Nothing in this course is marked against a standard you have not seen. Every rubric below is the sheet the work is actually marked on, published before the work is set.
| Criterion | Weight |
|---|---|
| Research + Discovery | 20 % |
| Process + Practice | 20 % |
| Presentation + Production | 30 % |
| Criticality + Agency | 10 % |
| Class Participation | 20 % |
| Total | 100 % |
| Instrument | Weight | Set | Due | Criterion |
|---|---|---|---|---|
Research Pack + Project Notebook research.md | 20 % | W01 · pack W03 | Pack W05 · W06 read · final | Research + Discovery |
Project Notebook — decisions.md / patterns.md | 15 % | W01 | W06 read, final | Process + Practice |
| Probe — Intelligent Interface | 5 % | W06 | W09 | Process + Practice |
| Proposal (presentation + one page) | 5 % | W04 | W06 | Presentation + Production |
| Final Intervention + presentation | 25 % | W06 | Documentation W14 · presentation in the assessment period | Presentation + Production |
| Position Paper (1,200 words) | 10 % | W06 | W13 | Criticality + Agency |
| Participation — crit log | 15 % | W01 | continuous, W01–W11 and the final presentation session | Class Participation |
| Participation — peer-directed weekly write | 5 % | W01 | weekly | Class Participation |
The marking sheet for every one of these is published in Week 5 — three rubrics, in two versions each (the sheet the instructor marks on, and an identical self-assessment version), plus one checklist and one published statement of the participation bands. Nothing in this course is marked against a standard you have not seen.
The Research Pack and the Project Notebook are checked and banded rather than scored on a criterion matrix: a tick list of what has to be there, then one of four bands — complete and rigorous, complete, partial, not done. The Proposal, the Probe, and the Final Intervention with the Position Paper are scored on criterion rubrics.
Required to pass, ungraded: the Week 6 proposal, the Adjacent-Design checkpoint (W11), an AI Use Statement on every submission, and the Week 12 refusal write, which is returned with comment.
Eight things about these sheets that are not obvious from looking at one.
Fillable PDFs. Set and due dates, weights and criteria are in the table above.
Checked and banded rather than scored: a list of what has to be there, then one of four bands — complete and rigorous, complete, partial, not done. Version history and traced sources are a pass/fail gate checked before anything else is read.
Whether a hundred seconds and one page make the case for seven weeks of work. Worth 5% and required to pass — the real cost of a vague proposal is the seven weeks, not the five marks.
Marked on the four lenses the Week 9 panel reviews you through, so the feedback in the room and the mark are the same instrument.
One sheet, two criteria, marked in two passes — the Position Paper after Week 13 so you have it back before the closing session, the intervention after you present. Documentation fidelity, not artifact polish: polish tracks equipment and time more reliably than it tracks ability.
The one graded criterion with no sheet. This page is the instrument — it is published in week one and it is what the mark is made against.
Marked against the criterion’s own wording: being useful to other people’s work. Not attendance, not talking the most. Two records, both kept as they happen — a band from the crit log, and a pass/fail tally of the weekly write. There is no sheet to fill in; you get a written mark and a paragraph twice, after W06 and after W11, so nothing in December is a surprise.
There is no participation rubric to fill in. This page is the instrument. It is published in week one and it is what I mark against.
15% OF FINAL GRADE
A 0/1/2 grid per student plus a note when something is notable, filled in for two minutes at the end of each session. Observation window is W01–W11 plus the finals session — eleven instructor-present sessions.
4 — Generative. Other people’s work is measurably better for your presence. You bring material to crit that is not your own. You name the thing nobody wanted to say, and you do it kindly.
3 — Reliable. Present, prepared and specific. You respond to the actual work in front of you rather than to the topic in general. This is what doing it properly looks like, and it is a good mark.
2 — Passive. You attend, and you respond when asked directly. Your contributions are general enough that they would apply to any project in the room.
1 — Absent. Not there, or there and disengaged. Other people’s work is unaffected by your being in the room.
5% OF FINAL GRADE
150 words. It must respond to a named classmate’s work from that session. That is what keeps it from becoming a reflective diary, which measures Criticality and is already covered elsewhere. If it could have been written without attending, it does not pass. W10 is reading break.
Sessions counted: W01, W02, W03, W04, W05, W06, W07, W08, W09, W11, W12, W13, W14.
What each piece of work is, as opposed to how it is marked. Each brief’s “how this is marked” section is generated from the rubric, so the two cannot disagree.
Go and talk to people in your domain, and come back with something a model could not have told you.
One page and five slides. A hundred seconds to say what you are going to make, for whom, and why it is worth seven weeks.
Build an interface with a model in the loop that asks a question about your domain. It does not have to work. It has to be legible.
1,200 words saying what you think a technology should or should not do, and defending it against the best argument on the other side.
The intervention you proposed at week six, built — plus five minutes to make the case for it.
Given out in the room. Here as well, so losing one is not a problem.
The alternatives and accommodations, on one page. Week 1.
What goes into a context that actually works, and what it becomes. Week 2.
The template that goes on every submission — including the part that gets read, what you rejected.
Where the model helps, where it is plausibly wrong, and what a trained eye catches.
Built as rings in this course, because the distance from the transaction is the point.
A journey map is about a person. A flow is about a system.
Written before the session, not after.