THE REFERENCE LIBRARY
A growing collection of ideas for understanding the world — and your own thinking.
Never let the enemy overwhelm you; keep your spirit high and resolve strong, aiming to cut him down regardless of his approach. Through constant practice and deep study, you internalize this mindset until action becomes instinctive.
Taleb uses Poincaré’s work on the three-body problem to explain a limit of forecasting: in a system sensitive to initial conditions, small measurement errors can grow until a precise long-term prediction is no longer useful.
The authors recommend checking three contexts before judging unfamiliar online information: the source, the claim, and your own knowledge and reactions. They report that students sometimes improved their judgments after as little as 30 seconds of context-seeking; the time is an illustration, not a guarantee.
People react differently when their beliefs are challenged. Grant contrasts distress and enjoyment in a harsh historical study, then asks how admitting error might become an opportunity to learn rather than a threat to one’s identity.
To properly understand the Titanic tragedy, one must look beyond the immediate actors and consider the many factors that set the stage—including regulations, design, route, and atmospheric conditions—which often go back weeks, months, or even years before the event.
Taleb argues that filling children’s lives with scheduled activities leaves too little room for trial and error, curiosity, and learning to handle ambiguity. He contrasts this with self-directed learning that allows for unplanned encounters and discoveries.
The Toyota Way is a management philosophy and organizational culture with two pillars—continuous improvement and respect for people—and five foundation values: Challenge, Kaizen, Genchi Genbutsu, Respect, and Teamwork. It instills a culture for achieving organizational goals by leveraging the Toyota Production System.
The Tragedy of the Commons describes how individuals acting independently and rationally according to their own self-interest can deplete a shared resource, leading to outcomes contrary to the common good.
The Tree of Life represents descent from common ancestors. Human relationships appear within a much larger history connecting humans with other animals, plants, and other living organisms.
A classification of systems and entities into three categories based on their response to volatility and stressors: fragile (harmed), robust (unaffected), and antifragile (benefiting). The categories are relative and context-dependent.
Extreme results often contain both persistent differences and temporary luck. When the temporary component changes, later results tend to be less extreme. Secrist mistook this regression toward the mean in business data for evidence that competition forces superior firms into mediocrity.
Our judgments of value often depend on the alternatives in front of us. In the decoy effect, adding an option that is clearly worse than a similar option can make that similar option more attractive, even if no one chooses the decoy. Ariely connects this pattern with other comparison effects in shopping, pay, and social life.
A turkey expects its daily feeding to continue, then is killed before Thanksgiving. Taleb uses the story to show how a repeated pattern can support confidence while leaving an important risk unexamined.
The Turkey Problem illustrates how reliance on past evidence can lead to catastrophic surprise when a Black Swan event occurs, as in the case of a turkey being fed for a thousand days before Thanksgiving. It warns against mistaking absence of evidence for evidence of absence.
The Turkey Problem illustrates a limit of induction: past observations do not guarantee that a pattern will continue. In Taleb’s example, a turkey fed daily grows increasingly confident of human benevolence as slaughter approaches.
A structured method for constructing a timeline of events to analyze a decision or problem, emphasizing listing dates first to build a chronology, with a special form called 'placement' used to understand the mindset of decision-makers.
Parrish offers two complementary ways to manage weaknesses: develop strengths through practice and use safeguards where effort alone is unreliable or cannot remove a limitation. Safeguards can support both learned habits and biological limits.
Different corrective actions vary in their effectiveness at addressing root causes, ranging from weak reminders to strong systemic redesign. Combining multiple actions that work on the system rather than in the system is more likely to produce lasting change.
Taleb contrasts domains where many ordinary observations determine the total with domains where a few extreme observations can dominate it. He calls them Mediocristan and Extremistan. The distinction affects how much past experience can tell us about future outcomes. It is a rough framework: an extreme event need not be an unpredictable Black Swan.
Taleb argues that career incentives can sustain flawed methods when people defend them by saying everyone uses them. He criticizes academic conformity and asks who bears the cost when established teaching or advice causes harm.
Taleb argues against treating the Gaussian bell curve as a default description of the world. Where rare outcomes can dominate totals, a standard deviation may conceal important features of risk. The task is to establish where the model fits before relying on it.
The passage labels a feminine style of appearance and behavior the ultra-girl strategy and interprets it through the halo effect. The account relies on gender stereotypes and public personas; it does not establish the motives of individual women or what all men prefer.
Regularly rethink and update your beliefs and methods. Philosopher Robert Nozick taught a new class each year to engage in fresh thinking; his practice suggests that learning and growth come from challenging old ways and considering new perspectives, both in education and life.
Cognitive busyness can reduce people’s correction of an initial impression for situational constraints. In the studies described here, participants later used information held in memory to revise their judgments when the competing task had ended.
The understanding/acceptance principle uses correlations with cognitive ability to examine disputed standards for reasoning tasks. Its proponents argue that positive correlations support the scoring model, while absent or negative correlations give reason to reconsider the model or the task’s interpretation. The correlation alone does not settle which answer is normatively correct.
A model that explains how we often believe in systems and theories that we cannot prove, and how these models can become mistaken for reality, influencing our actions and perceptions.
A causal map can reveal several points where an incident might have unfolded differently. Cheng uses a passenger’s forced removal to show how crew scheduling, airline decisions, and security escalation can interact. Recognizing those connections does not assign equal responsibility to everyone involved.
Conic sections—ellipses, parabolas, and hyperbolas—are simple curves that recur across nature, science, and even literature because mathematical simplicity forces them into diverse roles.
The Unsubscribe Trap is a form of sludge where companies make it significantly easier to sign up for a subscription than to cancel it, using intentional friction as a retention strategy.
The empty space in an object makes it useful: the center of a wheel, the hollow of a vessel, and the interior of a room each serve a function. Lao Tzu illustrates this in the Dao De Jing.
Opposition and change can expose weaknesses or create opportunities to develop. The entry connects Nietzsche’s view of political opposition with physical and storytelling analogies, whose limits matter when applying them elsewhere.
Open-source intelligence (OSINT) uses publicly available information to inform decisions. Analysts combine sources, check what they show, and place them in context; public reports can reveal more than their authors intended.
Systems thinking helps recognize when things appear fine but may be approaching a tipping point, and identifies vicious circles by unpacking feedback loops, enabling better analysis of cause and effect.
Even with fractal and power-law models, the problem of induction persists; complexity theory should make us more suspicious of precise models because history does not reveal its generator, and failing to distinguish forward from backward processes is a major epistemological sin.
A warrior must respect and maintain their tools through constant training and study, understanding all aspects of their craft and planning for contingencies, allowing the 'spirit of the thing' to guide mastery.
Musashi connects combat readiness with practiced stance, balance, concentration, and decisive movement. He urges warriors to make their bearing consistent with their training and purpose.
Charlie Munger argues that hidden investment costs and accounting practices create a disguised 'wealth effect' from rising stock prices, functionally equivalent to undisclosed embezzlement, which amplifies economic booms and busts.
Overconfidence is common in judgment, but easy questions and sequential updating can also produce underconfidence. Griffin and Tversky examine these patterns and the role of feedback in calibrating confidence.
Familiar routes feel shorter due to routine, causing underestimation of travel time for habitual journeys and overestimation for new ones.
People often give predictions more certainty than the evidence supports. Allowing for a wider range of outcomes can improve planning, especially when an unlikely event would be costly.
The world is more random than we perceive, and humans are systematically fooled by randomness. Inequality in superstar economies often results from luck and imitation, not just skill, as demonstrated by the winner-take-all effect.
The AND model looks for ways to meet both parties’ needs. It encourages collaboration without requiring people to give up their own interests or assume that every conflict has a mutually satisfying solution.
Calling a procedure a best practice can discourage people from questioning it. Grant argues that organizations should evaluate how decisions are made alongside their results and give people room to challenge assumptions, report problems, and test improvements.
A four-step process (Widen Your Options, Reality-Test Your Assumptions, Attain Distance Before Deciding, Prepare to Be Wrong) to counteract cognitive biases and improve decision-making. The model is illustrated with examples like Priestley's choice, Steve Cole's 'AND not OR', Andy Grove's successor question, and David Lee Roth's M&M tripwire.
Recording a concern can help others understand a decision, but it may also strain working relationships or expose sensitive information. Leaving the concern unwritten creates a different risk: decision-makers may fail to address a problem that people already know about.
The wrong side of right is defending an idea because it is yours after a better option emerges. The model asks you to judge proposals by their likely consequences and change your position when the evidence warrants it.
A decision-making heuristic that uses a set of clear yes/no questions to quickly evaluate options, particularly under time pressure and risk. It relies on predefined parameters to simplify complex choices.
Grant uses the Yoda Effect to describe revising beliefs and assumptions when evidence changes. His forecaster examples emphasize testing a favored view and keeping personal identity separate from past predictions.
In Ariely’s sharing experiments, more people accepted a free item, but each participant took fewer units than when a small price was charged. He interprets that restraint through social norms: people may treat a free shared resource differently from a purchase. The result depends on the setting and does not imply that every free offer reduces consumption.
Dennett calls the intuition that physical mechanisms must leave consciousness unexplained the Zombic Hunch. Philosophical zombies—beings that act exactly like conscious people but lack experience—give the intuition a vivid form. He argues that its apparent force reflects a failure of imagination rather than a demonstrated limit on physical explanations.
Root Cause Analysis (RCA) can be scaled to an organizational level to identify system-wide improvements, share good practices, and create consistency across processes. This thematic approach helps reveal cross-functional themes and supports sustained improvements.
The Theory of Constraints (TOC) is a management philosophy by Eli Goldratt that focuses on identifying and managing bottlenecks to improve system performance, using the Five Focusing Steps and Throughput Accounting.
The book proposes taking a break when deliberate analysis stalls. Its coin-toss exercise uses your reaction to an imagined outcome to explore preferences; that reaction is one input to a decision, not proof of which option is best.
Experts can defend a failed conditional forecast by questioning its conditions, invoking a near miss, extending its deadline, or dismissing forecasting itself. Any such objection may be valid. Tetlock examines whether experts apply these defenses selectively and update confidence consistently with their earlier probability judgments.
Banasiewicz distinguishes theory-guided hypothesis testing from exploratory analysis that seeks patterns without a predetermined explanatory theory. He favors using established knowledge to guide and check new findings. Both approaches depend on data quality, and exploratory findings need further testing before they support confident decisions.
The concept that hidden information or unknown factors are always present, influencing outcomes and decisions, and that acknowledging this can improve probabilistic thinking and reduce overconfidence.
Senge argues that problems should be understood through relationships within a shared system. Blaming an isolated opponent can divert attention from improving the interactions that sustain the problem.
Public opinion is often contradictory and inconsistent, with people simultaneously expressing support for broad goals like cutting spending while opposing cuts to specific programs they favor, making it difficult to discern a coherent public will.
Expecting to control other people’s choices can add pressure to an already difficult situation. The author’s sales experience illustrates the value of directing effort toward preparation and actions while recognizing the limits of control over results.
Thermodynamics describes energy transfer, entropy and thermal equilibrium. Its account of the energy needed to sustain processes can also suggest metaphors for maintenance, though social order and motivation are not physical entropy or heat.