W1.6: landing converged on new shuffle reference - expanders 7→0, text -21% - #432
Conversation
…rs, cut text Paul's named complaint was "too many expand-to-read-more components" plus "avoid too much text / reorganize into better components." Converge the landing on the new shuffle2 reference (40.28) while keeping the approved W1.5 hero band, obsidian course-window card, centered container, and section-band rhythm. Expanders 7 -> 0: - Module map: 5 per-module <details> "N lessons" + the "Modules 3-5" <details> group -> 5 flat wide rows (kicker+title, ruby Deliverable, inline CHAPTERS list, "You leave with" side panel), matching reference R2.2. Cut off-landing (reachable on lesson pages + kit): per-lesson routers, optional-lesson skip-notes, template lists, per-module Start buttons. Kept reachable: lesson links, one compact Mia walkthrough line per module, deliverables + glossaries. - NOT-cover: 7-item list + "Why these are out" <details> -> light 2x2 scope cards + one residual line (reference R2.3, mapped to LIGHT not the ref's dark - the 3-dark-zone budget is spent on hero card + mistake CTA + endcap). 7 exclusions fold into 4 cards + residual; no fact deleted. Text cuts / component polish: - Mistake cards: bold scannable title lead-in per card (reference R2.1); module deep links preserved in the descriptions. - Intro: dropped the redundant 5-module narrative wall (the module map now carries it); kept the Founding Hypothesis gloss + 2-3 week checkpoint. Scope: content _index.md + course-list.css only (no shared/theme CSS, no JS). Two style.css !important leaks neutralized in the landing bundle (the green module-leave checkmark; the h3 divider margins). Test: desktop interaction test no longer clicks the removed #module-1 <details> summary - clicks the now-inline lesson link. Desktop macos landing baseline re-recorded (intentional). Mobile baseline unchanged (that test captures only the hero fold, which is untouched). Gates: bin/hugo-build 8/8 validators pass; bin/qtest green; zero console errors (pre-existing webmanifest CORS quirk only); no mobile overflow at 390px. Side-by-side vs reference: 40.29-w16-result-full.jpeg. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
Warning Review limit reached
Next review available in: 21 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Plus Run ID: ⛔ Files ignored due to path filters (3)
📒 Files selected for processing (4)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
Honest bin/dtest (main checkout, detached at branch tip): clean — the only failures are the 7 known mobile-codeblock QEMU emulation diffs (deterministic on Apple-Silicon Docker, green on CI-native amd64). Zero landing regression. Note: the landing visual-regression test is hero-fold-scoped and W1.6 preserved the W1.5-approved hero, so the below-fold redesign (module map / mistakes / scope) passed the pixel suite; that below-fold rendering was covered by the chrome-devtools scroll gate + the 40.29 side-by-side instead. Ready to merge on Paul's taste verdict — the open call is the module-map density (flat-all-chapters as shipped, or one-more compaction pass to match the reference's brevity). |
…dule-map-compaction lever noted Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
1 participant
W1.6: landing converged on the new shuffle reference — expanders gone, text cut 21%
Answers Paul's directive: "make the landing look much closer to [the new shuffle reference] and avoid too much text / reorganize into better components," plus the standing "too many expand-to-read-more" complaint. Landing-owned files only (content
_index.md+course-list.css); no shared CSS, no JS — campaign-window safe.Headline results
<details>in.post-content). The named complaint, fully resolved.What changed (components from the new reference, JetVelocity-mapped)
<details>→ 5 flat scannable rows (kicker + title + ruby Deliverable + inline chapters + "You leave with" panel + timeline dots).Evidence
40-49-review/40.28-reference-shuffle2-full.jpeg(left) vs40.29-w16-result-full.jpeg(right).Guards (all verified live)
8 inbound anchors present; both stub H2s byte-identical; hero verdict chip + endcap "Start Lesson 1.1 →" intact; no 390px horizontal overflow; ≤3 dark zones (NOT-cover kept light on purpose — the 3-dark budget was already spent). Two
style.css !importantleaks neutralized inside the landing bundle (out-of-scope files untouched). The landing journey test updated (it clicked a removed<details>summary → now clicks the inline lesson link) and passes.⚑ One open taste call for Paul (the reason the page is still ~2× the reference height)
The reference is short partly because it does not list every chapter on the landing — it shows compact module summaries only. This PR flattened all 25 chapters into visible rows (that's what killed the expanders). Net effect: expanders gone, but the module map is now a long flat list.
Recommendation: to fully match the reference's brevity and your "less text" directive, compact the module map to module summaries + a chapter count, with chapters reachable via each module's lesson links / the syllabus — rather than every chapter inline. That's a follow-up if you want it; it's a genuine "show the full syllabus vs. keep it short" tradeoff that's yours to call. The other un-named length levers (relocating "Take this course if" / "Who built this" off the landing) are also available.
Gates: hugo-build 8/8 + clean; qtest green; macos desktop-landing baseline re-recorded in-commit; honest bin/dtest from the main checkout (result in a PR comment). Content/landing change — Paul's taste gate decides ship vs. one-more-compaction.
🤖 Generated with Claude Code