How Salt Destroys Leather (And How to Stop It)

Public Housing in America: A State-by-State Breakdown

What is public housing, really? It’s not the high-rise projects of decades past. Today, it’s a patchwork of 3,000 local housing authorities, each with its own rules, waitlists, and funding crises. With the midterms approaching and renters squeezed nationwide, state-level policy divergence has never mattered more. We break down how eligibility works, why New York has six times the units of Texas, and what it actually takes to get on a waitlist — from Washington’s streamlined portal to Illinois’s fragmented maze.

Moving Secrets from Film Scouts & Museum Pros

Most moving advice comes from professional movers — but the best systems come from professions that move complex, irreplaceable things under pressure. This episode unpacks how film location scouts document every power outlet and cable path before a single truck arrives, how museum registrars achieve a 0.1% damage rate on priceless artifacts, and how yacht crews maintain floating inventory systems. You'll learn practical techniques you can steal: strike kits organized by setup sequence, dependency maps that prevent mid-move chaos, and six-angle photography that reveals hidden problems. Whether you're moving a smart home network or a single room, these professions have already solved the problems you're about to discover at 2 AM.

Housing as Anchor vs. Safety Net

This episode unpacks a provocative proposal: strip the job guarantee from the state's safety net and instead guarantee just housing, food, and healthcare — unconditionally, without means-testing, and as a legal right. We explore how this differs structurally from existing programs like Section 8 and SNAP, why waitlists and welfare cliffs create chronic anxiety even when nothing's gone wrong, and what Israel's month-to-month rental market reveals about the psychological cost of insecurity. Drawing on cortisol studies, UBI meta-analyses, and international comparisons (Germany's indefinite leases, France's eviction moratorium, Denmark's social housing), we ask whether the real problem isn't the safety net's size but its design — and whether a universal anchor could actually be cheaper than our current patchwork of conditional assistance.

How to Break an LLM's Bad Verbal Habits

Every large language model develops a fingerprint of overused phrases — "delve," "failure modes," "it's worth noting." Blacklists leak into output, regex silently inverts meaning, and feeding previous episodes into context only compounds the problem. This episode explores three practical approaches to breaking the reinforcement loop: compressed style summaries with decay, dual-model editing architectures, and context-aware regex using dependency parsing. No silver bullets, just real trade-offs for production pipelines.

Pharmacokinetics vs Neural Nets: Two Meanings of "Model

The word "model" now triggers thoughts of AI, but clinicians have used pharmacokinetic models for decades — differential equations with physically meaningful parameters, no training data required. This episode unpacks the fundamental differences between classical mathematical models, machine learning models, and algorithms, using drug dosing as a concrete example. We explore how theory-first PK models (built from known physiology) compare to data-first neural networks, why extrapolation risk differs dramatically between them, and how the FDA regulates each category differently. A practical taxonomy for anyone buying, building, or relying on "models" in 2026.