lindy-effect

# Lindy Effect ## One-Liner The older something is, the longer it's likely to survive—future life expectancy is proportional to current age for non-perishable things. ## Core Insight The Lindy Effect states that for non-perishable items (ideas, technologies, books, institutions), every additional period of survival implies a longer remaining life expectancy. A book that's been in print for 40 years will likely remain in print for another 40 years; if it survives another decade, its expected life extends to 50 more years. This inverse relationship between fragility and time creates a powerful heuristic: what has endured longest is least likely to disappear soon. **Key Distinction**: Applies only to non-perishable items without natural expiration dates. Humans age toward death (perishable), but ideas age toward immortality (non-perishable). ## Mental Model ``` Time as Filter: Year 1: 1000 new ideas emerge Year 10: 100 remain (90% eliminated) Year 100: 10 remain (weak ideas filtered out) Year 1000: 1 remains (only the robust survived) Remaining Life Expectancy: 10-year-old idea → expect 10 more years 100-year-old idea → expect 100 more years 1000-year-old idea → expect 1000 more years The longer it has survived, the more it has proven its robustness against time's disorder. ``` **Power Law Distribution**: Without natural upper bounds, survival follows a power law where remaining life expectancy ∝ current age. ## When to Use - **Learning priorities**: Decide which knowledge to invest time acquiring - **Technology choices**: Select tools and frameworks for long-term projects - **Reading lists**: Filter signal from noise in information overload - **Investment decisions**: Evaluate business model durability - **Tradition evaluation**: Decide which customs to preserve vs. discard - **Career skills**: Choose skills with lasting vs. ephemeral value **Apply when**: Dealing with non-perishable information products, choosing between old and new approaches, filtering high-volume decisions. **Don't apply when**: Dealing with perishable items (food, people), in rapidly shifting technical domains where novelty is advantage, when recency is the value proposition. ## Execution Steps ### 1. Classify Perishability **Non-Perishable (Lindy applies)**: - Ideas, theories, mental models - Books, philosophical texts - Recipes, techniques, practices - Institutions, religions, customs - Languages, mathematical proofs - Classical music, art movements **Perishable (Lindy doesn't apply)**: - Humans, animals, organisms - Food, consumables - Machines with wear patterns - Things with planned obsolescence - Fashion trends (deliberately ephemeral) ### 2. Assess Survival Time - How long has this idea/technology/practice existed? - In what form? (original vs. evolved) - Has it survived competitive pressures? (alternatives existed) - Has it survived environmental changes? (context shifts) **Mechanism**: Longer survival = more exposure to Black Swan events without dying = revealed robustness ### 3. Apply Proportional Life Expectancy **Simple Heuristic**: - 1-year-old framework → skeptical, likely fad - 10-year-old framework → cautiously optimistic - 50-year-old framework → high confidence - 500-year-old framework → near-certain durability **Mathematical Form** (for the curious): If no natural expiration, remaining life T ∝ current age t Expected remaining life ≈ current age (power law exponent near 1) ### 4. Combine with Skin in the Game Lindy Effect is strongest when: - Survival had real consequences (not artificially preserved) - Alternatives existed (competitive selection occurred) - Environment changed (tested against novelty) - Users could abandon it (market discipline existed) **Red Flags**: Government mandates, monopoly lock-in, lack of alternatives weaken Lindy signal. ### 5. Apply Decision Filters **For Learning**: - Prioritize 100-year-old knowledge over last year's framework - Read classics before bestsellers - Study fundamental principles before hot techniques - Exception: Rapidly evolving fields (AI, genetics) need recent knowledge **For Technology Choices**: - Unix tools (50+ years) > JavaScript framework du jour (2 years) - SQL (40+ years) > NoSQL fad (10 years) for most use cases - HTTP (30+ years) > new protocol with no adoption - Exception: When new tech solves previously impossible problem **For Business Models**: - Subscription (centuries old) > complex revenue innovation - Marketplace (millennia old) > novel platform play - Direct sales (eternal) > untested growth hack - Exception: When innovation creates new category ### 6. Discount Novelty Bias Humans overweight recent information: - Newest book gets attention despite weak Lindy - Old wisdom ignored due to availability bias - Fads cycle because each generation rediscovers - Lindy corrects this: weight by survival time, not recency ### 7. Identify Fragility in the New **Young ideas are fragile**: - Haven't faced diverse conditions - May succeed due to temporary tailwinds - Untested against competition - Could be survivor of small sample (early luck) **Inversion**: Most new things will die young. Bet on the small fraction that won't by waiting for Lindy signal. ## Real-World Examples **Books and Reading** - Read Meditations (1,800 years old) before latest productivity bestseller (1 year old) - Stoicism (2,300 years) has longer expected life than "Hustle Culture" (5 years) - Homer's Odyssey (2,700 years) → expect 2,700+ more years - Most current bestsellers → expect forgotten in 5 years **Technology and Tools** - C language (52 years) still dominant; Pascal (50 years) dying → slight Lindy difference, network effects matter - Unix philosophy (50+ years) outlived countless replacements - Relational databases (50 years) > document stores (15 years) for Lindy bet - Vim/Emacs (40+ years) → will outlive modern IDE fads **Business Practices** - Double-entry bookkeeping (700 years) → immortal - Joint-stock companies (400 years) → extremely robust - Apprenticeship models (millennia) → still effective despite education innovations - Most management fads (5-10 years) → already dying **Investing** - Warren Buffett applies Lindy: "Only buy businesses that have been around a long time" - Coca-Cola (138 years) more Lindy than food delivery startup (3 years) - Real estate (5,000+ years as asset class) > cryptocurrency (14 years) ## Common Traps **Trap 1: Applying to Perishables** - "This 90-year-old will live another 90 years" → NO, humans are perishable - Lindy requires non-perishable items without natural decay **Trap 2: Ignoring Context Shifts** - "Latin survived 2,000 years, so it'll be spoken 2,000 more" → NO, utility context changed - Lindy assumes continued selection pressure; artificial preservation breaks it **Trap 3: Survivor Bias Confusion** - "Most old ideas are good because they survived" → Correlation, not causation - Only dead ideas teach us what fails; survivors are biased sample **Trap 4: Freezing in Amber** - "Never change old practices" → Wrong - Lindy guides bets, doesn't prohibit innovation. Update when evidence overwhelms prior. **Trap 5: Novelty Never Wins** - "Always choose old over new" → Ignores paradigm shifts - Occasionally new dominates (electricity > candles). Lindy is Bayesian prior, not law. **Trap 6: Forgetting Competition** - "My tradition is 1,000 years old" → Was there selection pressure? - Monopoly or mandated traditions have weak Lindy (not tested) ## Relationship to Other Frameworks **Antifragility** - Lindy is evidence of antifragility through time - Things that exhibit Lindy have benefited from temporal volatility - Time = disorder; Lindy survivors gained strength from it **Chesterton's Fence** - Both warn against removing old things without understanding - Lindy: Old has proven value - Chesterton: Don't remove until you know why it exists **Via Negativa** - Lindy suggests subtracting new, unproven additions - What survived time needs less justification than what's novel - Default to old, require proof to adopt new **Second-Order Thinking** - Lindy forces long time horizons: "Will this matter in 50 years?" - Filters out short-term fads from durable value - Time-tested beats recently popular ## Cross-Domain Applications **Engineering**: Prefer battle-tested libraries over shiny new frameworks (unless new solves impossible problem) **Medicine**: Traditional remedies with centuries of use deserve investigation despite lack of RCTs **Investing**: Long-lived companies with durable moats > hot startups with no profit **Education**: Teach timeless skills (writing, math, logic) before vocational training for current job market **Personal Development**: Study ancient philosophy before modern self-help **Product Design**: Classic designs (Eames chair, Leica camera) reveal enduring aesthetic vs. trend ## Key Principles - **Time is the ultimate filter**: What survives has revealed robustness - **Non-perishable only**: Applies to ideas, not organisms - **Proportional expectancy**: Remaining life ≈ current age - **Discount novelty**: New things are fragile until proven otherwise - **Context matters**: Must have faced selection pressure to be meaningful ## Further Reading - Taleb, Nassim Nicholas (2012). "Antifragile: Things That Gain from Disorder" (Chapter on Lindy) - Erenow (Blog). "The Lindy Effect" (mathematical treatment) - Toby Ord (2023). "The Lindy Effect" (Oxford paper, arxiv.org/pdf/2308.09045) - Luca Dellanna. "The Lindy Effect" (practical applications) - Albert, Bartlett. "The Most Important Video You'll Ever See" (exponential vs. Lindy thinking) --- **Source Domain**: Military Strategy, Ancient Wisdom & Hidden Gems (07) **Pattern Type**: Heuristic / Time-Based Selection Filter **Practitioner Value**: 9/10 | **Clarity**: 9/10 | **ROI**: 9/10 | **Novelty**: 8/10 | **Cross-Domain**: 10/10 **Total Score**: 45/50