The author argues that most life-changing advances—from infant survival to indoor plumbing—happened before the digital era. He says smartphones, AI, and 3D baby scans feel novel but are refinements of century-old inventions, not signs of a new age of breakthroughs.

• Infant mortality dropped from ~100 per 1,000 births in 1900 to under 6 today, with nearly all the improvement occurring between 1870 and 1970, before digital technology existed.
• Smartphone features (cameras, GPS, telephony) are refinements of inventions 35-130 years old, not new breakthroughs.
• Post-war productivity gains have halved since 1970 despite the internet, and daily routines like cooking, housing, and travel look largely unchanged from 50 years ago.
• Modern innovations like streaming, digital payments, and AI chatbots are real but don't match the scale of past breakthroughs like electrification or sanitation.

The article explores the concept that AI advancements follow a predictable pattern, which the author refers to as “straight lines on graphs.” It discusses the uneven capabilities of AI across different tasks while suggesting that the rate of improvement remains consistent. The author also speculates on the impact of reinforcement learning and compute resources on future AI development.

• AI task-horizon length (how long a task the AI can complete) doubles every 3–7 months, and this rate holds fairly steady even though absolute capability varies wildly across domains.
• The apparent RL-driven acceleration in benchmarks may be an illusion caused by measured tasks overlapping with what labs specifically post-trained on, not genuine broad capability gains.
• Progress splits into general pre-training gains (lift everything) versus targeted post-training gains (lift specific benchmarks labs choose to optimize for cost/PR reasons), and that targeting can shift over time.
• Even modeled compute slowdowns are unlikely to meaningfully delay key AI milestones, since current capability growth is fast enough to hit them first.

Progress in software development often leads to complexity and waste rather than clarity and efficiency. As experience grows, the focus shifts from cleverness to simplicity, empathy, and effective communication, emphasizing that successful systems are built on strong relationships and understanding among team members.

• Most added complexity in software (extra abstractions, layers, frameworks) counts as motion, not real progress, and often just creates future cleanup work.
• Senior engineers trend toward writing simpler, more boring code rather than clever code, because cleverness usually costs the team more in maintenance than it earns in elegance.
• Technical success depends more on communication and trust between people than on tooling or architecture choices—most project failures trace back to misalignment, not bad code.
• Genuine progress looks like deletion and simplification (removing code, cutting dependencies, clarifying intent) rather than accumulation of new features or systems.