Online coding and Python classes in Bracknell

The learner models a tipping bucket that tips once for every 0.2 millimetres of rain, a common size for such gauges rather than a figure from the museum's record. Six invented showers fall over a week, from 0.05 millimetres to 12.35. Python keeps a running total of true rain, uses whole-number division to count completed tips, and reports each shower as its tips times 0.2. The key design choice is that water left in a part-full bucket stays there and counts towards the next tip.

Each single shower is only accurate to one step, and tiny showers vanish or appear late. But because the part-fill is carried forward, nothing is lost: the week's total is out by less than one tip. Throw that remainder away after every shower, the slip in the last column, and the errors pile up in one direction, leaving the week 0.63 millimetres short. The same idea explains rain rates. Counting tips in five-minute blocks, the rate can only be a multiple of 2.4 millimetres per hour, so a gentle drizzle reads as zero, then suddenly as 2.4.

The learner writes two tests: the reported total must never differ from the true total by a whole tip or more, and a shower of exactly 0.2 millimetres must produce exactly one tip, not zero because of a rounding slip in floating point. We also convert the classic gauge's size: a 5-inch opening is 127 millimetres across, about 12,668 square millimetres, so each millimetre of rain puts about 12.7 millilitres of water in the bottle.

Asking how finely a sensor can really measure is a strong reason for a Bracknell teenager to keep coding in 2026. The longer argument is in why teenagers should still learn to code in 2026.