Daily Inspiration
Design excellence, every day.
Jury-selected work from the A' Design Award, presented fresh each morning.
A random yes or no. Fair 50/50 probability on every decision. Your browser decides; the server never knows.
A yes-or-no question is the smallest possible unit of information: one bit. Two outcomes, each with probability 0.5, each completely independent of every previous answer. Ask a hundred times and the expected split is fifty-fifty. The standard deviation for 100 trials is 5, so seeing between 45 and 55 "yes" answers covers roughly 95% of all possible outcomes. The mathematics is identical to a coin flip. The psychology is entirely different.
Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso published a landmark study in 2011 examining 1,112 judicial rulings by experienced Israeli parole judges. Judges approved parole at rates approaching 65% immediately after meal breaks and declined at rates near zero just before the next break. The severity of the crime, the length of the sentence, and the prisoner's ethnicity all had less influence on the outcome than the simple question of when the judge last ate.
The finding illustrates decision fatigue: when humans make too many consecutive choices, the decision-making mechanism itself degrades. Tired minds default to the status quo. A random generator experiences no such degradation. Decision number ten thousand draws from the same entropy source as decision number one. The answer is always fresh.
Each answer on this page calls crypto.getRandomValues(), the Web Cryptography API specified by the W3C and implemented in every modern browser. Your device draws from hardware-level entropy: thermal noise, electrical jitter, and timing variations in silicon. These physical processes are fundamentally unpredictable at the quantum level. The result is an exact 50/50 split, generated entirely on your device. The server delivers the page. Your browser delivers the verdict.
Dan Ariely, a behavioral economist at Duke, has documented how people struggle with binary choices when both options carry roughly equal weight. The deliberation consumes time and mental energy without improving the outcome. Committing in advance to follow a random answer eliminates this cost entirely. The decision happens instantly. The mental resources freed up become available for choices that genuinely benefit from deliberation.
There is a second, subtler benefit. When you flip the decision to a random source and notice a flash of disappointment at the result, you have discovered your actual preference. The random answer did not make the choice. It revealed the choice you had already made. This technique works precisely because the generator is verifiably fair. A biased process would undermine the insight.
A yes-or-no generator demonstrates the Bernoulli trial, the foundation of probability theory. Project this page on screen and run 30 decisions. Record each result on the board. Ask students to predict the longest streak of consecutive identical answers before starting. The actual streak almost always exceeds the prediction, which opens a productive conversation about human intuition and genuine randomness.
For a deeper exercise, have half the class write a "random-looking" sequence of 30 yes-and-no answers by hand while the other half uses the generator. Compare the two sets. Human-generated sequences alternate too frequently and avoid the natural clumps that true randomness produces. Jakob Bernoulli formalized this observation in his Ars Conjectandi of 1713. Three centuries later, the bias persists in every classroom that runs this experiment.
Every answer generates inside your browser. The server delivers the page; your device creates the outcome. Decision history stays in localStorage under your control. The server stores no answers, sets no tracking cookies, and keeps no user accounts.
Sharing this URL sends the tool, never your results. The recipient receives the same page and generates their own independent answers from their own device. Two people clicking "Decide" at the same moment produce completely separate outcomes. The URL carries the tool. Your device carries the randomness.
Send this link. They get the same tool, their own answer. Compare your streaks.
Daily Inspiration
Jury-selected work from the A' Design Award, presented fresh each morning.
