P=NP Explained: Why Class Schedules and Circuit Routing Are the Real Hard Problems
thesaraharminta · x · 2026-09-11
A popular-science thread arguing that most people don't actually understand the P=NP problem.
- Before the 1960s, engineers assumed faster computers would eventually solve any hard problem.
- They then hit a class of problems that crash or take billions of years regardless of compute power.
- These are everyday problems, not exotic physics: scheduling a high school timetable where no teacher is in two places at once, or drawing the shortest non-overlapping trace on a circuit board.
- The kicker: once a solution is found, it can be verified very quickly — the core tension of P=NP.
Note: only the opening of the thread is included in this post.
More from Research
- Steerable Visual Representations Presented as ICML Long Oral — y_m_asano · 2026-09-11
- OpenCVL: a satellite-to-photo registration dataset at ECCV 2026 — ducha_aiki · 2026-09-11
- Diverse VPR work submitted to ECCV 2026 — ducha_aiki · 2026-09-11
- EASE: evidence-anchored spatial attention lifts multimodal RLVR by up to 3.1 points, EMNLP 2026 — jiqizhixin · 2026-09-11
- Hypothesis: ASI Has a Mathematical Incentive to Preserve Human Diversity — No_Cause_2731 · 2026-09-11
- MutexaGPT: LLM agents plus MD simulations hit 40% on enzyme design, 4x the baseline — bravo_abad · 2026-09-11