Step 1: Solution Assessment — Board List Column Reflow

Problem statement

At half-wide window widths, Board's fixed-width From (9rem) and Date (13rem) columns are often wider than their content needs — especially Date, now that relative timestamps ("2 days ago") are much shorter than the old full-format text it was sized for — leaving empty space while the flexible Subject column gets squeezed.

forte_mobile_friendly_step1_solution_assessment.md (pending approval) targets a narrower failure: Subject collapsing to unreadable widths (~16px) at ~400px, fixed with an overflow-x scroll fallback. This report is a different symptom at wider ("half-wide") widths — wasted space, not unreadable squeeze — so it's written as its own assessment; Step 3 can decide whether the two share an implementation.

Options

  • Option A — Re-tune the fixed column widths

    • Pros: Smallest change; one CSS edit re-sizing .paned-list-date-head/.paned-list-date (and .paned-list-from-head/.paned-list-from if needed) to fit the new shorter content.

    • Cons: Still a static guess — wastes space differently at other widths, and drifts out of sync again if date/label text length changes (e.g. "11 months ago" vs "2 days ago"); doesn't help From when usernames are short.

  • Option B — Size From/Date/Replies/Score to their content (intrinsic width) instead of a fixed rem value

    • Pros: Each column takes only what its actual content needs at any width, automatically freeing the rest to Subject; no magic numbers to re-tune when content length changes.

    • Cons: Needs a per-column cap so a single unusually long value doesn't blow out the column (e.g. clamp(min, ch-based, max)); still layered on the current flexbox row, so behavior is a bit more implicit than a single layout mechanism.

  • Option C — Move the row layout from flexbox to CSS Grid with content-sized tracks

    • Pros: One layout mechanism where Subject is the single flexible track and every other column is max-content-sized — solves the reported wasted-space complaint directly, and is a natural base to later layer the forte_mobile_friendly narrow-viewport fix on top of (same track definition, just added a min-width floor + scroll fallback below it).

    • Cons: Larger touch than A/B — changes how Board's list head/rows are laid out, not just column widths, so it needs its own regression pass (Users/Activity untouched since this is Board-scoped).

Recommendation

Option B, scoped to Board only. It solves the reported complaint (content-driven widths instead of static guesses) without changing the underlying layout mechanism, keeping risk and review surface small. Option C is the more thorough fix and worth revisiting if/when forte_mobile_friendly is approved and the two efforts get reconciled, but is more than this specific complaint needs right now. Option A is rejected as a re-run of the same static-guess problem that caused this report.

Share this document for review — Approved Step 1 is required before Step 2.