answering policy questions like that in a smart way requires gathering and processing feedback from up to 5,000 people spread across the Defense, Veterans Affairs, Transportation and Education departments; dozens of businesses, nonprofits and advocacy organizations; and from disabled veterans and students themselves, the office’s policy adviser Michael Reardon says.
By gathering that feedback through IdeaScale rather than mass emails and public forums, the office not only saves time and money, it also removes barriers to participation and receives continual feedback as it watches constituents hash out ideas in public, Reardon says.
“If you’re someone with a disability and you want to provide input on a transit authority that provides services to you, [typically] you’d have to show up at a meeting site, take off work, go in front of a group for three minutes and then there’s no feedback mechanism,” Reardon says.
Meeting Your Audience
A critical element of interactive social media is that the platform must fit the audience, Kraft says. The cobbled together first draft of Madison was ideal for a Web savvy crowd that was eager to weigh in on SOPA and other Internet-focused legislation, he says. The OpenGov Foundation is working on an updated and more user friendly version due out this summer.
Madison 2.0 will offer “all the functionality of Facebook,” Kraft says. Everyone who uses the site will be able to create a personal page with legislative issues, bills and edits. Lawmakers will be able to drag and drop proposed bills into the system and outside contributors can collaboratively write legislation from scratch.
Similarly, GitHub made sense for the developer crowd interested in collaborating on the White House open data policy while IdeaScale is a better fit for reaching a more general constituency.
Finding the right platform to reach a narrow audience doesn’t necessarily mean searching out some dusty corner of the online world. When Michelle Lee was asked to gather feedback about proposed changes to Philadelphia transit lines, she considered numerous online platforms. The problem was that about 40 percent of Philadelphia homes lacked broadband Internet access, which meant the people most affected by the changes were unlikely to take part in an online survey.
Lee worked on the project with Code for America, a nonprofit that pairs technology entrepreneurs with municipal governments. She launched a text message survey of frequent bus riders along routes that might be served by a new rapid transit system. Signs urged riders to text a number to take a three-question survey, which asked about their transit habits and how those might change if a rapid transit option was available.
About 90 percent of the people who started the survey finished it. When the city updates its transit plan, Lee’s team plans to text survey participants to let them know how their feedback was used.
“Text appealed to us because the status quo to participate at that point was to go to a two-hour meeting and this pulled that commitment level to the other extreme,” Lee says.
Lee is now the chief executive of Textizen, a for-profit business housed, until recently, in the San Francisco-based Code for America’s startup incubator. Textizen has run surveys for other cities, including one for the Flint, Mich., planning commission, which was rewriting its 20-year strategic plan.
Most attendees at planning commission meetings were in their mid-50s, Lee says, which meant they’d have little stake in questions like whether the city offered good elementary schools. Via text, the commission was able to collect a broader base of citizen input, she says.
A Civic Universe Online
What many people see as the endgame for projects like Madison and Textizen is a vibrant civic culture in which people report potholes, sign petitions and even vote online or through mobile devices.
The Internet is great at gathering and processing information, but it’s not as good at verifying who that information is coming from, says Alan Shark, a Rutgers University professor and executive director of the Public Technology Institute, a nonprofit that focuses on technology issues affecting local governments.
“Star Trek is here,” Shark says. “We have these personal communicators, their use is continuing to grow dramatically and we’re going to have broader civic participation because of it. The missing piece is trusted identities.”
This is a stumbling block the Federal Communications Commission ran into in 2009 when it sought to use IdeaScale and an FCC blog not just to gather informal policy recommendations but also to collect official comments on a proposed regulation aimed at ensuring net neutrality.
Net neutrality—the idea that all information packets sent through the Internet should be treated equally rather than larger players being able to buy preferential treatment—is a passionate issue for many Web entrepreneurs and consumers, most of whom are unused to posting comments in the Federal Register. The FCC wanted those people to have a say, so it met with them someplace more accessible, says GitHub’s Balter, an intern in FCC’s technology office at the time.
Accepting comments through the blog and IdeaScale proved nearly as great a bureaucratic hurdle as publishing the proposed rule, Balter says. The 107-page notice of proposed rulemaking includes several legalistic paragraphs explaining that the commission would only accept IdeaScale comments that included the commenter’s identifying information.
The question of trusted identities online stretches far beyond civic participation. The Commerce Department’s National Institute of Standards and Technology is focused on the problem, largely to stem cyber espionage and identity theft in the private sector. But the institute also is interested in helping government provide better online services, such as processing tax bills, Social Security benefits and veterans’ disability claims.
Once NIST is satisfied with digital credentialing standards, that could be the key to online voting and other forms of civic participation, Shark says.
Some deeper form of credentialing also could help state and local governments and Congress members assess feedback they receive through Facebook, Twitter and other social media, by sifting out comments from people who aren’t constituents, according to Balter.
“Otherwise I could tweet to my congressman that I support gun control or something like that and say I’m Ben from Chicago, but I’m actually in Russia or China,” he says.
Such credentialing could drastically change the voting landscape, Lee says.
“If people could vote through text messages or some other convenient media, if they didn’t have to go to a polling site and verify their identity and fill out a form every few years that would transform who votes and how we think about voting,” she says.