Repository source: docs/plans/backlog/page_asset_performance_step1_solution_assessment.md
Page Asset Performance — Step 1: Solution Assessment
Problem Statement
After theme splitting, standard pages still ship shared CSS, scripts, inline critical CSS, and static-artifact assets that may not serve their route; reduce transfer and parse cost without minifying source assets.
Option A: Extract About as a page-CSS pilot
Move the About-only .about-section rules into an About stylesheet and add per-page CSS support to the standard layout.
Pros:
Small, low-risk proof of the shared page-asset mechanism.
Immediately stops non-About routes loading About glyphs and layout rules.
Cons:
Leaves the larger CSS, script, and artifact opportunities untouched.
Repeated one-page changes can create inconsistent asset ownership.
Option B: Route-family CSS and JavaScript ownership pass
Create shared route-family asset contracts and extract every selector/script whose template use is exclusive to that family.
Pros:
Captures About plus the current page-specific CSS and layout-wide no-op scripts.
Keeps assets readable and cacheable without a bundling/minification system.
Cons:
Requires a complete ownership inventory and route regression coverage.
Extra requests can outweigh savings for very small assets if grouped poorly.
Option C: Measured non-minification performance program
Combine route-family assets with critical-CSS reduction, HTML/payload review, deferred heavy tooling assets, and static-artifact/deployment hygiene, guided by page-size and request measurements.
Pros:
Addresses CSS bytes, HTML bytes, JavaScript execution, and artifact bloat together.
Makes prioritization evidence-based and protects each improvement with budgets.
Cons:
Too broad for one implementation pass; should be delivered as focused follow-up stages.
Some gains depend on server/CDN configuration outside application code.
Recommendation
Option C, starting with Option B's route-asset contracts and the About pilot. The companion inventory identifies the complete current source-level opportunity set; use measurements to select the first small implementation feature, beginning with shared per-page CSS support and about.css.
Inventory: non-minification performance inventory.
Reply Approved Step 1 to proceed to the feature description.