APPENDICES
Tools, setup, and the working vocabulary for this course.
Building Your Capstone Workbench
Getting a repository, a board, an hours log, and a place to run code — including on a locked-down laptop with no administrator rights
The on-ramp for the whole course, built around the cheapest path that works. A decision table keyed to your actual situation — admin rights or none, capable machine or not — leads to a browser-only path (GitHub plus a cloud development environment, nothing installed, no admin) or a local path for students who have the rights and the hardware. Then the four pieces every capstone needs regardless of stack: the repository and its structure, the project board wired to requirement identifiers, the hours log, and a diagramming tool that produces something you can put in a specification. Closes with a milestone-to-minimum-environment table so nobody over-provisions, a free-tier watch list with a teardown checklist, and a verification pass that proves the workbench works before Week 1 ends.
The Document Kit: Templates and Worked Examples
Every artifact this course requires, as a fill-in template and as a worked example — with the bad version shown next to the good one
The reference you will open every week. Each of the course's deliverables appears three times: a blank template you can copy, a fully worked example from a realistic student project, and a deliberately bad version annotated with what is wrong and what it will cost you. Covers the project charter, requirements specification with functional and non-functional sections, acceptance criteria, architecture decision record, work breakdown structure, schedule and burn-down, risk register, test plan, defect report and log, hours log, README, runbook, onboarding and handoff guide, change log, and retrospective. This appendix is where the course's promise to show you what good actually looks like is kept.
The Grading Contract
Exactly how the 50 / 25 / 15 / 10 split is earned, the full final rubric, the submission checklist, the pace table, and the AI-use policy
No surprises in Week 16. This appendix states the grading contract in full: the final implementation and its documentation at 50 percent with the complete rubric line by line and what evidence earns each line, the fourteen weekly milestones at 25 percent, the 30-minute presentation at 10 percent with its own rubric, and the sixteen weekly quizzes at 15 percent. It gives the week-by-week pace table that tells you whether you are on track, the complete final submission checklist, the repository structure a grader expects to find, the AI-use and disclosure policy with the ai-usage log format, the academic-integrity line, and a plain account of what falling a week behind actually costs — plus how the milestones and the final grade relate, so nobody mistakes a weekly milestone for optional work.
Fifty Capstone Ideas, Sized
A catalog of right-sized senior projects across a dozen domains, each with a scope estimate, the hard part, and how to shrink or grow it
For the student who has the skills but not the idea, and for the student whose idea is three times too big. Fifty candidate projects spanning web and mobile applications, data and analytics, machine learning and applied AI, developer tooling, security, systems and networking, embedded and hardware, games and simulation, accessibility, civic and nonprofit software, education technology, and research replication. Each entry names the problem and the user, sketches a minimum viable scope that fits roughly 240 hours, identifies the genuinely hard part, lists the skills it will stretch, and gives one way to make it smaller and one way to make it more ambitious. Use it to generate candidates in Week 2 — then do the scoping work yourself.
Glossary
The vocabulary of the software development life cycle, project management, and delivery — in the sense this course uses it
Every term the course uses as a term of art, defined the way a practitioner means it and cross-referenced to the week that teaches it: from acceptance criteria, architecture decision record, and baseline through burn-down, change control, definition of done, elicitation, functional and non-functional requirement, MoSCoW, regression, release candidate, risk register, traceability matrix, walking skeleton, and work breakdown structure. Includes the AI-era additions this course adds — ai-usage log, novelty load, prompt contract, clean-machine test, and bus factor.