Skip to content

Make the models page readable at a glance, and give the list motion - #191

Merged
vahid-ahmadi merged 1 commit into
mainfrom
redesign/models-page-first-view
Aug 26, 2026
Merged

Make the models page readable at a glance, and give the list motion#191
vahid-ahmadi merged 1 commit into
mainfrom
redesign/models-page-first-view

Conversation

@vahid-ahmadi

Copy link
Copy Markdown
Contributor

The problem with the first view

The first thing on the page was the core model's accordion, open, with four dense bullets. The actual map of the suite — the seven one-line rows telling you what each model answers — sat below them, mostly below the fold.

So the first view was the details of one model rather than the shape of seven.

Closed the core. The first view is now the map: seven rows, each a question and a model id, with the detail one click away where it was always meant to be.

Less text

The hero lede and six of the wordiest facts are cut, keeping every number. The £50,000 example no longer itemises allowance, income tax and NI on the way to the net figure — the net figure was the point.

Motion

Rows animate in sequence rather than as one block, so the eye reads the list as seven choices rather than a wall.

  • --qa-i restarts per .qa-list, so the Combine list doesn't inherit the Models list's delays and arrive half a second late
  • hover and open states move the border
  • the + rotates into a × on open

The existing prefers-reduced-motion rule disables all of it.

Cache key

Bumped v=4v=5 across 63 files, generator template included — style.css changed, and #189 is the lesson about what happens when that's skipped.

Testing

Site suite 1564 passed; contract and topic-page generator clean.

🤖 Generated with Claude Code

https://claude.ai/code/session_01Lcj9DDqam9KmVCfhEdnJcJ

The first thing on the page was the core model's accordion, open, with
four dense bullets. The actual map of the suite -- the seven one-line
rows that tell you what each model answers -- sat below them, mostly
below the fold. So the first view was the details of one model rather
than the shape of seven.

Closed the core. The first view is now the map: seven rows, each a
question and a model id, with the detail one click away where it was
always meant to be.

Text: the hero lede and six of the wordiest facts are cut, keeping every
number. The £50,000 example no longer itemises allowance, income tax and
NI on the way to the net figure -- the net figure was the point.

Motion: rows animate in sequence rather than as one block, so the eye
reads the list as seven choices. --qa-i restarts per .qa-list so the
Combine list does not inherit the Models list's delays and arrive half a
second late. Hover and open states now move the border, and the "+"
rotates into a "x" on open. The existing prefers-reduced-motion rule
disables all of it.

Cache key bumped v=4 -> v=5 across 63 files, template included: style.css
changed, and #189 is the lesson about what happens when that is skipped.

Site suite 1564 passed; contract and generator clean.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Lcj9DDqam9KmVCfhEdnJcJ
@vercel

vercel Bot commented Aug 26, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
macromod Ready Ready Preview Aug 26, 2026 2:32pm

Request Review

@vahid-ahmadi
vahid-ahmadi merged commit 407501a into main Aug 26, 2026
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant