Skip to main content

Book Exercise Lanes

This module is judgment-heavy, so the "exercises" are produced artifacts rather than book problem sets. Use these four lanes to get volume on the skills after you have already learned them from the concept pages.

How To Use This Page​

  1. Finish the relevant concept page first.
  2. Try the artifact from memory once.
  3. Only then open the matching chunks for shape reinforcement.
  4. Keep a mistake log with tags such as patterned before I had pressure, review comment with no evidence, ADR without alternatives, over-mocked test, irreversible decision not flagged, or PR too large / commits too messy.

Lane 1: Integration -- Patterns in a Real Feature​

Use this lane when you can name individual patterns but still freeze when building a real feature.

Chunk reinforcements:

Target outcomes:

  • 1 complete PricingEngine-style feature (or equivalent) with each abstraction justified by a current pressure and cost; no minimum pattern count
  • 1 pressure table in the README (pressure, smell without pattern, pattern, cost accepted)
  • 3+ tests per pattern
  • 1 ADR attached

Lane 2: Review -- Comments on Real Code​

Use this lane when you can identify problems in code but struggle to write a review comment that lands.

Chunk reinforcements:

Target outcomes:

  • evidence-based findings or explicit requirement questions for every provided sample (Practice 2); no comment quota
  • Each labeled by bucket (correctness / clarity / design / tests) and nit/substantive
  • At least 2 review comments written on a real PR (open-source or a teammate's)
  • 1 "receive feedback" reply written for a disagreement, using the 5-step pattern

Lane 3: Documentation -- ADRs, Design Docs, Diagrams​

Use this lane when your code is fine but nobody else understands why it looks that way.

Chunk reinforcements:

Target outcomes:

  • 2+ ADRs in Context / Decision / Alternatives / Consequences format (one per realistic one-way door)
  • 1 two-page design doc for a feature you are actually building
  • 1 Mermaid sequence diagram with at least one failure branch
  • 1 C4 container diagram of a real system (yours or an OSS one)

Lane 4: Stewardship -- Shape the History and the Workflow​

Use this lane when you ship correct code but it is painful to review and read later.

Chunk reinforcements:

Target outcomes:

  • 1 branch rewritten with git rebase -i into 3-7 one-intent commits
  • 1 PR description using the full template (Summary, Risk/Rollback, How to Verify, Not in Scope)
  • 1 pre-commit config and CI job wired up in a real repo (formatter + linter + tests)
  • 1 broken-window cleanup commit landed in a real codebase, called out in the PR body

Completion Checklist​

  • Completed at least two lanes in full
  • Produced at least 2 ADRs and 1 design doc in Lane 3
  • Reviewed every Lane 2 sample with supported findings or explicit requirement questions, without a comment quota
  • Retained actual mistakes, their evidence and corrections; if none occurred, label a deliberately faulty example and show how to detect it. No minimum personal-error count.
  • Reattempted at least 2 artifacts after peer or self-delayed critique

Required review-and-repair evidence​

Complete the evidence-based review extension and all AR1–AR6 gates. Reuse its packet in the project; HCI/accessibility work remains separate and required.