Skip to content

FAQ.en.md

Latest commit

 

History

History
377 lines (261 loc) · 11.3 KB

FAQ.en.md

File metadata and controls

377 lines (261 loc) · 11.3 KB
Raw

Frequently Asked Questions (FAQ)

Languages / 语言: English | 简体中文

General Questions

What is DLRS Hub?

DLRS Hub (Digital Life Repository Standard Hub) is a standardized system for managing digital life records. It provides a framework for storing, organizing, and controlling access to digital representations of individuals, including voice samples, images, videos, and personal data.

Who is DLRS Hub for?

DLRS Hub is designed for:

  • Individuals who want to preserve their digital legacy
  • Families managing memorial records of deceased loved ones
  • Researchers studying digital identity and AI ethics
  • Organizations building digital human applications
  • Developers creating AI voice/avatar cloning systems

Is DLRS Hub free to use?

Yes, DLRS Hub is open source under the MIT License. You can use, modify, and distribute it freely.

Technical Questions

Why not put videos directly in GitHub?

Answer: Raw audio, video, facial data, voiceprints, and identity documents are highly sensitive data. GitHub is not suitable for storing large binary files because:

  1. Size limitations: Git repositories become slow and bloated with large files
  2. Permanent history: Once committed, files are difficult to completely remove from Git history
  3. Privacy risks: Sensitive biometric data should not be in public version control
  4. Compliance: Many jurisdictions require specific security measures for biometric data

Instead, DLRS Hub stores pointer files (.pointer.json) that reference data in secure object storage (S3, OSS, etc.), while keeping metadata and audit information in Git.

What is a pointer file?

A pointer file is a JSON file that contains metadata about a file stored elsewhere:

{
  "artifact_id": "voice_001",
  "type": "voice_sample",
  "storage_uri": "s3://my-bucket/audio/voice.wav",
  "checksum": "sha256:abc123...",
  "size_bytes": 1048576,
  "sensitivity": "S3_BIOMETRIC"
}

This approach separates sensitive data from version control while maintaining traceability.

What object storage services are supported?

DLRS Hub supports any S3-compatible object storage:

  • AWS S3 (International)
  • Alibaba Cloud OSS (China)
  • Tencent Cloud COS (China)
  • Google Cloud Storage (International)
  • MinIO (Self-hosted)
  • Backblaze B2 (International)

How do I validate my record?

Run the validation tool:

python tools/validate_repo.py

This checks:

  • JSON schema compliance
  • Required fields presence
  • File structure correctness
  • Sensitive file detection
  • Consent documentation

Privacy and Consent

Must human records be public?

No. Records are private by default (visibility: private). Public indexing requires:

  • Explicit authorization (allow_public_listing: true)
  • Verified consent badge
  • Manual review approval

You can also use public_unlisted for records accessible via direct link but not searchable.

How do I withdraw my consent?

Contact the endpoint specified in your manifest.json:

"consent": {
  "withdrawal_endpoint": "mailto:privacy@example.org"
}

The system will:

  1. Freeze the runtime state immediately
  2. Preserve audit logs (required by law)
  3. Delete personal data after retention period
  4. Remove from public indexes

Can minors create records?

Yes, but with restrictions:

  • Guardian consent required (consent.guardian_consent: true)
  • Cannot be publicly indexed
  • Additional privacy protections apply
  • Must comply with COPPA (US), GDPR (EU), or local laws

What about deceased individuals?

Records of deceased individuals require:

  • status: "deceased" in manifest
  • profile/inheritance_policy.json specifying heirs
  • Executor contact information
  • Default action: freeze the record

Usage and Rights

Does a digital life equal the actual person?

No. This is critically important to understand:

  • All outputs are AI-generated or AI-assisted content
  • They do not represent the real person's immediate authentic thoughts
  • They are simulations based on training data
  • All outputs must be labeled as AI-generated

Think of it as a digital memorial or assistant, not a replacement for the actual person.

Can I use someone else's record commercially?

Only if explicitly authorized. Check the record's manifest.json:

"rights": {
  "allow_commercial_use": true,  // Must be true
  "allow_voice_clone": true,     // For voice applications
  "allow_avatar_clone": true     // For avatar applications
}

Unauthorized commercial use is:

  • Violation of the record's license
  • Potentially illegal (right of publicity, personality rights)
  • Subject to takedown and legal action

What can I do with a public record?

It depends on the permissions granted:

Permission Allowed Use
allow_public_listing View in public index
allow_model_finetune Train AI models
allow_voice_clone Create voice synthesis
allow_avatar_clone Create digital avatars
allow_commercial_use Commercial applications
allow_research_use Academic research

Always check the specific record's permissions before use.

How do I verify a record's authenticity?

Check for these indicators:

  1. Verified Consent Badge (review.verified_consent_badge: true)
  2. Provenance Information in audit/provenance.json
  3. Consent Evidence in consent/ directory
  4. Review Status (review.status: "approved_public")
  5. Audit Trail in audit/access_log.pointer.json

Be cautious of records without these verifications.

Compliance and Legal

Is DLRS Hub GDPR compliant?

DLRS Hub provides tools and frameworks for GDPR compliance, but:

⚠️ You are responsible for ensuring your specific implementation complies with GDPR and other regulations.

The framework supports:

  • Right to access (data export)
  • Right to erasure (deletion policy)
  • Right to rectification (version control)
  • Data portability (JSON format)
  • Consent management (consent records)

Recommendation: Consult with a lawyer before deploying in production.

What about CCPA, PIPEDA, or other privacy laws?

The same principle applies: DLRS Hub provides the framework, but you must ensure compliance with applicable laws in your jurisdiction.

Consider:

  • US: CCPA (California), COPPA (children)
  • Canada: PIPEDA
  • China: PIPL (Personal Information Protection Law)
  • Japan: APPI (Act on Protection of Personal Information)
  • South Korea: PIPA (Personal Information Protection Act)

Do I need a lawyer?

Yes, if you're deploying this in production. Especially for:

  • Public-facing services
  • Commercial applications
  • Cross-border data transfers
  • Services involving minors
  • Memorial/deceased records

The templates in policies/ and templates/consent/ are starting points only, not legal advice.

Badges and Review

What are badges?

Badges indicate verification status:

  • 🔵 verified-consent: Consent has been verified by reviewers
  • 🟢 public-data-only: Contains only public information (no biometrics)
  • 🟡 research-approved: Approved for academic research
  • 🔴 commercial-licensed: Licensed for commercial use

How do I get a verified-consent badge?

  1. Submit a record with complete consent documentation
  2. Include consent statement (consent/consent_statement.md)
  3. Provide consent video or signature (consent/consent_video.pointer.json)
  4. Pass automated validation
  5. Pass manual review by maintainers
  6. Badge is issued in review.verified_consent_badge

What is the review process?

  1. Automated validation: Schema, structure, sensitive files
  2. Human review: Consent verification, content moderation
  3. Badge issuance: Verified badges added to manifest
  4. Public indexing: Record added to registry/humans.index.jsonl
  5. Ongoing monitoring: Periodic re-verification

Troubleshooting

My validation is failing. What should I do?

  1. Read the error message carefully
  2. Check the example record: humans/asia/cn/dlrs_94f1c9b8_lin-example/
  3. Verify all required fields are present
  4. Ensure JSON syntax is valid
  5. Run python tools/check_sensitive_files.py
  6. Check the Getting Started Guide

I accidentally committed a large file. How do I remove it?

# Remove from Git history (use with caution!)
git filter-branch --force --index-filter \
  "git rm --cached --ignore-unmatch path/to/large/file" \
  --prune-empty --tag-name-filter cat -- --all

# Force push (only if you're sure!)
git push origin --force --all

Better: Use .pointer.json files from the start!

How do I update an existing record?

  1. Edit the files in your record directory
  2. Update audit.last_modified_at in manifest.json
  3. Run validation: python tools/validate_repo.py
  4. Commit and push changes
  5. If public, the registry will be updated automatically

Contributing and Community

How can I contribute?

See the Contributing Guide for details. You can:

  • Add your own record
  • Improve documentation
  • Fix bugs in tools
  • Translate to other languages
  • Propose standard improvements

Where can I get help?

Can I use DLRS Hub for my project?

Yes! DLRS Hub is open source (MIT License). You can:

  • Use it as-is
  • Fork and modify it
  • Build commercial services on top
  • Contribute improvements back

Just ensure you comply with applicable privacy laws.

Advanced Topics

What is the difference between derived/ and artifacts/?

  • artifacts/: Original source materials (audio, video, images)
  • derived/: Processed data derived from artifacts (embeddings, memory atoms, entity graphs)

Derived data is typically generated by AI/ML pipelines and is less sensitive than raw artifacts.

How do cross-border transfers work?

Check the record's rights.cross_border_transfer_basis:

  • none: No cross-border transfer allowed
  • consent: Allowed with explicit consent
  • contract: Allowed under contractual terms
  • adequacy: Allowed to jurisdictions with adequate protection
  • scc: Allowed under Standard Contractual Clauses

Always verify compliance with GDPR Article 44-50 or equivalent local laws.

Can I host my own DLRS Hub instance?

Yes! DLRS Hub is designed to be self-hosted:

  1. Clone the repository
  2. Set up object storage (S3, MinIO, etc.)
  3. Configure access controls
  4. Deploy validation tools
  5. Set up CI/CD for automated checks

See Architecture for deployment guidance.

Still Have Questions?

If your question isn't answered here:

  1. Check the complete documentation
  2. Search GitHub Issues
  3. Open a new issue with the question label
  4. Email us at https://github.com/Digital-Life-Repository-Standard/DLRS/discussions

We're here to help! 💙