The Future of Democracy

Imagine a government that engages with the public to craft policy, rather than listening to a few well-connected stakeholders and lobbies. A government that fosters unity and compromise, rather than outrage-driven tribalism. A government that trusts its citizens, and is trusted by them in return. Imagine a government committed to radical transparency, where most meetings are recorded, transcribed, and published. Where freedom of information requests are met with persistent public APIs rather than heavily redacted printouts. Imagine the future of Democracy as an upgrade to the social technologies we use to collectively reach consensus and direct our resources toward human thriving.

Today I learned that the future is here, and the future is in Taiwan. Yes, Taiwan (pop 23M), the country that recorded 900 Covid deaths (0.004%) without lockdowns or business closures. Compare this to California (pop 40M), with 86,135 deaths (0.2%) despite aggressive shelter-in-place policies and business closures. It turns out that in the last 6 years, Taiwan has quietly but dramatically upgraded their Democracy. And at the center of it all is Audrey Tang, a 40-year-old, non-binary, conservative-anarchist hacker who currently serves as the Digital Minister of Taiwan.

I’ve been excited about Democracy 2.0 for a long time, but somehow Audrey and Taiwan slipped below my radar. While I’ve been reading about Quadratic Voting, Liquid Democracy, and better coordination mechanisms, Audrey has managed to actually deploy these tools to a country of 23M. And despite the apparent success, neither I nor any of my friends had heard about it.

I want to share some of the things Taiwan is doing, and maybe you will be equally inspired.

1. Pol.is is a light-weight online tool for driving groups of people towards consensus. Existing social networks make it hard to have productive discussions for a variety of reasons. They foster in/out group dynamics, drive attention toward conflict/disagreements, and even the best threads tend to degenerate to ad-hominem attacks. Pol.is takes a different approach. People share their sentiments, and can upvote / downvote the sentiments of others. There is no “comment” feature, so there is no way to start or maintain an argument. As people agree / disagree with other sentiments, they are clustered into groups and are shown the main areas of agreement and disagreement in their group, and between their group and other groups. Sentiments that achieve plurality are promoted, so the incentives are to craft statements that both your group and other groups would support. This allows for groups to explore the solution space and focuses attention on areas of agreement. It allows a large number of people to brainstorm together, and surfaces core values and areas for agreement. Taiwan has successfully run 100+ conversations using this tool, and uses it in the early phase of policy making. They used it to help draft the Uber / AirBnb policies for the country, and it resulted in a widely supported policy. Pol.is is open-sourced and has been used by outside of Taiwan. If you want to learn more, check out this detailed blog post and demo.
2. vTaiwan is an online-offline consultation process which brings together government ministries, elected representatives, scholars, experts, business leaders, civil society organizations and citizens. The process helps lawmakers implement decisions with a greater degree of legitimacy. It has various touch points such as a website (vtaiwan.tw), a combination of meetings and hackathons along with the consultation process. It starts by using Pol.is to crowdsource the agenda, and then engages legislators and experts to draft the policy in a transparent (live-streamed) manner while staying engaged with the public.
3. Radical Transparency — All meetings with Audrey undergo a 10-day co-editing period and then are released publicly. This serves to make conversation participants mindful of what they say because they know people in future generations will be watching. Any non-private data is released publicly and persistently via API. There are hundreds of datasets from various agencies, and new ones are being added.
4. g0v is a top level domain that allows hacktivists to “fork” government websites and provide better interfaces. The first .g0v website presented the budget as an interactive visualization, allowing citizens to engage with the budget rather than having to grok hundreds of pages of PDF. Because Taiwan shares so much data via APIs, citizens can (and do) build better versions of government websites and the government can forgo expensive IT contracts.
5. Humor over Rumor The government hires comedians to quickly respond to misinformation by making funny memes. The memes spread faster than the rumors, essentially inoculating the population. When the pandemic first hit, some netizens started circulating the rumor that the state would run out of toilet paper soon, since the resources were being used to make medical masks, she shared. Within hours, Taiwan’s Premier posted an infographic debunking the rumor and dissuading citizens from stockpiling. The infographic even featured a caricature of him shaking his bum, playing on the similar-sounding Mandarin words for “stockpile” and “bum”.
6. Presidential Hackathon invites teams to propose open data solutions for delivering a better and more sustainable environment worldwide. The projects must address UN’s SDGs, and citizens use quadratic voting to select the projects. Nearly half of the population (10M) citizens vote. Initially people log in to vote for their friends’ project, but end up spreading their votes across 5–8 projects, often finding synergies. The five winners are given a presidential promise that their idea will be turned into national policy.

Now, to be fair, these are tiny improvements compared to what’s possible. But it’s remarkable to see a country actually implement modern governance tools, and satisfying to see them succeed.

If you want to learn more, start with Audrey’s interview on the 80000 Hours Podcast, and go from there. If you’re interested in experimenting with these technologies in San Francisco, get in touch.

N.B. Some snippets in this write-up are blatantly copied from other posts/articles. But I have no doubt that Audrey would support the informational re-use.