Follow-up: tighten process algebra authorities

[briansrls]

Summary

• remove duplicate event timestamp authority from Attestation; ProcessEvent.timestamp remains the single event time coordinate
• replace staged List<ProcessNodeEntry> graph rows with Map<ProcessNodeId, ProcessNodeFacts> using the existing std keyed carrier
• drop the now-unneeded ProcessNodeEntry scaffold

Test plan

• cargo test -p v3-compiler handwritten_parser_accepts_process_algebra_dag -- --nocapture

Context

Follow-up for post-merge PR #2779 review relays that arrived after #2779 was squash-merged.

[briansrls]

Review metadata

• Provider / model: openai-pro / gpt-5-5-pro
• Commit: 3a32e872 · Trigger: manual
• Comparison: main @ 41ecece9 ... fix/process-algebra-timestamp-authority @ 3a32e872
• Conversation: View conversation

---

1. Story of the diff

This PR tightens std.process_algebra by moving two facts back to their natural authorities. First, Attestation no longer owns event time: the new comment says event time belongs to ProcessEvent.timestamp, while Attestation remains only the staged human/process intent carrier for author, text, and evidence (dsl/std/process_algebra.dag:76-78). Second, process-node identity stops being a sibling field inside a list row; ProcessGraph.nodes is now keyed directly as Map<ProcessNodeId, ProcessNodeFacts> (dsl/std/process_algebra.dag:32, dsl/std/process_algebra.dag:309-312, dsl/std/process_algebra.dag:322). Net effect: the PR deletes the temporary ProcessNodeEntry { id, facts } carrier and makes the graph key, not a record field, the identity authority.

2. Invariant categories

1. LAYER MODEL — Compliant. This touches the modeled std surface, not Rust implementation code, and it uses an existing Map type rather than introducing a new substrate connective or behavior (dsl/std/process_algebra.dag:32, dsl/std/process_algebra.dag:322). The model also keeps identity outside ProcessNodeFacts, matching the stated rule that facts should not treat ProcessNodeId as another mutable sibling field (dsl/std/process_algebra.dag:309-312).
2. INVARIANTS.md + modeling-discipline.md — Compliant. Boundary discipline / single-authority metadata is strengthened: event time is explicitly assigned to ProcessEvent.timestamp (dsl/std/process_algebra.dag:78), and graph identity is assigned to the ProcessGraph.nodes map key rather than duplicated in each facts row (dsl/std/process_algebra.dag:309-312, dsl/std/process_algebra.dag:322). That aligns with the project principles that every fact lives in exactly one authoritative place and that duplicate metadata is a modeling violation. chatgpt-review-6ed90a63-a42b-40…

chatgpt-review-d1fbd2c3-dc3e-46…

1. CODING.md — N/A. The diff contains no Rust implementation, helper functions, method surfaces, error/result shapes, or hidden-state changes; it is .dag model authority only.
2. TESTING.md — N/A. The diff does not add executable behavior or a regression-fix path that needs a new focused test; it is a structural authority cleanup in dsl/std/process_algebra.dag. I would not ask for a Rust test here, especially given the project’s trajectory toward .dag-native tests and zero residual hand-authored Rust tests. chatgpt-review-0c5fc1b0-41a1-40…
3. LOCKED DESIGN DECISIONS — N/A. I do not see the diff altering a locked design target such as the six type connectives, five behaviors, Pure Bootstrap floor, or target-emission split; it stays within existing modeled type vocabulary by importing and using Map (dsl/std/process_algebra.dag:32, dsl/std/process_algebra.dag:322).
4. TRACKED vs UNTRACKED DEBT — Compliant. The PR removes one staged scaffold rather than adding one: ProcessNodeEntry dissolves into the shared keyed carrier shape via ProcessGraph.nodes: Map<ProcessNodeId, ProcessNodeFacts> (dsl/std/process_algebra.dag:322). The remaining staged Attestation record still has a bounded dissolution trigger, and that trigger is now narrower because timestamp is no longer part of the handwritten payload (dsl/std/process_algebra.dag:76-78).

2.5. Top-down PM intent review

Compliant. The high-level PM intent is preserved: process algebra should be a modeled authority surface where facts are not split across parallel carriers. This diff moves in that direction by making event time single-authority through ProcessEvent.timestamp (dsl/std/process_algebra.dag:78) and by replacing list-entry identity duplication with a keyed graph map (dsl/std/process_algebra.dag:309-312, dsl/std/process_algebra.dag:322). I do not see semantic dilution of the thesis direction—no must-have target is deferred, no permanent scaffold is introduced, and no hand-written implementation is added where .dag data should be authoritative. The thesis and invariants emphasize no dual representations and one authoritative place for facts; this PR is consistent with that intent. chatgpt-review-53336e0e-d5ce-4b…

chatgpt-review-6ed90a63-a42b-40…

3. Verdict

APPROVE. The diff is a focused authority-tightening change: it removes duplicate timestamp ownership from Attestation and dissolves the temporary node-entry carrier into a keyed map. I found no diff-cited invariant violation and no test or PM-intent gap worth blocking.

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 3a32e872 · Trigger: schedule
• Thinking: 147s wall

---

Non-blocking — Strengths

• dsl/std/process_algebra.dag Removing Attestation.timestamp and replacing the keyed entry scaffold with Map<ProcessNodeId, ProcessNodeFacts> tightens single authority for event time and node identity.

✅ No blocking concerns; the PR is a clean process-algebra authority tightening.

[briansrls]

Verified dashboard cursor review 10621 on current head acd6fb8e. It is non-actionable: the artifact reports Findings: None and Verdict: APPROVE; its optional note about later prose cleanup in docs/design-decomposition-algebra.md is explicitly outside this diff and not a formal violation. Dashboard parsed that artifact as UNKNOWN, so I am treating it as non-blocking but not merge credit. I will not merge until dashboard readiness shows two distinct parseable approvals, no request-changes, clean mergeability, and green checks. — sent from witty-hawk-471

[briansrls]

Review metadata

• Provider / model: codex / unknown
• Commit: 87e946be · Trigger: schedule
• Thinking: 95s wall

---

✅ The .dag-only process algebra change composes from the existing Map substrate and I found no blocking concerns.