Every explainer on this site follows the same process. It is written down here so you can judge the work, and hold me to it.
How topics are chosen
A topic earns a page when it is a genuine load-bearing idea, one that a curious person keeps bumping into and that rewards being understood properly rather than merely name-dropped. The bias is towards ideas that are settled enough to explain honestly but deep enough to be worth three levels: the shape of a proof, the mechanism behind a phenomenon, the reason a famous result is true. I do not add a page just because a term is being searched for. If I cannot explain something without either hand-waving or copying, it does not get published.
How the three levels are pitched
Each page presents the same idea at three depths, and every level is written to stand on its own.
- Level 1 is for a curious twelve-year-old. Plain words, a concrete picture, no notation. The goal is a true mental model, not a simplified-to-the-point-of-wrong one.
- Level 2 is for a reader who has just finished school. It introduces the key equation or mechanism and explains what each part is doing, without assuming a degree.
- Level 3 is pitched at an undergraduate. It states the result precisely, uses the standard notation, and is careful about the conditions under which the result actually holds.
Moving up a level should add rigour, never contradict the level below. If Level 1 has to be untrue to be simple, the page is wrong and gets rewritten.
What counts as a primary source
Where a page rests on a specific result, I go to the primary literature rather than to a chain of summaries. In order of preference, the sources I treat as primary are:
- the original peer-reviewed paper or its preprint on arXiv, cited by author and year;
- the official statement of a problem or result from the body responsible for it, such as the Clay Mathematics Institute for the Millennium Prize Problems or the Particle Data Group for particle physics;
- a review article or standard graduate textbook, where the point is settled consensus rather than a single result.
Encyclopaedias and news write-ups are useful for orientation but are not cited as authority. The specific works a page relies on are listed in its Sources section, and those links are followed, not hidden, because pointing at the real literature is part of the job.
How the science is kept honest
Where a field is unsettled, contested, or resting on a prediction that has not been observed, the page says so in plain terms rather than presenting an open question as closed. Status lines on the technical level mark whether something is established, predicted, or genuinely open.
How errors get corrected
I will get things wrong. When that happens I want to know. If you spot an error, email me at corrections@tldrscience.net. I read every message. When a correction is warranted I fix the page, note what changed, and record it in the public corrections log with the date. The log is deliberately open: a site that never publishes a correction is not more accurate, only less honest.