# Legal Policy & Copyright Guidelines

**Project**: Universal Science Discovery Repository (USDR)  
**Last Updated**: May 2026

## Core Principle

**We never host full-text copyrighted scientific papers or other restricted content.**

This repository is designed to be 100% legally compliant while maximizing scientific value. We achieve this by focusing exclusively on:

- Metadata, DOIs, titles, authors, abstracts (where fair use or open license applies)
- Community-created summaries, analyses, and syntheses
- Structured data (YAML, JSON-LD, RDF)
- Code, notebooks, and derived datasets under open licenses
- Links to legal sources

## What Is Allowed

### ✅ Permitted Content
- **Metadata & Bibliographic Information**: Titles, authors, DOIs, PMIDs, publication dates, journal names, abstracts (when permitted by publisher policy or fair use)
- **Open Access Content**: arXiv preprints, PubMed Central articles, DOAJ journals, CC-licensed works (with proper attribution)
- **Community Contributions**: Original summaries, key findings tables, critiques, hypothesis formulations, gap analyses (transformative works)
- **Public Domain & CC0 Data**: Government datasets, CC0 releases, your own original work
- **Derived Works**: Re-processed data, re-plotted figures (with attribution), code that reproduces results
- **Links & Pointers**: URLs to legal sources, with clear "Open Access Status" indicators

### ❌ Prohibited Content
- Full-text PDFs or HTML of paywalled papers
- Systematic scraping or bulk downloading of publisher content
- Content that violates the terms of service or license of any linked resource
- Personally identifiable information from research participants (even if "public")
- Content that could enable re-identification of anonymized subjects

## Licensing

| Content Type              | License                          | Notes |
|---------------------------|----------------------------------|-------|
| Code & Scripts            | MIT                              | See `LICENSE` file |
| Structured Data (YAML, etc.) | CC0 1.0 Public Domain         | Maximum reuse |
| Original Markdown Content | CC0 1.0 (recommended) or CC-BY-4.0 | Contributor choice |
| Summaries & Syntheses     | CC-BY-4.0 (default)              | Requires attribution |
| Third-Party Metadata      | Original source license          | We add no new restrictions |

## Third-Party Sources & Attribution

When adding content from external sources:

1. Always include the original DOI or stable identifier
2. Clearly state the license/status (e.g., "Closed access – metadata only", "CC-BY-4.0 via arXiv")
3. Use tools like Unpaywall, OpenAlex, or CORE to identify legal open versions
4. Never bypass paywalls or use unauthorized copies

## Enforcement & Takedowns

- We respond promptly to valid DMCA notices and copyright complaints
- Content will be removed or modified within 48 hours of a verified complaint
- Repeated or willful violations may result in contributor restrictions

**To report a potential copyright issue**, please open a private issue or email **legal@usdr.dev** with:
- URL of the problematic content
- Description of the claimed infringement
- Your contact information (for follow-up)

## Data Protection & Privacy

- We do not collect or store personal data beyond what is necessary for Git attribution (name + email in commits)
- Research data involving human subjects must be properly anonymized or have explicit consent for open sharing
- We respect "right to be forgotten" requests for contributor information where legally required

## International Considerations

This project is hosted in the United States. Contributors and users are responsible for complying with the laws of their own jurisdictions, including:
- Database rights (EU)
- Moral rights (some jurisdictions)
- Indigenous knowledge protections
- Export control regulations (for dual-use research)

## Questions?

Open an issue with the `legal` label or email **legal@usdr.dev**.

We are committed to being good stewards of open science while respecting the legal rights of publishers, authors, and data subjects.