Nominations are open for the evaluation committee of the PQC Migration Prize. Nominations are collected now; nothing is decided until W1 ratifies the composition table, and seats are filled at W2.
How: the Committee nomination issue template at Sign in to GitHub · GitHub, or a reply here and it will be carried over.
What the committee does
Two evaluation tracks, following the NIST-style split named in the GDC26 report:
- Theoretical security review — assumptions, parameter choices, composition (hybrids, aggregation, ZK wrapping), key-management risk, side-channel posture, migration-window threat model. Written review against a checklist; no testbed run needed.
- Implementation and performance — measured cost and behaviour on the neutral testbed, reproduced across nodes.
Both tracks need people. So does the work before either starts: the metric register, the rubric weights, and the scheme scope rulings are committee decisions, and they happen in the next three months.
What is being looked for
The composition table in GOVERNANCE.md §3 is a draft: 12–18 seats across academic cryptography, protocol engineering on Bitcoin and Ethereum, wallet and custody, measurement and testbed operations, and standards. It is a draft, and arguing with it is a legitimate way to respond to this post. The quorum rule is open too.
Self-nomination is normal and expected.
What a nomination involves
A nomination records the lane proposed, relevant experience, and declared interests. Anyone seated files a conflict-of-interest disclosure before taking part in evaluation work — the policy is public at evaluation/COI_POLICY.md; the disclosures themselves are not.
That is the general shape: decisions are published, deliberation is not. A seated member’s roster entry carries only what that person agreed to publish, which is asked at nomination. The fact that a member recused, and what from, is published; the disclosure behind it is not. Nominations that are declined are kept on the record with the reason, because a competition that discards that record cannot show it was neutral.
Conveners
Conveners are named on ratification at W1. That is one of the things W1 is for. Before W1, one seat is needed immediately: a second convener with a signing key, because the repository’s review rule (a code-owner approval that is not the author’s, no bypass) means a single convener cannot merge anything — including the committee’s own nominations. Setup is docs/SIGNING.md §1–3; it takes an afternoon.
Names for the two track leads — theoretical security, implementation and performance — are the next ask, and nominations for them are read first.