Belonging to a democracy ensures an inalienable right to participate in public matters, whether presidential elections, town hall meetings or political protests. Arguably the most effective democracy is one in which every citizen can make informed decisions when exercising this right, upholding the most reasoned political arguments while dissolving those that are lacking or based purely on partisan judgments. The decisions made by congressmen elected to office should, in turn, represent the character and will of the people; if not, those individuals lose their place in government to more faithful politicians.

http://simplifycongress.org

Informing the public, however, has become an increasingly difficult ordeal, as much of the reportedly objective news landscape is punctuated by claims of bias, partisanship and favoritism towards special interests. Citizens rely on news media to understand the complexities of the government and its policies because, lacking a specialized education or experience in Washington, the raw facts make little sense without expert interpretation. But when political commentators in media often align themselves with a single party and its polarizing ideology, rather than base their opinions on factual information and make partisanship secondary, should the American public turn to a different source of analysis to understand how specific legislation will impact daily life?

Knowledge-based networks like Wikipedia have already paved a possible direction: information and analysis provided by the wisdom of crowds, or in this case, of constituents who will contribute their particular expertise to understanding these issues. I consider networks throughout this thesis to be what Yochai Benkler defines in The Wealth of Networks as 'commons-based' communities that collectively produce, discuss and edit self-generated information, as opposed to news media which broadcasts in a one-way form of information exchange. With original documents of important legislation made available through SimplifyCongress.org, citizens will be able to read, compose and share with other users their opinions and interpretations to distinct sections within each proposed bill, thereby providing everyone with simple and accessible tools for understanding Congress and disintermediating the often biased and opaque analysis of political commentators.

In this thesis I will discuss the current obstacles that prevent the formation of a more informed public, from the perspective of news media, as well as the challenges evoked by the polarized and highly partisan political environment in the United States. After contextualizing this project with previous work done by organizations that provide legislative information to the public or build networks around commons-based production or citizen reporting, I will describe the technical implementation of SimplifyCongress.org and the types of users likely interested in participating in its first iteration. By outlining future improvements to the service as they pertain to the ultimate goals of the project, I hope to ultimately show that interactive technologies have the potential to empower individual citizens to shape their own political understanding and retain a greater public awareness of government.

SimplifyCongress.org

Interpreting legislation as a community.

With these criteria in mind -- information, citizen media and transformation -- I created a platform giving anyone with a basic Internet connection and an interest in learning about congressional politics the opportunity to create, read and vote on summaries of current legislation created by citizens, called SimplifyCongress.org. Regardless of political affiliation or familiarity with the legislative process, SimplifyCongress.org offers a collaborative approach to deciphering the most confusing legislation in a way that relies on transparent, community-driven knowledge, rather than one-way broadcasts from special interests, lobbyists or polarizing commentators. In this section I will describe the five development phases SimplifyCongress.org underwent: originating ideas; redefining the scope to include both citizen input as well as information exchange; utilizing data and technology; building community oversight; and finally, incentivizing participation.

Origins

SimplifyCongress.org began as a final project for the seminar Mashups: Remixing the Web, taught by Google engineer Daniel Aminzade. The class was dedicated to "drawing upon content retrieved from external data services to create entirely new and innovative applications," and the appeal of solving the ever-worsening relationship between people and the government with data inspired me to take congressional data and present it to potential users. Despite the fact that OpenCongress had already achieved such a project, I felt that perhaps there were directions that the service had ignored or forgotten about and that needed to be explored, specifically with regard to tailoring its vast storehouse of legislative data for the laziest user rather than the most active who would actively sift through it.

The final project, which had no name and very little styling or presentation, was a simple form that included name to identify users, zip code to link users with their congressmen, party to affiliate the user, and a brief introduction that was passed through a Yahoo term extraction Application Programming Interface (API) to suggest issues and related bills that the user may be interested in following. As opposed to OpenCongress, which does not present tailored suggestions for users upon registering, the form exemplified a more active approach to building a bridge between the myriad interests of the public and the many bills affecting not only broad-strokes changes in America, but also the smaller transformations of specific issues or local areas. Once linked to legislation that matters on a personal level, users would feel more obliged to return to the service and follow the bills through their lifetime.

Designing for Citizen Input

Though the form-based final project could succeed in informing the public and generating interest in legislation, there was still a major component that both OpenCongress and the project lacked: a transformational effect on the legislation itself, much like what PublicMarkup aimed to achieve with legislation with the 72 hour reading period; and participation from users in summarizing and analyzing bills, in the vein of Global Voices. Participation in and transformation of original legislative documents are arguably the principle activities that can shift reliance away from polarizing political commentary in news media to more transparent discussion and collaboration by fellow citizens with like-minded goals in understanding Congress. Designing these additions, however, were beyond the capacity of the original project and required an expansion of scope to emphasize user input as much as information exchange within the service.

The other challenge was to identify how citizens could engage with and summarize the data, as the size and magnitude of legislative documents constitute a major obstacle in itself. OpenCongress added a feature allowing users to comment on each line of a bill, but that created too much fragmentation within the bill as often single lines do not contain much information, as well as too broad an entry to contribute to the service; PublicMarkup's emphasis on sections as points of entry into the bills, however, was ideal because they fully represent individual ideas within the bill and are neither too small (at least a paragraph) nor too big (at most perhaps ten paragraphs). By separating out and giving each section its own space for text, user input and discussion, the site frames legislative information so users can most efficiently extract simple meaning and understanding from the bits within bills.

Mechanics of Simplification

A project of this scope requires reliable data from a reputable source like THOMAS, but unfortunately Congress' in-house service currently has no public-facing API that could provide SimplifyCongress.org with protocols for retrieving this information. Luckily, Joshua Tauberer, a PhD graduate student and software developer interested in transparency in government, started a service in 2004 called GovTrack that allows other developers to easily access a storehouse of raw legislative information scraped from THOMAS through a software application called rsync. SimplifyCongress.org ensures that all data on the site is the most recent available by making automated requests to GovTrack in 12-hour periods, enough to follow the often slow pace of Congress while not burdening the popular service that Tauberer supplies the public.

As the main goal of the site to create communities around simplifying and discussing legislation, each section is displayed in a list on each bill page and has an attached form that allows registered users to fill out a summary that contains fewer words and is easier to understand than the original. Summaries should describe the legislative purpose of that section, in terms that are as unbiased as possible, although the purpose of the summaries is to also empower citizens to voice their reactions as well. Users are also asked to submit tags along with their summaries so as to allow others to search effectively through the bill page for summaries that match popular terms; for example, users who want to search for what happens to their beneficiaries with the health care reform bill can search for summaries tagged with those words rather than search through the hundreds of sections in that legislation.

Community Oversight

Once summarized by a registered user, others can view that summary alongside the original text and decide if that user's interpretation is too slanted, incorrect or misleading by voting to support or deny it and discussing it in section-specific discussion threads. These methods are an important factor when implementing oversight in an online, content-generating community because as opposed to one-way broadcasts in news media, once a critical mass is established the group can effectively check submissions and enforce edits and correction through discussion or votes. Like the vibrant community of Wikipedia, where in the words of founder Jimmy Wales "you are likely to be contacted and challenged if you have made a flawed argument or based your conclusion on faulty premises," users on SimplifyCongress.org will ideally watchdog inaccurate assumptions or poorly written summaries.

To avoid duplicated votes that would lead to inaccurate results, the service only allows users to vote once per summary. Each summary displays the average score beside the user who submitted it, as well as a clickable up and down arrows that symbolize assent or dissent. If users have given the summary a certain score, it is marked with a yellow medal to indicate exemplary work; if users have collectively voted the submission below a certain score, that summary is automatically removed from the service and the section is reopened to users.

Influence and Motivation

In addition to community oversight, motivation is a key factor in creating an active community and supporting a critical mass of users; at SimplifyCongress.org, encouragement to participate takes the form of a point system known as Influence, which users can collect and spend throughout the site. After writing a summary for a section inside a bill, users receive Influence based on the reading difficulty of the section's original text, generated by readability tests like SMOG. Users can then use Influence to vote up or down on submissions that inspire them to react, thereby creating an economy of effort and judgment in which users only vote on those they genuinely feel need criticism or support.

Incentivizing participation, as well as ensuring all users must act before they vote, separates those who merely visit the site to understand legislation, and those who want to contribute and be known as excellent sources of political knowledge. How users spend Influence and contribute to the site can also be the foundation of additions to the site's social dynamics, like badges for individuals who reach a certain threshold or spend the most points. Competition will hopefully spur greater participation and create an active and ambitious community that can progressively simplify the complexity of Congress.

The Power of Voting

Collecting influence gives users the power to spend their points by voting on summaries written by others, but are given only one vote per summary. If a user believes that another citizen's work is exemplary and should be rewarded, she will spend a couple influence to raise that summary's score by one, which in turn gives a couple influence to the user who wrote the summary; if a user believes the opposite, that the summary exhibits partisan bias, the negative vote will take away influence from the summary's creator. There may be concerns that many users adding and subtracting from each other's influence could cause, for example, experience extreme dips and unfair behavior, but daily caps on how much influence a user can lose or gain will hopefully mitigate such a dynamic.

Influence and voting could successfully maintain a nonpartisan atmosphere conducive to creating unbiased summaries, in that many will focus on proving they know a great deal about the actual facts of current legislation that are rewarded through points. Partisanship may be a successful strategy for winning influence in the political commentary arena within news media, but within the SimplifyCongress.org community the most rewarded participation will be one that includes consideration of the issues and attempts at accuracy. It is, therefore, important to build a foundation of users who understand these goals and enforce them through their interaction with the site to check possible partisan or unreliable opinions.