The Seat Allocation Protocol (Schedule C, Orbital Habitats Compact)

SCHEDULE C — ALLOCATION OF CONSULTATIVE COUNCIL SEATS

Standing instrument of the Orbital Habitats Compact. This Schedule governs every allocation of consultative-council seats in every member habitat. It is written to be performed by anyone and trusted because of that, not despite it.

---

PREAMBLE. Given the electors of a habitat, each holding graded preferences over the declared candidates, this Schedule distributes a fixed number of seats among those candidates: proportionally, reproducibly, and without ever creating or destroying a seat along the way. Every clause below serves one of those three words, and the Schedule names which.

C-1. The ballot. An elector grades each candidate they wish to grade. A grade cannot be negative; the ballot form is constructed so that the question cannot be asked. A ballot that grades every candidate at zero is refused and returned, not silently set aside — a null ballot is an answer, and the count is forbidden to invent what the elector declined to say. (Reproducibly: a count that quietly drops inputs cannot be re-performed from the record.)

C-2. Normalisation. Each accepted ballot is reduced to fractional weights summing to exactly one. Every elector therefore enters the count with precisely equal total weight, however many candidates they graded and however enthusiastically. (Proportionally: proportionality is between electors before it is between candidates.)

C-3. Aggregation, and the first proof. The normalised weights are summed per candidate. Before the count proceeds, it proves that the grand total equals the number of accepted ballots — no more, no less. If it cannot prove this, it halts. (Without creating or destroying: the conservation the Schedule promises at the end is checked at the beginning, and at every boundary between.)

C-4. The threshold. The seating threshold is one more than the whole part of electors divided by seats. Any candidate whose aggregated weight reaches the threshold takes a seat, and the surplus above the threshold is redistributed — proportionally, within each contributing ballot, across that elector's remaining candidates. Every transfer conserves exactly what it transfers. Weight is moved; weight is never minted and never burned.

C-5. Remainders. Seats the threshold could not fill are distributed to the largest remaining aggregates, one seat each, until none remain. (The Schedule prefers two plain rules performed exactly over one clever rule performed approximately.)

C-6. Ties. Wherever two candidates stand equal, the seat falls by enrolment-mark order — the fixed ordering of the candidate register — and never by lot, discretion, or the order names appeared on any ballot paper. Position in a list influences nothing in this Schedule; only accumulated weight orders candidates. The consequence is the Schedule's central promise: the same ballots always produce the same seats, on any desk, in any habitat, in any year.

C-7. Exactness. The count is performed in whole fractions, exactly, from first weight to final seat. Approximate arithmetic is prohibited — not as pedantry, but because the surplus transfers of C-4 multiply fractions upon fractions, and it is precisely there, where the numbers grow inconvenient, that approximation would do its damage. The ledger carries the long denominators. The count does not round until it is finished, and when it is finished there is nothing left to round.

C-8. The proofs, and the halt. At every boundary between phases the count proves, and records that it proved: that total weight still equals the number of accepted ballots; that every surplus transfer conserved its weight exactly; that seats allocated so far plus seats remaining equals seats requested; and that every tie so far broke by rule. These proofs are not removed when the count is performed in earnest. They are the count. On any failure the count halts, loudly, and reports the boundary at which it failed — because a count that fails honestly can be repeated, while a count that succeeds dishonestly poisons every count after it.

C-9. Verification. The result is published together with the full anonymised ballot record and this Schedule. Any elector may re-perform the count from the published record and must arrive at the same seats; a verifier who arrives anywhere else has found either their own error or the count's, and the Compact's electoral regulations treat the second possibility as the emergency it is. The Returning Officer's authority derives from this clause and not from the office: the officer is trusted because the officer is, by construction, barely necessary.

---

Registry note: Schedule C is filed with its own full revision history, per the record-keeping conventions the Survey Archive's doctrine states for holdings generally — a standing instrument being, among other things, a holding about how a habitat has agreed to count. On the family resemblance between this Schedule, that doctrine, and the commissioning gates, the Archive's Senior Archivist has written separately; this office takes no position, and notes only that she is right about the counting.