A computational lab notebook meets a scientific atlas: warm paper, deep ink, capillary-blue flow lines, and one sharp coral annotation colour.
- Hero — one promise: learn to reason with code rather than memorise syntax; the five-step reasoning loop is the dominant visual.
- Support — show the quick and full routes, with the same course spine.
- Detail — map each lesson to a scientific decision and a tangible output.
- Final action — start lesson zero, inspect the repository, or teach the course.
- The hero copy enters as one calm staggered sequence.
- The reasoning-loop path draws once, turning an abstract workflow into motion.
- Curriculum rows reveal their output on hover or keyboard focus; the route spine remains sticky on wide screens.
All motion respects prefers-reduced-motion.
Give future CoMPhy students enough computational judgement to enter a real research repository without treating Python as a bag of spells. The interface should feel rigorous, open, and unintimidating.
Editorial scientific field notebook, with a restrained Swiss grid and a small amount of capillary-flow character.
- Static MkDocs site deployed by GitHub Actions.
- Canonical URL:
https://comphy-lab.org/comphy-python101/. - Responsive from 320 px upwards, keyboard navigable, WCAG AA contrast.
- Minimal JavaScript and no decorative dependency bundle.
- Fast enough to remain useful on institutional Wi-Fi.
The memorable object is the reasoning loop:
question → representation → transformation → verification → evidence
It appears in the hero, lesson structure, exercises, and capstone assessment.