pairwise-primary-sources-2026-05-11

Pairwise primary sources

Collected: 2026-05-11 UTC

Primary sources reviewed

• https://www.pairwise.vote/
  • Official homepage
  • Used to confirm the project’s self-description as a governance / funding-decision tool built around pairwise comparisons, delegation, and BudgetBox aggregation
• https://app.pairwise.vote/
  • Live app landing page
  • Used to confirm the product was deployed as a real voting interface for Retro Funding 6 and retained an explorable application surface after the round closed
• https://paragraph.com/@pairwise/pairwise-in-retro-funding-round-5-your-voting-tool
  • Official Pairwise post for RF5
  • Used to confirm the RF5 role as official ballot-building tooling, plus the star-rating layer used before finer pairwise ordering
• https://paragraph.com/@pairwise/empowering-superchain-with-pseudonymous
  • Official Pairwise architecture post for RF4
  • Used to confirm the pseudonymous voting stack: Thirdweb AA wallets, Semaphore / Bandada proof flow, sponsored transactions, IPFS ballot payloads, and EAS attestations
• https://paragraph.com/@pairwise/pairwise-retro-funding-round-6-liquid-democracy-experiment-1
  • Official Pairwise post for RF6
  • Used to confirm category-specific liquid-democracy delegation, Farcaster-based delegation paths, and the project’s move from ranking UX toward expertise-routing middleware
• https://paragraph.com/@pairwise/deep-funding
  • Official Pairwise post on Deep Funding
  • Used to confirm that Pairwise also serves as a pairwise-comparison layer for jury spot-checking and human calibration inside AI-assisted capital-allocation systems
• https://github.com/metagov/interop
  • Metagov interop grant repository
  • Used to confirm an earlier public framing of Pairwise as an open-source offchain voting dapp similar to Snapshot, focused on two-option comparisons and result aggregation

Notes

• Pairwise cleared the bar because it is not just another governance frontend; it exposes a distinct upstream control layer where category design, comparison sequencing, star filtering, delegation paths, anonymity plumbing, and ranking aggregation shape outcomes before the final treasury or grants settlement surface.
• The most useful comparison is not to base voting chains but to ballot-construction middleware. Pairwise sits closer to Snapshot-like preference interfaces, but adds a more opinionated decision pipeline: pairwise micro-judgments, optional pre-bucketing by stars, and BudgetBox-based ranking output.
• The pseudonymous RF4 architecture is especially worth keeping because it decomposes account abstraction, reputation import, Semaphore-group identity proof, IPFS ballot storage, backend badge mapping, and EAS publication into separate layers instead of flattening the tool into a generic anonymous voting label.
• The RF6 delegation experiment is also analytically important because it makes Pairwise a routing layer for expertise and social trust, not only a ranking interface. Category-level delegation via Farcaster identities turns ballot construction into a partial liquid-democracy surface.
• The Deep Funding use case matters because it shows Pairwise can function as a human-judgment checkpoint inside larger model- or graph-assisted capital-allocation systems, not only inside one DAO’s internal voting workflow.
• One historically useful tension across sources is the shift from an early public framing as an offchain open-source voting dapp to later official emphasis on onchain attestations and pseudonymous wallet flows. That evolution helps separate the core mechanism (pairwise ranking middleware) from the changing publication / identity stack wrapped around it.
• This entry belongs in the active corpus because it gives the governance-tooling and public-goods-funding clusters a better comparison point for upstream ballot construction, delegation routing, and ranking middleware — surfaces that often matter as much as the final onchain vote or payout contract.