Instructor Guide and Review Status
This is an authored candidate with executable reference checks. Independent subject/code review and learner pilots are pending. No percentile outcome or measured teaching effectiveness is established by the amount of content.
Scope and scheduling
Allow 95–135 hours, an author estimate for a learner who meets the diagnostic:
| Work | Estimated hours |
|---|---|
| Diagnostic, lessons, contracts and bounded reading | 33–45 |
| Five labs, eight core implementations | 26–38 |
| Independent practice and feedback | 20–28 |
| Assessment, review and remediation | 8–12 |
| Integrated capstone | 8–12 |
Do not add the same exercise time twice when reusing an artifact. Remediation, optional implementations, research papers and full-book reading can add time. At ten hours weekly, plan roughly ten to fourteen weeks, then adjust from logs. Teach amortization lessons 11–14 before advanced heaps when that dependency helps.
The core codes AVL, DSU, binomial heap, doubling array, Fenwick, lazy sum tree, persistent sum tree and Bloom filter. Red-black, skiplist, treap, Fibonacci, pairing and succinct structures still require conceptual reasoning; their full implementations are depth options. Do not turn every option into an implicit gate.
Running a cohort or self-study review
- Use the prerequisite diagnostic without expecting unfamiliar advanced structures. Assign the named repair path before interpreting a failure as module difficulty.
- Ask for a predicted trace before running a checker. Review invariants and resource models separately from returned values. Inspect destructive ownership and old versions.
- Keep the assessment key closed during the initial attempt. Record assistance, actual time and unattempted tasks; adjust accommodations explicitly.
- Score each competency using assessment. Require the level-2 evidence for every competency; a high average cannot hide incorrect persistence or a false-negative membership shortcut.
- Recheck repaired misconceptions with changed data or assumptions. After 7–14 days, ask for a fresh transfer explanation or implementation change without the old solution.
Independent review request
Reviewers should inspect the AVL deletion/metadata proof, binomial ownership and carry bounds, DSU theorem statement, initial potential, random-model qualifications, lazy tag composition, persistence sharing, and Bloom publication lifecycle. Run the checks but also read code: finite tests cannot establish all asymptotic claims. Confirm the assessment key and capstone fixtures independently. Record source/edition disagreements and the exact change needed, with a minimal counterexample where possible.
Pilot record to collect
Record participant prerequisites, weekly hours, time per stage, hints used, first failing sequence, rubric disagreement, recheck result and delayed transfer. Start with a small cohort spanning relevant prerequisite strength; describe its limits rather than generalizing to all learners. Have two reviewers independently score a sample, reconcile disagreements and revise ambiguous criteria.
Current status: authored and locally executable; independent review, cross-platform execution, observed workload and learner transfer remain pending. Record real dates/results here when those activities occur. Website builds verify rendering and downloads, not classroom effectiveness or a deployed service.