docs: market decentralization design (the §10.1-B shape)#16
Open
leyten wants to merge 1 commit into
Open
Conversation
…e market The concrete shape for NETWORK_ARCHITECTURE §10.1 option B (market-as- optimizer, decided): posted per-layer-token asks, adequacy-floor pricing of the slow-node externality, deterministic auditable ring formation around the unchanged shard.plan, RingCharters, chunked client-ack settlement that closes the token-count gap, DHT/gossip discovery on the sidecar, and a 6-stage migration where each stage is locally testable and visibly decentralizing (flagship: a swarm forms with the orchestrator dead). Names the honest central residue and the hard open problems (emissions wash-trading first).
This was referenced Jul 9, 2026
leyten
added a commit
that referenced
this pull request
Jul 12, 2026
…ider (#21) Panel of three design angles + adversarial verification, written to the skeptic's rework list: capability records are hints (re-measure at formation), member verify-and-sign is the load-bearing check, demand-side signature is the legitimacy anchor. Canonical shard.form determinism contract (reference-impl + cross-version goldens; the PYTHONHASHSEED test would have been a no-op), three-way measurement sourcing split, warming-abandonment as the real formation cost, gapped-settle kept OUT of v0 (accuser-forgeable evidence), equivocation stated as detectable-not-slashable. M0->M4 migration with M1 scoped honestly (function-honest, not input-honest). leyten forks in §10, incl. the #16 market handoff points.
leyten
added a commit
that referenced
this pull request
Jul 12, 2026
…rit the money-flow (#24) Decisions (2026-07-12): (1) earnings = inherit c0mpute's existing 70/30 demand-funded money-flow, realized per-shard (pay-by-layers #16), USDC via the existing worker-earnings ledger — no new points economy, no supply subsidy; hard gate = bind pay to a verified token count before USDC. (2) FULLY PERMISSIONLESS from day one — corrects the earlier gated-phase-1 lean. The admission we built is ALREADY permissionless (capability function, never an identity gate); sybil defense is per-job economics (rake) + verification (spot-check), NOT admission-gating, so key-rotation largely dissolves as a threat. Reputation = non-load-bearing (spot-check frequency + seat preference on measured misbehavior only); stake gates nothing at launch (privacy tier decided-off). Launch = placement M1 (verifiable-centralized), M4 is the road not day one. Consequence: leg 8.7 (spot-check scheduler + auditor) is now day-one critical path. Reconciled the io.net lesson: the hole was paying for claimed capacity, not open entry — open + measured-pay avoids it.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The concrete design for the decided market end-state — how the central orchestrator becomes a compute market, staged so each step is locally testable and visibly decentralizing.
Core moves:
shard.plan+ signed RingCharters — members verify-and-sign instead of obeying; no auctioneer, no consensus anywhere.paid = min(claim, client ack)) — closes the known token-count gap and deletes the central settler; bounded exposure, no escrow needed for the PoC.6 stages; flagship demo at stage 3: kill the orchestrator, a swarm still forms and serves.
Leaving this OPEN for review before any market code is built — it shapes product direction.