25%, Individual. All semester.

Participation

Participation here is three graded activities, each with its own weight, and two that are worth bonus points only. It is graded on showing up prepared with something real, not on who talks the most.

Attendance and discussion (8%)

This course happens in the room, in the debates where your assigned side is deliberately your unassigned view, and in the labs that produce the deliverables on the Relay Wall. Being present and prepared, arguing the side you are given, and doing the lab work with your group. Nothing else is asked for here. Absence logistics and accommodations live on the the attendance policy.

Problem Inventory (10%)

The inventory is a shared, running record of real service problems you have seen with your own eyes: at a job, a volunteer role, a board you serve on, or from your own experience of these systems, alone or with family. It opens in Week 2 and lives as a shared document linked on Canvas for the rest of the term. Entries are written in your own time and posted there; no class time is spent drafting them.

Entries log a situation rather than a solution. The four fields below ask what happened, who it affected, why it persists, and whether technology is appropriate, in that order; there is no field for a proposed fix, and “tech seems irrelevant here” is a complete, welcome answer.

Entry format

Each entry has four fields, plus a rough domain tag:

1. The situation. What happens, concretely. A specific scene beats a general complaint: “new clients tell their full story three separate times during intake” beats “intake is inefficient.”
2. Who is affected. Name the people who bear the weight: clients, frontline staff, supervisors, the organization, funders.
3. Why it persists. What sustains the problem: workload, funding rules, old systems, policy, habit, turnover, or no one owning it.
4. Is tech relevant? Yes, no, or unsure, with one line of reasoning. “No, this is a staffing problem” is one of the most useful entries the inventory can contain.

Requirement and grading

Five entries across the semester: at least two by Week 6, and at least two more after Week 9, once the interview with your organization is underway and pairs start seeing its problems firsthand. Entries are graded on quality, not volume: a specific scene, a specific who, an honest account of why the problem persists, and reasoned tech relevance. The Week 8 check looks for at least two entries in, with feedback and no separate grade.

Norms

• No client-identifying details. No names, and no combination of details that could identify a person, even without a name.
• Org names optional. Name the organization only if you are comfortable and the entry says nothing they’d consider a confidence.
• Firsthand preferred, secondhand labeled. If you heard it from a colleague rather than saw it, say so.
• No blame entries. Describe the system rather than a colleague: “notes get thinner as caseloads grow” is an entry, and “my supervisor is lazy about notes” is a complaint about a person.
• Lived experience counts. Your own passage through a benefits system, for yourself or family is direct observation, and logging it creates no pressure to disclose more than the entry needs.
• No entry is a request for solutions. Logging a problem is not volunteering to have the class solve it.

Uses across the course

Entries feed the Ethics Case (case picks and mitigation reality-checks), the Strategy Pitch (problem definitions at partner intake), and the two Debate Labs, where inventory entries substitute for hypotheticals.

Weekly question (7%)

Answer that week’s question on its discussion board by Tuesday at 5:00pm, in three to five sentences or two minutes of audio. The reply editor records audio, so either mode is a reply on the same board. We open Wednesday’s session with a few of the answers, so take something from the week, an argument, a case, a tool you tried, a reading that annoyed you, and answer from it.

There is one question a week, and they are asked in order:

1. Whose decisions end up inside a technology, and can the people using it see them?
2. What does a case record do to the person it is about?
3. When a decision moves into software, where does the discretion go?
4. What should a machine be allowed to decide without a person checking it?
5. What would you teach the people you work with about when to trust a machine’s answer?
6. What is lost when a task is made easier?
7. When is it right to leave someone out of the data?
8. Who is the real user of a system, the person in front of it or the one who bought it?
9. What has to already be true in an organization before new technology helps it?
10. Who has standing to tell a story that belongs to someone else?
11. When a design is called accessible, who is it accessible to?
12. What does a technology make invisible, and to whom?
13. Whose decisions end up inside a technology, and can the people using it see them?

Every one of them can be answered about almost anything the session presents. The question remains broad while the material under it changes, so a December answer and a September answer to the same kind of question can be compared.

Answer on that week’s board in Canvas by 5:00 pm on the Tuesday before class. Three to five sentences is enough. If you would rather talk than type, the reply box takes a recording straight from a phone or laptop browser, and a voice memo uploaded as a file works too; either counts the same.

The deadline falls before the session on purpose. You commit to an answer before the seminar works on it, and we open Wednesday with a few of them, so your classmates read what you wrote. Two unrehearsed minutes of thinking about a system is closer to how you will talk about technology in a staff meeting than a polished paragraph is.

Answers are graded complete or incomplete. An answer is complete when it takes the week’s question seriously; there is no penalty for changing your mind or for refusing to change it. Week 1 is not graded, and of the twelve that are, the four lowest drop, so the best eight count and a missed week needs no explanation. The Week 13 answer is due by the end of finals week.

Design Walls (bonus)

Optional, and worth up to three bonus points on the participation grade. Nobody loses anything by never posting.

Four shared class boards are open all semester: the Wall of Shame (design that excludes), the Wall of Curb Cuts (design aimed at one group that benefited everyone), From the Field (tech-for-good news you spot as it happens), and What It Got Wrong (the errors your own AI tools produced). At least four substantive posts across the semester, spread over at least two walls.

The walls live online as four discussion boards on Canvas, one per wall, alongside the physical wall in the classroom. Post online so the walls continue to grow between sessions; the best finds are pinned to the physical wall in class.

A post is an image or a link and one sentence. The grade is on the sentence, which has to name the decision behind the design and say who that decision includes and who it excludes. Four of them, one per wall:

• Wall of Shame: “This state benefits portal logs you out after 15 idle minutes and deletes the half-finished application, a session-timeout decision that excludes anyone applying on a shared phone, between shifts, or with a screen reader that moves slower than the timer.”
• Wall of Curb Cuts: “OXO’s Good Grips peeler was designed for one woman with arthritis, and the fat rubber handle decision now includes anyone with wet, cold, or tired hands.”
• What It Got Wrong: “I asked an assistant for the appeal deadline on a Medicaid denial and it gave me 30 days with a confident citation to a state page that says 90, which I only caught because the page it named did not exist.”
• From the Field: “My county library now lends wifi hotspots for weeks at a time, a circulation-policy decision that includes households who lost home broadband when the Affordable Connectivity Program ended.”

A session timeout, a handle spec and a lending policy are decisions, while a benefits portal is a product.

In Week 13 the class votes the Curb-Cut of the Semester and the Most Instructive Failure. Strong posts become case seeds for the Ethics Case and as examples for the Strategy Pitch, so the walls compound into the graded work instead of standing apart from it.

Event reflections (bonus, up to 3 points)

Attending an event I share on Canvas and posting a short reflection within a week earns a bonus point, up to three. The points go to the final course grade rather than to this component, so they can offset a loss anywhere in the course. Full terms are under bonus points.