Comments
Hi everyone,
I’m part of a small team working on a project that could be beneficial to EA and the world, and we’re looking for people willing to help refine the idea and make it a reality!
The project (which doesn’t have a name yet--suggestions welcome) hopes to reinvigorate democracy by creating a nonprofit web and mobile platform for civil public deliberation, where people can add their opinions to topics or issues regarding different organizations—not just national governments, but also regional and local governments, supranational organizations, private corporations, nonprofits, community groups, universities, industry associations, and more.
I’ll organize the rest of the post as follows: first, I’ll briefly explain the context of the project, which most of you are probably already familiar with. I’ll then lay out the goals of the project and the theory behind what we think needs to happen, before explaining the details of what we’re actually trying to build, how we envision our project functioning, and how it’s different from existing projects. After that, I’ll give some more background on the team, and I’ll provide a very rough plan for the project as well as a description of how we think it could be utilized by the EA community. The end of the post will contain information about how you can get involved.
I’m happy to answer any questions or provide more detail in the comments or over private message or email, and any thoughts, suggestions, or feedback would be greatly appreciated.
Project Background
It’s no secret that contemporary politics is dysfunctional. Ideological groups have been polarized into echo chambers, public trust in government is dangerously low, and there are often sharp discrepancies between public opinion and policy. Trust in democracy is at a low point internationally, and democratic backsliding is occurring in many places that were previously considered to be established democracies. Many leaders are increasingly empowered to act in ways that defy the public’s interests or wishes, without many adverse consequences. Citizens, all the while, are often unaware of the stark discrepancies between public opinion and the actions of leaders.
At the same time, many people want to be more involved in civil society, but simply don’t know how or don’t have the time.
Project Goals
This project is guided by three overarching goals, based on the problems observed in the background:
For the first goal, we define “accountability” as a healthy balance between democratic input and expert judgment--the public should be a central component of decision-making, and leaders should generally face strong incentives to represent the beliefs and interests of their constituents, but there should also be wiggle room for leaders to break with the public if there are good-enough reasons to do so. More robust forces of democracy are desirable, but it’s also important to prevent this tool from becoming a breeding ground for populism. For this reason, we chose to focus on developing a platform to enhance conversations about the actions of leaders and institutions, rather than trying to create a mechanism for direct democracy.
The second goal is naturally limited in scope, since “public discussion” encompasses most human communication, in some form or another. Instead of trying to change the nature of “public discussion” in general, we are instead concentrating on providing an inclusive, easily-accessible space for people to express and discuss their thoughts--and we’re developing new moderation techniques to keep the conversation constructive, as is explained later in this document.
The third goal stems from a side benefit of our answer to the first two goals: we realized that our proposed platform could be easily leveraged to offer local advocacy groups an opportunity to collaborate with other advocacy groups in different regions, and to share knowledge, campaigns, and resources. This goal is not as central to the project as the other two, but we think that it would nonetheless be valuable to help build up an infrastructure to turn grassroots efforts into larger-scale change.
We can further break down the goals of the project into the benefits it seeks to provide for different stakeholders. This project will seek to provide citizens:
It will provide leaders:
Theory
In order to hold anyone accountable for anything, one must have some sort of control over something that person desires. Without such control, nobody would have any external incentive to act how others want them to. Different categories of leaders have different root goals, and therefore face different incentive landscapes--but luckily, the public has control over the root desires of most types of leaders.
The main goals of public, elected, or community officials are usually a mix of the following: serving their constituents, promoting their ideology through policy, gaining power, and making money.
For leaders who truly prioritize serving their constituents, the barriers to accountability are often informational and structural. Information about public opinion on national hot-button issues may be widely available, but polling data at lower levels of government or in different forms of organizations is much more scarce, especially for mundane topics. It can be prohibitively costly and inconvenient for an elected official representing a smaller constituency to obtain accurate information about public opinion.
A lack of information can create compounding barriers when considered in the context of organizational structure. As an example, representatives in partisan legislatures may face high costs for voting against their party, such as being given poor committee assignments or losing re-election support from party leadership. Without clear evidence of their constituency’s beliefs, representatives may not be able to present their party leadership with a reason for their vote that is strong enough to avoid punishment.
For leaders who prioritize anything else, their ability to meet their goals hinges, in most cases, on re-election--and since the public retains control over re-election, it is a clear target for increasing accountability.
If the public already has this power, though, why is non-accountability an issue in the first place? Outside of electoral system structure, there seem to be two main reasons, both centered around a lack of transparency: 1) voters are largely unaware of how their representatives vote on issues that they find important, and 2) representatives are often unaware of public opinion. The first reason explains why we find ourselves re-electing politicians who don’t follow our views: we’re often only aware of how they voted on flashy issues with lots of press coverage, even though those issues might not be our primary concerns. The second reason explains why even good-natured representatives may fail to accurately represent the public. Without clear evidence of their constituency’s beliefs, representatives may not have a good enough reason to break with their party. Since representatives can face high costs for breaking with their party (especially in terms of re-election support from party leadership), representatives face no incentives to vote counter to their party unless there is enough strong evidence to show that their constituents believe otherwise.
Leaders of private organizations will often have slightly different goals than elected officials or community leaders. Most leaders of for-profit corporations are expected to add value for shareholders, grow the business, and gain investors, while also making money for themselves, and incorporating corporate social responsibility. Nonprofit leaders face a similar incentive landscape, except with more focus on social welfare, and fundraising to replace profit-making.
As with elected officials, the public has some control over these desires. Private organizations rely on the public for their income, giving consumers the power to constrain how these corporations act. Boycotts, for example, enable the public to temporarily cut off a chunk of a corporation’s income, prompting it to negotiate or comply with the public’s demands. Even if only a small portion of the public takes issue with the corporation, the potential damage to the corporation’s brand may be threatening enough for the corporation to change their actions.
As with re-election, though, boycotting isn’t being utilized to its full potential as a tool for accountability. There are powerful examples of boycotts creating momentous change, but most consumers do not have the time or attention necessary to boycott all corporations who have taken actions that they disapprove of. There are far too many corporations in the world for consumers to pay attention to, especially given the time and effort required to learn about and act on each concern individually. Instead, it would be beneficial for citizens to express their general attitudes simultaneously to a class of related corporations, saving significant amounts of time and energy, and empowering busy citizens to contribute to movements they care about. Many people also feel that petitions and boycotts are generally not effective anyway, and therefore don’t actively seek out opportunities to engage in them. It might thus be necessary to rebuild confidence that civic action can actually make a difference, as advocacy groups constantly try to do.
Based on this rough analysis, we can identify a few things that need to happen if our goals are to be met:
In other words, information about the beliefs of citizens and the actions of leaders needs to be compiled side-by-side into a single location, so that leaders can see what their citizens think and citizens can see clearly where their leaders acted against their wishes. Crucially, the latter component ensures that leaders will have a reason to pay attention to citizens’ opinions: armed with this previously-obscure knowledge, citizens are empowered to take action against the leaders that represent them poorly, by voting them out or refusing to support their business.
Project Output
Considering the essential factors listed above, we decided to achieve the goals of this project with a web and mobile application, since online services are accessible, convenient, and frequently used, and are ideal for facilitating more efficient flows of information across a wide number of people. The app will function doubly as a communication platform and as an information source, with the following features:
Users
Organizations
Leaders
Issue Pages
Moderation
An Example of How it Could Work
It is probably easiest to understand how the platform will work with a concrete example. We’ll consider the Emerald City government as our organization. A citizen notices that the roads in Emerald City needed repair, so she adds a page devoted to discussing this issue. Let’s say she also contributes her viewpoint to the page, arguing that the city government should fix the roads. She shares the page with her friends, who all lend their name to support her viewpoint.
Some other citizens, however, disagree. They think that the city government should spend its limited money on the more crucial problems of public health. One of these citizens posts this viewpoint to the page, and some other people browsing the page add their names to his viewpoint. A few crafty think tank researchers devise a third option: instead of spending public funds to work on road repair, the city could privatize its road system. They post this to the issue page, and garner support for it on the platform by mentioning the idea in their newsletter. As this issue is heating up in the community and a few more unique viewpoints are added, the platform is busy calculating estimates of what public opinion on the issue looks like.
The issue soon becomes popular enough that the Emerald City Council starts paying attention. One councilmember introduces a bill to allocate more funding to road repair, with some specific measures that are also contentious. Somebody adds an issue page for this bill, which extends the more general discussion on the road repair page. This new page sparks a few viewpoints, now that the issue is directly relevant to taxpayer money.
Data from the viewpoints suggested that most residents were in favor of the bill, except for those in the poorest district, who were unhappy that the bill offered little money for road repair in their neighborhood. The councilmember from this district wanted to vote for the bill, but her chief of staff saw that her constituents passionately opposed the bill. She knew that, if she voted for the bill, a local watchdog organization would input the voting results to the platform, which would then calculate an index to describe how much the city council members diverged from their constituencies when voting. This would mean that many of her constituents would get a notification on their phones that their leader diverged sharply from their views—ruining her re-election chances. To prevent this from happening, she brokered a compromise deal with the other council members that addressed her constituency’s concerns.
Comparison to Existing Projects
It’s a worthwhile question to explore how this platform will be different from existing options, especially since there are so many established platforms that have overlapping functionality with this project. There are a few different classes of existing platforms that may be similar: online polling and petition sites, social media, direct voting systems, and leader activity-compilation sites. This project will attempt to extract the successful components of these services, combining them into a single platform:
The unique purpose and combination of features that this project possesses sets it apart from these existing services, so we’re confident that we aren’t reinventing the wheel, but instead developing a new tool that doesn’t yet exist.
By incorporating elements from each of these classes of organizations, citizens will be able to condense much of their civic activity into a unified place, giving them more time to focus on issues, and rendering each of their actions more impactful. Leaders will also benefit from having a unified place to learn about their constituents’ thoughts and communicate with the public.
Project Team
Currently, we are operating with a small group of a couple dedicated volunteers, and we’re seeking to add some more people to the team.
I have a background in politics, research, and nonprofit settings, and other team members have deep experience in business leadership and software architecture in both startups and established organizations. Together, we have a network spanning advocacy, research, and charity nonprofits, multiple private corporations in different industries, and actors at all levels of government, including city council members, state officials, and former members of Congress.
As is, we’re confident that this project will not lose momentum or fall apart, but we think that it would be much stronger with the additional support of a few more tenacious volunteers.
A Brief Project Plan
Below is a rough outline of the plan for this project. Right now, we are between stages 0 and 1. We’re working on getting feedback on the idea, finding people interested in helping, and developing a very basic proof of concept.
Stage 0: We will ask experts and community members for feedback on the idea, and any suggestions for improvements. We will also engage in “market” research in order to determine whether citizens and leaders would use such a platform, and how often they believe they would do so.
Stage 1: In this stage, we will recruit volunteers to help build the minimum viable platform for this project. It is expected that the MVP will take 3-5 months to develop, depending on the size of the volunteer force. Building this platform is within the capabilities of a small but dedicated team of volunteers, since longer development timelines are not a concern at the moment.
The MVP will likely contain the core features of the platform: issue pages for organizations, viewpoint contributions, and displays of community-inputted information on leader actions. Since these features are the most critical for reaching the project’s goals, we think that they will constitute enough to complete a first test and evaluation of the project. We will replace many of the bells and whistles of the full version with simpler implementations: as examples, member validation will likely be based only on email address, moderation will likely be completed by designated users, and organization announcements and workspaces will either be omitted or limited to simple text. Trimming these additional features will allow us to build a basic release-ready platform in a shorter amount of time, with fewer people. More embellishments can be added later on, with a larger team or more external support.
Stage 2a: Once the MVP is built, we will target university student governments for a trial of this platform. Student governments were chosen because of their engaged constituencies, medium size, and dynamic challenges, but other settings may be appropriate too. Based on the network of our team members, we will seek to work with ~10 different student governments to roll out the basic platform for a semester, invite their students as members, and engage in intra-university marketing to make students aware of the new opportunity to share their concerns or ideas with their leaders. This stage will likely take 3-5 months to complete, following the 3-5 month development period.
Based on feedback from this trial, we will evaluate the project’s performance thus far, and improve the platform as needed.
Stage 2b: We will then target local governments and citizens’ groups and encourage them to use this platform to include more citizens into their discussions, again based on the collective network of our team.
Many local governments are seeking to increase citizen engagement, and may be more willing to test out new technologies than other sorts of organizations. At the same time, advocacy groups are constantly trying to get the attention of their local governments, and they are likely to find value in this project. In this stage, we will simultaneously work with the local government and existing advocacy groups of one or two cities at a time, concentrating our efforts on building active local communities of users. Our goal with this strategy is to seed multiple pockets of engagement as small examples of how the project could function at a larger scale. This approach is also designed to circumvent a potential concern with the project’s implementation: most citizens won't take interest in the platform unless they think their voices will be heard, but leaders won’t take interest until a substantial number of their constituents are using it. If only one of these groups is targeted, usage of the platform might fizzle out quickly. By initially focusing our marketing on specific local governments and advocacy groups at the same time, though, we’ll be able to offer citizens and activists assurance that leaders will pay attention to the platform, and we’ll be able to show leaders that their constituents are engaged on the platform.
This stage might take six months to a year to complete, depending on the enthusiasm of local organizations and our internal capacity to work on outreach. We will continuously conduct more evaluation and improvement rounds, as the user base of the platform grows.
Stage 3: Once there is an established user base in local communities, we will expand to other levels of government and types of organizations through social networks and additional targeted outreach. This stage is not well-defined quite yet, since it will depend heavily on the evolution of the project, but we expect that it will follow from the combination of natural spread and focused marketing.
For a short summary of the tentative timeline: if we can have a working product built by May 2021, we can spend the summer months setting up partnerships with institutions for our first trial round, starting in August or September. We can then assess and improve in December 2021, and roll out our second trial round in January 2022.
Funding and Finance
At least initially, this project will be volunteer-run, so there will be little need for external funds. As the project grows, there will be technology and maintenance costs involved, mostly pertaining to building the website and app.
There are two major options for funding:
If the project shows continued promise during Stage 2a (which we’ll do our best to make happen), we are expecting to incorporate as a nonprofit organization. Most likely, we will then search for private grants to cover increasing technology costs, using the data we collect in the trial runs to make a case for the value of the project. Contingent on our success in the grant search, we may also seek to hire a limited number of paid software engineers and nonprofit development staff to expand the functionality and reach of the platform, especially if the project takes off faster than a volunteer team can keep up with. This paid team would also be charged with implementing our second funding option, transitioning the nonprofit to a self-sufficient state.
Project Evaluation
This project could be evaluated according to a few key metrics: 1) user count, 2) quality of public discussion, and 3) self-reported user feedback. Most obviously, this project would be increasingly successful the more people use it. At least initially, user count will be the main metric. Once discussion picks up, it is foreseeable that quality will matter. Is the platform doing a good job of keeping discussion civil? Is there actually any fruitful deliberation happening? We can eventually supplement these metrics with feedback from users: do they think it is helping them make more informed decisions at the ballot? Do leaders factor insights from the platform into their decisions?
These questions are a sampling of the many that could be asked in order to measure the success of the project. More concrete evaluation metrics will be established more clearly as the platform takes shape, as well as plans for gathering and using feedback to guide the project.
Relevance and Benefits to EA
We think this project could be a very useful tool for the EA community. For starters, it will try to expand the respectful and nuanced discussions we have here to a broader number of people, with the aim of reducing undue polarization and kindling constructive discussion.
The platform could also be employed by EA-related advocacy groups to more effectively push for specific changes and to amplify the visibility of their ideas. An animal welfare group could raise an issue with a corporation about its factory farming practices, and conveniently garner and present public support for their concern. Groups interested in reducing global poverty could use the tool to campaign for different national aid interventions, and easily work with other groups to lobby multiple governments simultaneously. A city-based climate change group could do the same, using the platform to coordinate with many other city-based groups to lobby for similar policies in parallel--allowing smaller, decentralized actions to amount to the equivalent of a much larger top-down initiative.
It would be a great place to facilitate dialogue between technocratic and democratic forces, which is a relevant task for developing emerging technology policy. Some extended features could enable qualified experts or research organizations to add specially-designated viewpoints, or to quickly poll citizens for suggestions, ideas, and concerns on certain policies. As a very speculative potential use of this platform, it could even help guide AI alignment, by providing a bank of information on what people value.
At the broadest level, this project fits neatly into the scope of longtermism. If we want to ensure a secure future, it’s essential that we strengthen healthy forms of human organization where it already exists, and encourage it to grow where it doesn’t. This platform could help with both of these objectives by reducing the likelihood and resiliency of autocracy, increasing the efficiency of organizational communication and problem-solving, and fostering an engaged public. If nothing else, it could spark more interest in the different ways that network technology can be used to improve liberal democracy.
Getting Involved
If you’re interested in this idea, we would love to talk with you! Everyone is welcome (and encouraged) to get in contact, especially people who would be willing to help create and grow this project. Here are some of the specific skills that would be particularly helpful to us at the moment:
If you have questions or suggestions, or are interested in helping, you can either write a comment, private message me, or send me an email at [email protected]. All thoughts are appreciated.
Thanks!
Another type of related existing project is government-side community engagement tools, such as https://publicinput.com/. Their software makes it easier for city governments to ask questions and seek feedback from their residents. I know Jay started out with the goal of trying to get elected representatives in direct conversational contact with their constituents, and I think it'd be worth asking him why he went this direction instead.
Thanks for the tip! I'll try contacting him through the website you linked--it would be great to hear more from people who have attempted this sort of project before.