spb/wp20_uqo

1# Data rights, provenance and ethics — WP2023## 1. Assessment roll (ADMIN)45- Origin: « Rôle d'évaluation foncière du Québec », Ministère des Affaires municipales et de l'Habitation, published on Données6  Québec under **CC-BY 4.0** (CKAN metadata read 2026-10-20; `docs/institutional_sources.md`). Owner names, cadastre and postal7  addresses are redacted at source and re-identification is prohibited; our copy (`vraiprix.db` → `units`) contains no owner data.8- Use in WP20: unit characteristics and assessed land/building values as verified administrative fields; attribution given in the9  paper. Civic addresses are used only to match listings and are never printed.1011## 2. Transactions (Registre foncier via Vrai Prix)1213- Used only in aggregate (Experiment 15: ratio of recorded sale price to cost indication for ≈170 frame units sold after listing).14  No individual transaction is reproduced. Vrai Prix's own policy (Toveli integration matrix) is "not republished in bulk".1516## 3. Listing text and broker fields (TEXT)1718- Collected by the author's Immo-Ka connectors for the Vrai Prix/Immo-Ka platforms (not by this project). Descriptions are19  copyrighted marketing texts of brokers/owners; WP20 processes them automatically and reports only aggregate statistics and20  short anonymised fragments (≤ 15 words) in the error taxonomy. Prices, taxes and municipal assessments are masked before any21  model sees the text (`src/extraction/textprep.py`).2223## 4. Listing photographs (VISUAL)2425- Copyright belongs to the listing broker/owner/photographer. WP20 **does not download, store or redistribute** any26  photograph: images are passed to the model API by URL and are not kept (`data/images/` is git-ignored and unused); no listing27  photograph appears in the manuscript. Illustrations in the paper are schematic drawings or synthetic images clearly labelled.28- **Robots.txt policy**: hosts whose robots.txt disallows all automated retrieval (`media.remax-quebec.com`) were excluded from29  the extraction frame (rule R10) rather than circumvented; the model provider's fetcher enforces the same rule. Other hosts30  either publish no robots.txt or do not disallow image paths (`results/raw/robots_check.txt`). Terms of use of individual31  broker sites were not reviewed one by one; this is recorded as a residual legal uncertainty, and the analysis is a32  non-commercial academic study of aggregate extraction behaviour.33- Privacy: photographs may show personal belongings or people; the extraction schema contains no field about occupants, and the34  `evidence` free-text field is instructed to name building cues only. Outputs are reviewed for incidental personal information35  before release (`scripts/` check: no names, plates or faces are requested or stored).3637## 5. Model outputs and API usage3839- Three Anthropic models via the Message Batches API; prompts, schema versions and model ids are versioned40  (`configs/prompts.yaml`, `configs/models.yaml`, `results/raw/batches/run_meta.json`). API keys live only in an untracked file41  outside the repository; `results/raw/api_usage.jsonl` records tokens, never content.42- Model outputs are treated as **AI claims**, never as ground truth; they are compared with administrative and broker fields.4344## 6. What is NOT established4546- No human expert annotations exist; the annotation protocol (`docs/annotation_protocol.md`) is a proposal.47- The representativeness of the frame is limited (listings from 65 connectors, Aug–Oct 2026, RE/MAX excluded by R10, Kijiji48  and Facebook excluded by R8); no Québec-wide generalisation is claimed.49