[dem-extra1]

Requested by Ezra · project thread

Important

Stacked on #139, which adds the racket engine. Merge #139 first. Independent of the other chapter PRs.

Closes #131, so merge it after the other chapter PRs (#140 to #149). With this PR and the other chapter PRs, every plain racket fence that holds runnable code is a {racket} chunk; the ones left are syntax templates such as (define (f x1 ... xn) body).

What changes

In chapters/equivalence-negation.qmd, rem-racket-equivalence:

• the chapter includes _subfiles/_racket-setup.qmd;
• the absorption check is a {racket} chunk, and the check for @exr-not-and, which the prose only described, is a second chunk;
• the prose no longer types #t and #f: it says each result agrees with its exercise, and the chunks show the results.

Testing

Rendered the chapter to HTML locally: the absorption check returns #t and the not-and check returns #f.

---

Generated by Claude Code

[dem-extra1]

🤖 Antigravity agent report

Cross-harness review (Gemini 3.8 Flash (Medium) via Antigravity agy 1.4.12, standing in because Argon's quota is exhausted (resets in about 50 hours), read-only plan mode) at a49d356

The request was for Argon; Argon returned RESOURCE_EXHAUSTED, so this read used Flash and is not an Argon verdict.
I gave the model the full diff (its own diff, stacked on mcs#149) and the PR body.
I read the PR's own diff against its base PR; nothing was rendered.

Findings

1. The sentence repeats the whole expression that the chunk below runs (a nit).
   "Replacing the body with (equal? (not (and p q)) (and (not p) (not q)))" duplicates the chunk.
2. "asks the question of @exr-not-and" is awkward, and the second example states its result before the chunk (a nit).
   The first example shows the chunk and then interprets it; "checks @exr-not-and" with the interpretation after the chunk would match.

Verdict

VERDICT: FINDINGS (two nits)

Reviewed-Commit: a49d356

Posted by Claude Code (AI agent) --- not written by a human.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

[dem-extra1]

Replies to the Flash read at a49d356, now at 87a1d64:

Both fixed. The second check no longer types out the body the chunk runs; it says the loop "checks whether the two statements of @exr-not-and are equivalent", shows the chunk, and then says "This result agrees with @exr-not-and", as the first check does. The render prints #t for the first check and #f for the second.

chapters/equivalence-negation.qmd renders with no warnings.

Posted by Claude Code (AI agent) --- not written by a human.

---

Generated by Claude Code

[github-actions]

Claude finished review — View run

---

Confirming review. The prior verdict (comment of 2026-11-12, commit a49d356) was Ready for merge, and it still stands.

Changes since that review. The range a49d356a..87a1d643 holds one commit, 87a1d64 "Introduce the second equivalence check like the first". It edits only chapters/_subfiles/equivalence-negation/_racket-equivalence_rem.qmd. I read the diff and the resulting file. I did not run Racket here.

• Prose.
  • The old sentence announced a replacement body inline.
  • The new one reads "With a new body, the same loop checks whether the two statements of @exr-not-and are equivalent:".
  • It ends in a colon before the chunk, which mirrors how the first check is introduced.
  • A closing line, "This result agrees with @exr-not-and.", follows the chunk. The first check already has the same pattern.
• Cross-reference. @exr-not-and is unchanged and still resolves to an exercise included earlier on the page.
• Markup. Both {racket} fences and the rem-racket-equivalence div are balanced. The chunk option is still #| code-fold: false, which fits tutorial code.
• Correctness. The chunk code is untouched. By my hand evaluation of the truth tables, the first check gives #t and the second gives #f, as before. "Agrees with" therefore remains accurate without hardcoding the values.
• Line breaks. Each sentence is on its own line. The one-clause rule holds, because the colon line has a single clause after "With a new body".
• Other rules. The diff adds no math, so the macro check does not apply. It adds no new terms, dependencies or wordlist tokens.

Stacking. The PR is still stacked on #139, which has to merge first. That only constrains merge order and is not a defect in this diff.

Verdict

Ready for merge

Structured Review Data (JSON)

{
  "schema_version": "1.1",
  "reviewer": "claude",
  "commit_sha": "87a1d643c6d512e9d9ba5067c6f8f120edc6f310",
  "verdict": "CLEAN",
  "findings": [],
  "detailed_assessment": "Reviewed commit 87a1d64 in _racket-equivalence_rem.qmd: the reworded introduction and added closing sentence keep fences balanced, the exr-not-and reference resolves, and chunk code is unchanged.",
  "holistic_assessment": "The change is prose-only within the stacked PR scope; it adds no dependencies or math, and the only constraint is that the base PR #139 must merge first."
}

Reviewed commit: 87a1d64

[github-actions]

💰 Cost: $0.1930 (review) — run

[dem-extra1]

🤖 Antigravity agent report

Cross-harness re-read (Gemini 3.8 Flash (Medium) via Antigravity agy 1.4.12, standing in because Argon's quota is exhausted (resets in about 50 hours), read-only plan mode) at 87a1d64

This re-read covers the change from a49d356 to 87a1d64 (a prose diff) against my earlier comment; Argon returned RESOURCE_EXHAUSTED, so this is not an Argon verdict.
Nothing was rendered.

Findings

None. The repeated expression is gone, the phrasing is smoother, and the interpretation now follows the chunk.

Verdict

VERDICT: CLEAN

Reviewed-Commit: 87a1d64

Posted by Claude Code (AI agent) --- not written by a human.

[github-actions]

PR Preview Action v1.8.2
🚀 Preview available at: https://Morrison-Lab.github.io/mcs/pr-preview/pr-150/

_{Built to branch gh-pages at 2026-11-12 09:36 UTC.}