Polis: Open-source platform for large-scale civic deliberation

Overview and positioning

• Seen as an open-source consensus/participatory democracy tool, but some are annoyed the homepage doesn’t clearly link to the code despite “open source” branding.
• Several commenters are excited, referencing Taiwan’s experiments and Twitter/X “Community Notes” reportedly using similar algorithms.
• Others see it as essentially a “glorified online survey” that can at best distill proposals for later referenda.

Direct and liquid democracy

• Some want platforms like this to replace representative democracy with direct digital voting, arguing it would reduce lobbying and corruption.
• Others propose “assigned voting” / vote delegation chains (liquid democracy) so most people don’t need to vote on everything while retaining override rights.

Misinformation, extremism, and common ground

• Concern: how does this work when people believe “alternative facts” or hold dehumanizing views (e.g., wanting certain groups imprisoned or dead)?
• Reply: Polis-style mapping in Taiwan reportedly showed large areas of agreement among “median” participants; true extremists are a minority.
• Skeptics argue this consensus can be superficial (“cut government waste”–type platitudes) and that some priors are not revisable by deliberation alone.

Question design, bias, and manipulation

• Who writes the statements is seen as crucial: biased or innuendo-laden framing can poison results.
• One theory: include the full spectrum of positions so people gravitate toward slightly less-biased statements, revealing real overlaps.
• Critics worry unequal numbers of statements per “side” or buried/seeded items will still steer opinion, and that graphs of public sentiment can be used for targeted manipulation.

Identity, bots, and anonymity

• Major thread on spam/influence bots: suggestions range from invite trees, eID, proof-of-personhood / “soulbound” identity, to anonymous authentication via zero-knowledge proofs.
• Tension: anonymity is essential under authoritarianism, but anonymity also makes bot filtering harder.
• Some argue we may have to accept bot participation and rely on strong moderation (possibly AI-assisted).

Comparison with social media and moderation

• Enthusiastic commenters imagine Polis-like tools as an antidote to engagement-optimized social media that amplifies conflict.
• Value is seen in “atomic” statements and clustering by agreement rather than by outrage.
• Proposed quality controls: invite-only networks, karma thresholds, cooldowns, stricter bans/timeouts, text-quality heuristics, and non-binary voting reactions.

Scope, use cases, and gamification

• Use cases discussed: city planning, homeowner/strata associations, local ordinances, and exploratory opinion-mapping rather than binding lawmaking.
• Some doubt it can “fix” structural issues like control of money or cult-like radicalization.
• Debate over engagement: one side says it must be gamified and appified to reach average citizens; another says it only needs to serve those who care, not chase maximum “engagement.”