Know who is selling you bad science.

Health misinformation kills people. That is not hyperbole. Anti-vaccine rhetoric, cancer cure scams, and supplement grifters have measurable body counts. The people spreading this misinformation often have credentials (or claim to), platforms with millions of followers, and financial incentives to keep going. The people debunking them are scattered across journals, YouTube channels, and Twitter threads with no centralized resource.

Quack or Snack exists to change that. The project has three goals: raise awareness of who the pseudoscientists actually are and what they are claiming, drive real discussion about how we as a society police disinformation (what works, what does not, and where the lines should be), and have a little fun doing it. The duck puns and the voting system are not decoration. They are the mechanism that gets people to actually engage with the evidence.

The "Quacks" side profiles over 100 documented figures with danger ratings (low through critical), credential analysis (legitimate, fake, or misleading), claim-by-claim debunking (false, misleading, or unproven verdicts), and links to published takedowns from journalists and scientists. The "Snacks" side profiles 68+ science communicators, educators, and journalists who fight misinformation, with the same structured format: who they are, what they do, and why their work matters.

The annual Quacky and Snacky Awards add a community layer. Swipe-based voting (10 votes per day, fingerprint and IP rate-limited) produces real-time leaderboards. Top 16 per category advance to finals in December. The gamification is deliberate: making people engage with the profiles, read the evidence, and form opinions is the point. The voting is the mechanism, but the awareness and the discussion are the product.

Defamation risk is the elephant in the room. Every claim on the platform is sourced, every danger rating is editorially assigned, and the methodology is published. But profiling real people as "quacks" with "critical" danger ratings invites legal attention regardless of accuracy. The platform operates in the space between protected speech and actionable defamation, and the line between those is jurisdiction-dependent and expensive to test.

The content pipeline is manual. Each profile requires research, source verification, and editorial writing. At 100+ quack profiles and 68+ hero profiles, the catalog is substantial but nowhere near comprehensive. The pseudoscience ecosystem is enormous and constantly producing new actors. Scaling the editorial process without sacrificing rigor is the core operational challenge.

The voting system creates its own dynamics. Popular science communicators with large followings will naturally accumulate more Snacky votes. Controversial quacks with vocal supporters might receive defensive votes. Whether the vote totals reflect genuine community assessment or just audience size is an open question. The 10-vote-per-day cap helps, but does not eliminate, the influence of platform size on outcomes.

The Vercel KV backend handles current traffic, but if a profile goes viral (either because a quack's followers brigade the site or because a science communicator shares their profile), the rate limiting and vote integrity systems face their first real stress test.