[d-morrison]

Requested by Ezra · project thread

Important

Merge Morrison-Lab/mds#815 first, and wait until the mds site is republished.
This PR's links point at #def-harmonic-mean and #exm-harmonic-mean on the mds vectors page, and those anchors exist only after that PR is published.

Closes Morrison-Lab/mds#814.

Why

The harmonic mean is general math, so its definition and example move to the mds vectors page, next to the mean, in mds#815.
By the course sequence (mds -> pds -> sds), sds links that definition instead of repeating it.
Ezra approved the move in the project thread on 2026-11-13.

What

• Deletes _subfiles/classification/_def-harmonic-mean.qmd and _exm-harmonic-mean.qmd, and their includes in classification.qmd.
• Points the five links on the classification page at mds:
  • the two exercises (_exr-pos-agreement-harmonic.qmd, _exr-neg-agreement-harmonic.qmd);
  • the two theorems, where (@def-harmonic-mean) becomes a link on the words "harmonic mean";
  • _exm-pos-agreement-harmonic.qmd, where @exm-harmonic-mean becomes a link to the mds example.
• Removes wikipedia_harmonic_mean from references.bib, because nothing else here cites it. mds#815 adds the entry to mds.
• The section heading "Positive and negative agreement as harmonic means" and its theorems stay. The heading now shares its slide with the first exercise.

The theorems and solutions still write $H(\cdot, \cdot)$. The symbol is now defined on the linked mds page, not on this page.

Checks

• quarto render classification.qmd --to html: renders. #def-harmonic-mean and #exm-harmonic-mean no longer appear on the page, and no cross-reference is unresolved (no ?@). The one warning is about code-link: the session lacked the downlit and xml2 packages, and this PR doesn't change that.
• typos reports only the existing thm id prefix, in lines this PR leaves unchanged.

🤖 Generated with Claude Code

https://claude.ai/code/session_019xfc9q9AYY3SdrG17Xxjqb

[github-actions]

Claude finished review — View run

---

The PR is ordered correctly: it depends on mds#815, and the mds site doesn't yet carry the anchors, which the PR description says up front.

Review of PR #619 (commit 151e8a7d0ec51f7e1ddddd7628bd5b8e2e4ff69f)

The PR moves the harmonic-mean definition and example to mds and replaces the five in-page references with absolute links to the mds vectors page.

Findings

Blocking until mds#815 is published: the link targets don't exist yet

• Affected files:
  • _subfiles/classification/_exr-pos-agreement-harmonic.qmd:5
  • _subfiles/classification/_exr-neg-agreement-harmonic.qmd:5
  • _subfiles/classification/_thm-pos-agreement-harmonic.qmd:6
  • _subfiles/classification/_thm-neg-agreement-harmonic.qmd:6
  • _subfiles/classification/_exm-pos-agreement-harmonic.qmd:33
• I fetched the live mds site. linear-algebra-vectors.html returns 200, but it has no #def-harmonic-mean or #exm-harmonic-mean. The word "harmonic" does not appear on the page.
• No page in the mds sitemap defines def-harmonic-mean.
• Until mds#815 is merged and the site is republished, these five links land at the top of the page and the lychee link check may fail on the missing fragments.
• This matches the dependency the PR description already states. Hold the merge until that is done.

Suggestion: $H(\cdot,\cdot)$ is now used without a definition on this page

• The PR description says the theorems and solutions still write H(x, y), and its symbol is now defined only on the mds page.
• The links on the words "harmonic mean" cover this. Check that the first use of H in the theorem and solution divs sits close to such a link, so the reader can find the notation.

Not a finding: other changes check out

• No remaining references to def-harmonic-mean, exm-harmonic-mean or wikipedia_harmonic_mean exist in the repo apart from the new mds links, so no cross-reference is left dangling.
• The classification.qmd comment about the slide break is updated to match the new structure.
• Removing the bibliography entry is safe because nothing else cites it.
• The PR adds no new math, so I skipped the macro check.

Verdict

Needs more work — the mds anchors the links point to are not yet published. Once mds#815 is merged and republished, the change should be ready.

Structured Review Data (JSON)

{
  "schema_version": "1.1",
  "reviewer": "claude",
  "commit_sha": "151e8a7d0ec51f7e1ddddd7628bd5b8e2e4ff69f",
  "verdict": "NOT_CLEAN",
  "findings": [
    {
      "file": "_subfiles/classification/_thm-pos-agreement-harmonic.qmd",
      "line": 6,
      "category": "doc",
      "message": "Links to mds linear-algebra-vectors.html#def-harmonic-mean, but the live mds site does not define that anchor yet. The same applies to the links in _exr-pos-agreement-harmonic.qmd:5, _exr-neg-agreement-harmonic.qmd:5, _thm-neg-agreement-harmonic.qmd:6 and the #exm-harmonic-mean link in _exm-pos-agreement-harmonic.qmd:33. Merge only after mds#815 is published."
    }
  ],
  "detailed_assessment": "The five rewritten links target the def-harmonic-mean and exm-harmonic-mean anchors on the mds vectors page. I fetched the live page and the whole mds sitemap, and neither contains those ids, so the links resolve only to the top of the page until mds#815 is published. No dangling local references remain.",
  "holistic_assessment": "The scope is tight and follows the course sequence, and the bibliography and slide-structure cleanups are consistent. The merge depends on the mds republish stated in the PR description, and the notation H now lives only on the external page, so the change is not ready until that dependency lands."
}

@claude address review comments in [this review](https://github.com/Morrison-Lab/sds/pull/619)

Reviewed commit: 151e8a7

[github-actions]

💰 Cost: $0.2513 (review) — run

[d-morrison]

Annotation dispositions for head 151e8a7:

• check / check-exm-figures: "Example #exm-pos-agreement-harmonic has no figure." Deferred, out of scope. The example had no figure on main either. This PR changes only one line in it, which repoints @exm-harmonic-mean to the mds example. Adding a figure is new content, and the project's scope rule puts it in a separate PR, not in this move.

The Claude review's NOT_CLEAN verdict on this head is the dependency the IMPORTANT alert already states: the mds anchors appear only after Morrison-Lab/mds#815 is merged and the mds site is republished. I'll ask for a fresh review once they are live.

---

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

---

Generated by Claude Code

[github-actions]

PR Preview Action v1.8.2
🚀 Preview available at: https://Morrison-Lab.github.io/sds/pr-preview/pr-619/

_{Built to branch gh-pages at 2026-11-13 21:19 UTC.}