bi-03 step 01 — Predict the DOM

Goal

Produce four written DOM predictions before observing anything, then score them. The gap is the lesson; a prediction made after observation is worth nothing.

Tasks

  1. Open docs/observation.md and write predictions for cases A–D in src/ (a.html, b.html, c.html, d.html). For each: the full tree, every text node with exact contents, table.parentNode, and table.childNodes.length.

  2. For case C, additionally predict the total number of text nodes and whether <table> has any children at all.

  3. For case D, name the element present in the result that was never written in the source, and the spec rule that creates it.

  4. Do not use innerHTML or DOMParser to check. Those run the fragment parsing algorithm with a different context element and can produce a different tree.

Done when

  • Four predictions written, with exact text-node contents
  • Predictions timestamped before any browser was opened