docs: record the reduced-motion check — the hero half, and why that is not all of it - #85
Merged
Merged
Conversation
…s not all of it The owner checked reduced motion in Firefox on 2026-09-28 and the headline shows. That closes the oldest outstanding item in this workstream and, more importantly, **settles the hydration fix**: the mismatch is what left `opacity: 0` stuck inline, so a visible headline means server and client now agree. **It is deliberately recorded as half, not all, and the distinction is mechanical rather than pedantic.** The headline, subtitle and both calls to action come from `Hero`, which animates *on mount*. The featured course cards and the four "what you get" panels come from `Reveal` and `StaggerList`, driven by `whileInView` and an `IntersectionObserver` — and those were among the nine elements that blanked. A visible headline is evidence about the mount path and says nothing about the scroll-triggered one. Scrolling down with the setting still on closes it in about twenty seconds. Recorded in four places, because a verification that lives in one file is one somebody re-does: the runbook's verification log gains its first row and its §3.1 is rewritten; ADR-030 §6 no longer claims the fix has never been observed; roadmap A1 is marked half done with the remaining half stated precisely; and STATUS's unconfirmed table drops from "four, none looked at" to "four, one half confirmed". Also swept the four older places in STATUS that still asserted the check was outstanding — a stale claim in a status file is worse than no claim, because it is read as current. And ADR-030 §4 no longer lists 3.3.8 and 2.5.8 as *proposed* additions; both shipped in #80, and only the 2.4.11 test still waits on A2. Browser version not recorded, which the runbook asks for. Noted rather than invented. Docs only. 455 frontend tests unchanged; every ADR id and doc path resolves. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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 owner checked reduced motion in Firefox on 2026-09-28 and the headline shows. That closes the oldest outstanding item in this workstream and, more importantly, settles the hydration fix: the mismatch is what left
opacity: 0stuck inline, so a visible headline means server and client now agree.Docs only, four files.
Recorded as half, and the distinction is mechanical rather than pedantic
Hero— animates on mountReveal/StaggerList—whileInView+IntersectionObserverBoth groups were among the nine elements that blanked. A visible headline is evidence about the mount path and says nothing about the scroll-triggered one — so scrolling down with the setting still on closes it, in about twenty seconds.
I'd rather record this precisely than book the whole item as done, because a verification log that overstates what was seen is worse than none.
Recorded in four places, deliberately
A verification that lives in one file is one somebody re-does:
Also swept
Four older places in STATUS still asserted the check was outstanding. A stale claim in a status file is worse than no claim, because it's read as current.
And ADR-030 §4 no longer lists 3.3.8 and 2.5.8 as proposed additions — both shipped in #80, and only the 2.4.11 test still waits on A2.
One gap in the record
Browser version not captured, which the runbook asks for. Noted rather than invented.
455 frontend tests unchanged; every ADR id and doc path resolves.
🤖 Generated with Claude Code