Roles & Publication Modes
Roles & Publication Modes
Version: 0.9.70
Last Updated: December 21, 2025
Status: CORRECTED - Automation Philosophy Consistent
This page defines user roles and publication modes in FactHarbor.
1. User Roles
1.1 Contributors
Who: Community members who suggest system improvements
Responsibilities:
- Participate in sampling audits (analyze patterns, not individual outputs)
- Suggest algorithm/prompt improvements based on findings
- Document systematic issues observed
- Contribute to system improvement discussions
Can:
- Edit published content (changes apply immediately, Wikipedia-style)
- Flag quality issues (for sampling audit)
- Earn reputation through contributions
- Participate in RFC (Request for Comments) processes
Cannot:
- Approve content before publication
- Override quality gates
- Act as gatekeepers
- Manually fix individual AI outputs
Note: Contributors improve THE SYSTEM, not individual outputs.
1.2 Trusted Contributors
Who: Contributors with proven track record and domain expertise
Responsibilities:
- Same as Contributors, plus:
- Review complex algorithm changes
- Provide domain expertise on contested claims
- Mentor new contributors
Can:
- Everything Contributors can do, plus:
- Participate in higher-level system design decisions
- Review RFC proposals
- Access more detailed system metrics
Cannot:
- Approve content before publication
- Override quality gates
- Act as gatekeepers
Important: "Trusted" refers to judgment quality, NOT approval authority.
1.3 Moderators
Who: Team members focused on community health and abuse prevention
Responsibilities:
- Handle abuse, spam, and harassment
- Enforce community guidelines
- Respond to user reports
- Manage bans and appeals
Can:
- Hide abusive content
- Ban users for policy violations
- Review appeals
- Escalate serious issues to Governing Team
Cannot:
- Approve content for publication
- Review content quality before publication
- Override quality gates for content
- Act as editorial gatekeepers
Critical Distinction:
```
Moderators handle: ABUSE (spam, harassment, violations)
Moderators DO NOT handle: CONTENT QUALITY (that's automated)
```
1.4 Domain Trusted Contributors (Optional, Task-Specific)
Who: Subject matter specialists invited for specific high-stakes disputes
Not a permanent role: Contacted externally when needed for contested claims in their domain
When used:
- Medical claims with life/safety implications
- Legal interpretations with significant impact
- Scientific claims with high controversy
- Technical claims requiring specialized knowledge
Process:
- Moderator identifies need for expert input
- Contact expert externally (don't require them to be users)
- Trusted Contributor provides written opinion with sources
- Opinion added to claim record
- Trusted Contributor acknowledged in claim
Important: This is CONSULTATION, not APPROVAL. Their opinion is added to the evidence, not used as a gate.
User Needs served: UN-16 (Expert validation status)
2. Publication Modes
Fulfills: UN-1 (Trust indicators), UN-16 (Review status transparency)
FactHarbor uses TWO publication modes (not three). Focus is on transparency and confidence scoring, not gatekeeping.
2.1 Mode 1: Draft-Only
Status: Not visible to public
When Used:
- Quality gates failed
- Confidence below threshold
- Structural integrity issues
- Insufficient evidence
What Happens:
- Content remains private
- System logs failure reasons
- Prompts/algorithms improved based on patterns
- Content may be re-processed after improvements
NOT "pending human approval" - it's blocked because it doesn't meet automated quality standards.
2.2 Mode 2: AI-Generated (Public)
Status: Published and visible to all users
When Used:
- Quality gates passed
- Confidence ≥ threshold
- Meets structural requirements
- Sufficient evidence found
Includes:
- Confidence score displayed (0-100%)
- Risk tier badge (A/B/C)
- Quality indicators
- Clear "AI-Generated" labeling
- Sampling audit status
Labels by Risk Tier:
- Tier A (High Risk): "⚠️ AI-Generated - High Impact Topic - Seek Professional Advice"
- Tier B (Medium Risk): "🤖 AI-Generated - May Contain Errors"
- Tier C (Low Risk): "🤖 AI-Generated"
User Contributions:
- User edits apply immediately (Wikipedia model)
- All changes logged and versioned
- May be selected for sampling audit
- Reputation earned for quality contributions
REMOVED: Mode 3
V0.9.50 Decision: No centralized approval workflow.
Rationale:
- Defeats automation purpose
- Creates bottleneck
- Inconsistent quality
- Not scalable
What Replaced It:
- Better quality gates
- Sampling audits for system improvement
- Transparent confidence scoring
- Risk-based warnings
3. Content States
3.1 Published
Status: Visible to all users
Includes:
- AI-generated analyses (default state after passing gates)
- User-contributed content
- Edited/improved content
Quality Indicators (displayed with content):
- Confidence Score: 0-100% (AI's confidence in analysis)
- Source Quality Score: 0-100% (based on source track record)
- Controversy Flag: If high dispute/edit activity
- Completeness Score: % of expected fields filled
- Last Updated: Date of most recent change
- Edit Count: Number of revisions
- Review Status: AI-generated / Enhanced by contributors
Automatic Warnings:
- Confidence < 60%: "Low confidence - use caution"
- Source quality < 40%: "Sources may be unreliable"
- High controversy: "Disputed - multiple interpretations exist"
- Medical/Legal/Safety domain: "Seek professional advice"
User Needs served: UN-1 (Trust score), UN-9 (Methodology transparency), UN-15 (Evolution timeline), UN-16 (Review status)
3.2 Hidden
Status: Not visible to regular users (only to moderators)
Reasons:
- Spam or advertising
- Personal attacks or harassment
- Illegal content
- Privacy violations
- Deliberate misinformation (verified)
- Abuse or harmful content
Process:
- Automated detection flags for moderator review
- Moderator confirms and hides
- Original author notified with reason
- Can appeal to board if disputes moderator decision
Note: Content is hidden, not deleted (for audit trail)
4. Role Evolution Path
Contributor Journey:
- Visitor – Explores platform, reads documentation, may raise questions
2. New Contributor – Submits first improvements (typo fixes, small clarifications, new issues)
3. Contributor – Contributes regularly and follows project conventions
4. Trusted Contributor – Has a track record of high-quality work and reliable judgment
5. Auditor – Participates in sampling audits (pattern analysis)
6. Moderator – Focuses on behavior, tone, and conflict moderation (not content quality)
7. Domain Expert (optional) – Offers domain expertise without changing governance authority
Key Principle: Low barrier to entry, transparent criteria for advancement, clear separation between content quality (automated) and behavior moderation (human).
5. Principles
- Low barrier to entry for new contributors
- Transparent criteria for gaining and losing responsibilities
- Clear separation between content quality (automated) and behavioral moderation (human)
- Documented processes for escalation and appeal
- No gatekeeping for content publication
- Immediate application of contributions (Wikipedia model)
6. Related Pages
- AKEL - AI system
- Workflows - Process workflows
- Architecture - System architecture
- Decision Processes - Governance
- Requirements - All requirements
V0.9.70 CHANGES:
REMOVED:
- All references to "Mode 3" publication
- "Contributors validate quality gates"
- "Trusted Contributors validate outputs"
- "Moderators finalize publication"
CLARIFIED:
- Contributors suggest system improvements (not approve outputs)
- Moderators handle abuse only (not content quality)
- Trusted Contributors provide consultation (not validation)
- Domain experts provide opinions (not approval)
- Only 2 publication modes (AI-Generated / Draft-Only)
ADDED:
- Clear role boundaries
- Explicit "cannot" lists for each role
- Publication mode details
- Content state specifications