Changes for page Requirements

Last modified by Robert Schaub on 2026/02/08 21:23

From 2.6 to 2.5

From version 2.5

edited by Robert Schaub
on 2026/01/20 20:24

Change comment: Renamed back-links.

To version 1.1

edited by Robert Schaub
on 2025/12/16 21:42

Change comment: Imported from XAR

Raw
Rendered

Summary

Page properties (2 modified, 0 added, 0 removed)

Details

Page properties

Parent

@@ -1,1 +1,1 @@
--Archive.FactHarbor 0\.9\.40.Specification.WebHome
++FactHarbor.Specification.WebHome

Content

@@ -9,7 +9,6 @@
  **Who**: Anyone (no login required).
  **Can**:
--
  * Browse and search claims
  * View scenarios, evidence, verdicts, and timelines
  * Compare scenarios and explore assumptions
@@ -19,7 +19,6 @@
  * **Submit claims automatically** by providing text to analyze - new claims are added automatically unless equal claims already exist in the system
  **Cannot**:
--
  * Modify existing content
  * Access draft content
  * Participate in governance decisions
@@ -31,7 +31,6 @@
  **Who**: Registered and logged-in users (extends Reader capabilities).
  **Can**:
--
  * Everything a Reader can do
  * Submit claims
  * Submit evidence
@@ -41,7 +41,6 @@
  * Request human review of AI-generated content
  **Cannot**:
--
  * Publish or mark content as "reviewed" or "approved"
  * Override expert or maintainer decisions
  * Directly modify AKEL or quality gate configurations
@@ -51,7 +51,6 @@
  **Who**: Trusted community members, appointed by maintainers.
  **Can**:
--
  * Review contributions from Contributors and AKEL drafts
  * Validate AI-generated content (Mode 2 → Mode 3 transition)
  * Edit claims, scenarios, and evidence
@@ -62,7 +62,6 @@
  * Participate in audit sampling
  **Cannot**:
--
  * Approve Tier A content for "Human-Reviewed" status (requires Expert)
  * Change governance rules
  * Unilaterally change expert conclusions without process
@@ -69,7 +69,6 @@
  * Bypass quality gates
  **Note on AI-Drafted Content**:
--
  * Reviewers can validate AI-generated content (Mode 2) to promote it to "Human-Reviewed" (Mode 3)
  * For Tier B and C, Reviewers have approval authority
  * For Tier A, only Experts can grant "Human-Reviewed" status
@@ -79,7 +79,6 @@
  **Who**: Subject-matter specialists in specific domains (medicine, law, science, etc.).
  **Can**:
--
  * Everything a Reviewer can do
  * **Final authority** on Tier A content "Human-Reviewed" status
  * Validate complex or controversial claims in their domain
@@ -89,13 +89,11 @@
  * Override AKEL suggestions in their domain (with documentation)
  **Cannot**:
--
  * Change platform governance policies
  * Approve content outside their expertise domain
  * Bypass technical quality gates (but can flag for adjustment)
  **Specialization**:
--
  * Experts are domain-specific (e.g., "Medical Expert", "Legal Expert", "Climate Science Expert")
  * Cross-domain claims may require multiple expert reviews
@@ -104,7 +104,6 @@
  **Who**: Reviewers or Experts assigned to sampling audit duties.
  **Can**:
--
  * Review sampled AI-generated content against quality standards
  * Validate quality gate enforcement
  * Identify patterns in AI errors or hallucinations
@@ -113,19 +113,16 @@
  * Contribute to audit statistics and transparency reports
  **Cannot**:
--
  * Change audit sampling algorithms (maintainer responsibility)
  * Bypass normal review workflows
  * Audit content they personally created
  **Selection**:
--
  * Auditors selected based on domain expertise and review quality
  * Rotation to prevent audit fatigue
  * Stratified assignment (Tier A auditors need higher expertise)
  **Audit Focus**:
--
  * Tier A: Recommendation 30-50% sampling rate, expert auditors
  * Tier B: Recommendation 10-20% sampling rate, reviewer/expert auditors
  * Tier C: Recommendation 5-10% sampling rate, reviewer auditors
@@ -135,7 +135,6 @@
  **Who**: Maintainers or trusted long-term contributors.
  **Can**:
--
  * All Reviewer and Expert capabilities (cross-domain)
  * Manage user accounts and permissions
  * Handle disputes and conflicts
@@ -146,7 +146,6 @@
  * Oversee audit system performance
  **Cannot**:
--
  * Change core data model or architecture
  * Override technical system constraints
  * Make unilateral governance decisions without consensus
@@ -156,7 +156,6 @@
  **Who**: Core team members responsible for the platform.
  **Can**:
--
  * All Moderator capabilities
  * Change data model, architecture, and technical systems
  * Configure quality gates and AKEL parameters
@@ -168,15 +168,14 @@
  * Grant and revoke roles
  **Governance**:
--
  * Maintainers operate under organizational governance rules
  * Major policy changes require Governing Team approval
  * Technical decisions made collaboratively
++
  == 2. Content Publication States ==
  === 2.1 Mode 1: Draft ===
--
  * Not visible to public
  * Visible to contributor and reviewers
  * Can be edited by contributor or reviewers
@@ -183,20 +183,18 @@
  * Default state for failed quality gates
  === 2.2 Mode 2: AI-Generated (Published) ===
--
  * **Public** and visible to all users
  * Clearly labeled as "AI-Generated, Awaiting Human Review"
  * Passed all automated quality gates
  * Risk tier displayed (A/B/C)
  * Users can:
--** Read and use content
--** Request human review
--** Flag for expert attention
++  ** Read and use content
++  ** Request human review
++  ** Flag for expert attention
  * Subject to sampling audits
  * Can be promoted to Mode 3 by reviewer/expert validation
  === 2.3 Mode 3: Human-Reviewed (Published) ===
--
  * **Public** and visible to all users
  * Labeled as "Human-Reviewed" with reviewer/expert attribution
  * Passed quality gates + human validation
@@ -205,16 +205,15 @@
  * For Tier B/C, Reviewer approval sufficient
  === 2.4 Rejected ===
--
  * Not visible to public
  * Visible to contributor with rejection reason
  * Can be resubmitted after addressing issues
  * Rejection logged for transparency
++
  == 3. Contribution Rules ==
  === 3.1 All Contributors Must: ===
--
  * Provide sources for claims
  * Use clear, neutral language
  * Avoid personal attacks or insults
@@ -222,7 +222,6 @@
  * Accept community feedback gracefully
  === 3.2 AKEL (AI) Must: ===
--
  * Mark all outputs with `AuthorType = AI`
  * Pass quality gates before Mode 2 publication
  * Perform mandatory contradiction search
@@ -232,7 +232,6 @@
  * Submit to audit sampling
  === 3.3 Reviewers Must: ===
--
  * Be impartial and evidence-based
  * Document reasoning for decisions
  * Escalate to experts when appropriate
@@ -240,7 +240,6 @@
  * Provide constructive feedback
  === 3.4 Experts Must: ===
--
  * Stay within domain expertise
  * Disclose conflicts of interest
  * Document specialized terminology
@@ -247,10 +247,10 @@
  * Provide reasoning for domain-specific decisions
  * Participate in Tier A audits
++
  == 4. Quality Standards ==
  === 4.1 Source Requirements ===
--
  * Primary sources preferred over secondary
  * Publication date and author must be identifiable
  * Sources must be accessible (not paywalled when possible)
@@ -258,7 +258,6 @@
  * Echo chamber sources must be flagged
  === 4.2 Claim Requirements ===
--
  * Falsifiable or evaluable
  * Clear definitions of key terms
  * Boundaries and scope stated
@@ -266,7 +266,6 @@
  * Uncertainty acknowledged
  === 4.3 Evidence Requirements ===
--
  * Relevant to the claim and scenario
  * Reliability assessment provided
  * Methodology described (for studies)
@@ -273,14 +273,14 @@
  * Limitations noted
  * Conflicting evidence acknowledged
--== 5. Risk tier Assignment ==
++== 5. Risk Tier Assignment ==
++
  **Automated (AKEL)**: Initial tier suggested based on domain, keywords, impact
  **Human Validation**: Moderators or Experts can override AKEL suggestions
  **Review**: Risk tiers periodically reviewed based on audit outcomes
  **Tier A Indicators**:
--
  * Medical diagnosis or treatment advice
  * Legal interpretation or advice
  * Election or voting information
@@ -289,7 +289,6 @@
  * Potential for significant harm
  **Tier B Indicators**:
--
  * Complex scientific causality
  * Contested policy domains
  * Historical interpretation with political implications
@@ -296,17 +296,21 @@
  * Significant economic impact claims
  **Tier C Indicators**:
--
  * Established historical facts
  * Simple definitions
  * Well-documented scientific consensus
  * Basic reference information
++
++
++
  == 6. User Role Hierarchy Diagram ==
  The following diagram visualizes the complete role hierarchy:
++
++
  == 7. Role Hierarchy Diagrams ==
  === 7.1 User Class Diagram ===
@@ -328,14 +328,18 @@
  {{include reference="FactHarbor.Specification.Diagrams.Technical and System Users.WebHome"/}}
  **Key Design Principles**:
--
  * **Two tracks from Contributor**: Content Track (Reviewer) and Technical Track (Maintainer)
  * **Technical Users**: System processes (AKEL, bots) managed by Maintainers
  * **Separation of concerns**: Editorial authority independent from technical authority
++
++
++
++
  = Functional Requirements =
++
  This page defines what the FactHarbor system must **do** to fulfill its mission.
  Requirements are structured as FR (Functional Requirement) items and organized by capability area.
@@ -346,7 +346,6 @@
  === 8.1 FR1 – Claim Intake ===
  The system must support Claim creation from:
--
  * Free-text input (from any Reader)
  * URLs (web pages, articles, posts)
  * Uploaded documents and transcripts
@@ -374,6 +374,7 @@
  * Allow manual correction of cluster membership
  * Provide explanation why two claims are considered "same cluster"
++
  == 9. Scenario System ==
  === 9.1 FR5 – Scenario Creation ===
@@ -385,7 +385,6 @@
  === 9.2 FR6 – Required Scenario Fields ===
  Each scenario includes:
--
  * Definitions (key terms)
  * Assumptions (explicit, testable where possible)
  * Boundaries (time, geography, population, conditions)
@@ -403,6 +403,7 @@
  * Users can compare scenarios side by side
  * Show differences in assumptions, definitions, and evidence sets
++
  == 10. Evidence Management ==
  === 10.1 FR9 – Evidence Ingestion ===
@@ -414,7 +414,6 @@
  === 10.2 FR10 – Evidence Assessment ===
  For each piece of evidence:
--
  * Assign reliability / quality ratings
  * Capture who rated it and why
  * Indicate known limitations, biases, or conflicts of interest
@@ -426,12 +426,12 @@
  * Make dependencies explicit (e.g., "Scenario A uses subset of evidence used in Scenario B")
  * Use ScenarioEvidenceLink table with RelevanceScore
++
  == 11. Verdicts & Truth Landscape ==
  === 11.1 FR12 – Scenario Verdicts ===
  For each Scenario:
--
  * Provide a **probability- or likelihood-based verdict**
  * Capture uncertainty and reasoning
  * Distinguish between AKEL draft and human-approved verdict
@@ -449,6 +449,7 @@
  * Allow users to see "as of date X, what did we know?"
  * Maintain complete version history for auditing
++
  == 12. Workflow, Moderation & Audit ==
  === 12.1 FR15 – Workflow States ===
@@ -466,17 +466,17 @@
  === 12.3 FR17 – Audit Trail ===
  * Every significant action (create, edit, publish, delete/hide) is logged with:
--** Who did it
--** When (timestamp)
--** What changed (diffs)
--** Why (justification text)
++  ** Who did it
++  ** When (timestamp)
++  ** What changed (diffs)
++  ** Why (justification text)
++
  == 13. Quality Gates & AI Review ==
  === 13.1 FR18 – Quality Gate Validation ===
  Before AI-generated content (Mode 2) publication, enforce:
--
  * Gate 1: Source Quality
  * Gate 2: Contradiction Search (MANDATORY)
  * Gate 3: Uncertainty Quantification
@@ -494,6 +494,7 @@
  * Moderators and Experts can override
  * Risk tier affects publication workflow
++
  == 14. Federation Requirements ==
  === 14.1 FR21 – Node Autonomy ===
@@ -505,7 +505,6 @@
  === 14.2 FR22 – Data Sharing Modes ===
  Nodes must be able to:
--
  * Share claims and summaries only
  * Share selected claims, scenarios, and verdicts
  * Share full underlying evidence metadata where allowed
@@ -532,6 +532,7 @@
  * NEVER share model weights
  * NEVER override local governance
++
  == 15. Non-Functional Requirements ==
  === 15.1 NFR1 – Transparency ===
@@ -566,7 +566,6 @@
  * Release 1.0: **2000+ concurrent users** on a reasonably provisioned node
  Technical targets for Release 1.0:
--
  * Scalable monolith or early microservice architecture
  * Sharded vector database (for semantic search)
  * Optional IPFS or other decentralized storage for large artifacts
@@ -597,6 +597,7 @@
  * Support for saved views, filters, and user-level preferences
  * Progressive disclosure: casual users see summaries, advanced users can dive deep
++
  == 16. Release Levels ==
  === 16.1 Proof of Concept (POC) ===
@@ -626,10 +626,15 @@
  * Multi-node federation with full sync protocol
  * Mature audit system
++
++
++
  == 17. Related Pages ==
--* [[AKEL (AI Knowledge Extraction Layer)>>Archive.FactHarbor.Specification.AI Knowledge Extraction Layer (AKEL).WebHome]]
++
++* [[AKEL (AI Knowledge Extraction Layer)>>FactHarbor.Specification.AI Knowledge Extraction Layer (AKEL).WebHome]]
  * [[Automation>>FactHarbor.Specification.Automation.WebHome]]
  * [[Workflows>>FactHarbor.Specification.Workflows.WebHome]]
--* [[Governance>>Archive.FactHarbor.Organisation.Governance.WebHome]]
++* [[Governance>>FactHarbor.Organisation.Governance]]
++

Changes for page Requirements

Summary

Details

Applications

Navigation

Need help?