Getting started with Discovery — verifiable open science

Discovery is Turtini's verifiable open-science module. The whole point: turn "trust me" into "verify me". Every experiment carries an immutable evidence chain rooted in a hash-stamped pre-registration PDF — so any third party can confirm what you said you'd do BEFORE you ran the experiment, and what you actually did, and that nothing was rewritten after the fact.

**The crises Discovery attacks:**
- **Reproducibility** — about half of published bio results don't reproduce. Discovery ships container-on-demand re-runs of your analysis pipeline so anyone can re-execute and see a diff.
- **Negative results** — most labs never publish the experiments that didn't work, so the next lab re-runs the same dead-end. Discovery's public registry gives positive AND negative results the same DOI treatment.
- **P-hacking / HARKing** — pre-registration with a tamper-proof timestamp is the gold-standard counter. Discovery locks pre-regs as Verified-by-Turtini PDFs with a permanent SHA-256 you can re-check at turtini.com/verify.

**Where it lives:**
- Org-context lab? Settings → Modules → enable Discovery. The module appears in nav at /discovery.
- Independent researcher? Open Personal context → Account → Personal Modules → enable Discovery. Same /discovery URL; your experiments are owned by you, not by an org.

**The lifecycle:**
1. **Draft** — author the pre-registration (Hypothesis · Background · Methodology · Analysis Plan · Success Criteria · Predicted Outcome). Private to you / your org members while drafting.
2. **Lock** — click Lock Pre-Registration. Turtini renders a PDF, stamps it with a verification QR, registers its SHA-256, mints a DOI like `k7bvr3-9xqz`, and the experiment becomes public at /discovery/exp/{doi}.
3. **Active** — log daily notebook sessions (hash-chained), bind sensors so live readings stream into the chain, upload datasets. Wally analyzes incoming data and pings you if anything drifts past your pre-registered thresholds.
4. **Completed** — publish a Result (positive OR negative — same registry, same DOI weight). Pre-reg drift between what you promised and what you did is auto-flagged so reviewers can see it.
5. **Reproduce** — anyone can kick off a Python or R container re-run of your analysis to confirm the numbers.

**Verifying a Discovery PDF from outside Turtini:**
Drop the PDF onto turtini.com/verify. The page reads its SHA-256, looks it up in the pdfVerifications registry, and shows you when it was stamped, who issued it, and whether it's been revoked. This works for non-Turtini users — the verification endpoint is unauthenticated by design.

**Citizen Science crossover:**
Personal-context Discovery users can publish observations as Citizen Datasets that indie researchers (in org or personal context) can pull into a new experiment. A birder's bird-count log becomes an ecology lab's dataset. A backyard astronomer's transit photos become a planetary-science lab's reference set.