Work sign-up¶
How to use: tell the team what you want to work on before each sprint planning with the Work sign-up form. At planning, sign-ups become the assignment matrix in section 7. Then each person claims specific issues on the board.
1. Deadlines¶
| Planning meeting | Sign-ups due |
|---|---|
| Sprint 1 planning, Thu 2026-09-17 | Wed 2026-09-16, end of kickoff |
| Sprint 2 planning, Thu 2026-10-08 | Wed 2026-10-07, after the retrospective |
| Sprint 3 planning, Thu 2026-10-29 | Wed 2026-10-28, after the retrospective |
| Finalization planning, Mon 2026-11-30 | Wed 2026-11-25 |
2. Workstreams and P-01 deliverables¶
| ID | Workstream | Covers | Helpful skills | Primary sprint | P-01 deliverables | Seed issues |
|---|---|---|---|---|---|---|
| W1 | Research framing and related work | Research question, hypotheses, competing explanations, literature, threats to validity | Reading papers, writing | Sprint 1 (related work continues in Sprints 2 and 3) | Threat model; precise research question and sub-questions; competing explanations; related work on software supply chains and prompt injection | #12, #18, #32 |
| W2 | Data acquisition and data dictionary | Download scripts, sampling, inclusion rules, MANIFEST, DATA_DICTIONARY.md | SQL, pandas | Sprint 1 | Inclusion rules (content fetched, symlink stubs excluded, front-matter-valid sensitivity subset); DATA_DICTIONARY.md Part B; snapshot statement |
#13, #14 |
| W3 | Extraction pipeline and loaders | Code that turns raw tables into analysis tables | Python, SQL | Sprint 1 to Sprint 2 | Rule file reviewed; scanner and variant linking inside python -m msr_pipeline analyze; make pipeline end to end |
#15, #21 |
| W4 | Analysis and statistics | Comparisons, models, robustness and sensitivity checks | Statistics | Sprint 2 to Sprint 3 | RQ1 prevalence; RQ2 reach tests (Mann-Whitney U, bootstrap intervals, negative binomial regression); RQ3 variant drift; sensitivity checks | #24, #27, #28 |
| W5 | Validation and annotation | Annotation protocol, labelling, agreement, error analysis | Careful reading, domain knowledge | Sprint 2 (error analysis in Sprint 3) | Primary and second-rater labels with the annotation kit; blindness rule; precision per rule; inter-rater Cohen's kappa; adjudication of disagreements | #23, #29 |
| W6 | Testing, CI, and reproducibility | Unit tests, CI, fresh-clone runs, release | pytest, GitHub Actions | Every sprint | Rule regression tests; gemba walk each review; release tags | #22, #33, #37 |
| W7 | Visualization and results tables | Figures and tables generated by code | matplotlib | Sprint 1 and Sprint 3 | Exploratory prevalence figure; final figures/rq*_*.png from the pipeline |
#16, #30 |
| W8 | Report writing | report/draft.md, citations, observation versus interpretation |
Writing | Sprint 2 to Finalization | Background, method, and preliminary results (S2); complete draft with a safe-reporting section (S3); final report | #25, #31, #36 |
| W9 | Presentation and demo | Slides, live demo, rehearsal | Presenting | Sprint 3 to Finalization | Demonstration plan with speakers; demo that runs the pipeline from a fresh clone; rehearsals | #34, #38 |
| W10 | Process and Lean Six Sigma metrics | Board hygiene, metrics dashboard, retrospectives, kaizen | Facilitation | Every sprint | Planning, check-ins, reviews, retrospectives; metrics dashboard; FMEA-style ranking of capability categories | #19, #20, #26, #35, #40 |
3. How choices become assignments¶
At sprint planning the Product Owner and Scrum Master apply these rules in order and record the result in section 7.
- Every workstream active in the sprint gets one lead and at least one support.
- First choices are honoured where possible.
- Each member holds at least one research item (W1, W4, W5) and one engineering item (W2, W3, W6, W7) per sprint (charter section 8).
- Nobody leads more than two workstreams in the same sprint.
- If a member's contribution share was outside 15% to 45% last sprint (metrics dashboard), rebalance toward them or away from them.
- Tie-break: the member who received a lower choice last sprint goes first. Otherwise the Product Owner decides and the decision is logged in DECISION_LOG.md.
Across the semester, each member should lead at least one research workstream and one delivery workstream.
4. Claiming a specific issue¶
- Assign yourself on the issue, move it to In progress, and comment
Claiming, target <date>. - Hold at most two issues in progress at once (team WIP limit is five).
- If a claimed issue has no activity for 48 hours, the Scrum Master checks in. After 72 hours it returns to Todo unless you say otherwise.
- To swap work, comment on both issues and tag the other person.
- Stuck for more than a day? Open a Blocker issue.
5. Pairing¶
Pairing is encouraged, especially across strengths (for example a statistics-strong member with a writing-strong member on W4). List a pair in the matrix as Lead + Support. Both names go on the issue as assignees so the contribution is visible. Pairs cannot approve each other's pull requests for work they co-authored; a third member reviews.
6. Validation second raters¶
Decided at the kickoff on 2026-09-16 (decision D-020) and revised on 2026-09-23 (D-022) and 2026-10-16 (D-024). Jerad Dunne (jd) is the primary rater for every kind.
| Kind | Items | Second rater | Rater id | Round 1 due |
|---|---|---|---|---|
| Signals | 147 items, 209 label rows | none: Jerad Dunne re-rates (intra-rater, D-022, D-024) | jd round 2 |
Round 1 before scoring on Tue 2026-10-27; round 2 in Sprint 3, at least 14 days later |
| Lineage | 58 pairs | Hina Kramer (D-024) | hk |
Thu 2026-10-22 (proposed); jd commits round 1 after Hina Kramer's file is merged |
| Drift | 18 pairs | Leticia Aderhold | la |
Labels committed 2026-10-08, pull request #90 |
Protocol: docs/validation/README.md.
7. Assignment matrix¶
Filled at each sprint planning. Use names or handles. L = lead, S = support.
| Workstream | Sprint 1 | Sprint 2 | Sprint 3 | Finalization |
|---|---|---|---|---|
| W1 Research framing and related work | ||||
| W2 Data acquisition and data dictionary | ||||
| W3 Extraction pipeline and loaders | ||||
| W4 Analysis and statistics | ||||
| W5 Validation and annotation | ||||
| W6 Testing, CI, and reproducibility | ||||
| W7 Visualization and results tables | ||||
| W8 Report writing | ||||
| W9 Presentation and demo | ||||
| W10 Process and Lean Six Sigma metrics |
8. Where to see current assignments¶
- The project board (Assignees column) is the live view.
- DIGEST.md lists open issues per person and every sign-up. Regenerate with
python scripts/workspace_digest.py.