Benchmark Radar Day 16: Counting Accuracy and Progressive Disclosure
Published:
Single-adopter benchmarks were under-counted. Drill-downs were overwhelming. Day sixteen fixed both.
Author: Koutian Wu; GitHub: ktwu01
What shipped
Single-adopter benchmark count fix. Benchmarks adopted by only one model card were being under-counted. The fix ensures these benchmarks are counted accurately, even when their adoption is narrow.
Progressive disclosure restored. Drill-down sections in the leaderboard, Q&A, and homepage are now collapsed by default. Users expand what they want to read instead of being confronted with everything at once.
Daily social-checklist issue restored. The daily social-checklist issue was restored after being accidentally retired. The issue date is now derived from the report date, not the wall clock, preventing timezone-related misalignment.
workflow_dispatch publish default. The workflow_dispatch publish input now defaults to true (issue #88). Previously, manual workflow runs required explicitly setting the publish flag.
Brand icons in frontier circles. The adoption frontier circles now show real brand icons (OpenAI, Google, Anthropic, etc.) instead of generic markers. This makes the frontier chart immediately scannable.
Why it matters
The single-adopter count fix addressed a data accuracy issue. When a benchmark has exactly one adopter, under-counting it makes the adoption landscape look less diverse than it is. For a tool that claims to track benchmark adoption, this kind of error undermines trust.
Progressive disclosure was a UX principle applied to information density. The radar collects a lot of data; showing all of it at once overwhelms new users. Collapsing drill-downs lets the dashboard serve both casual browsers (who see summaries) and deep researchers (who expand what they need).
The brand icons in frontier circles were a visual identity improvement. When you see the OpenAI logo next to a benchmark adoption point, you immediately know who adopted it, without reading a legend.
Issues addressed
- #88: default workflow_dispatch publish to true, restore daily social issue, date from report
- #99: correct under-counted single-adopter benchmarks
- #183: restore progressive disclosure
- #184: restore daily social issue
- #185: workflow_dispatch default
- #187: single-adopter audit
- #188: progressive disclosure restore
- #178: brand icons in frontier circles
Day seventeen: audit hardening, masthead actions, and production polish.