User Class Diagram

author	version	line-number	content
		1	User Class Diagram
		2
		3	This diagram shows the user role hierarchy in FactHarbor, reflecting the current role model with Reader as the base role.
		4
		5	{{mermaid}}
		6	classDiagram
		7	class Reader {
		8	+String SessionID
		9	+String LocalPreferences
		10	+browse() View all published content
		11	+search() Search claims and scenarios
		12	+compare() Compare scenarios
		13	+flag() Flag issues or errors
		14	+submitClaim() Submit text for automatic claim extraction
		15	}
		16
		17	class Contributor {
		18	+String UserID
		19	+String DisplayName
		20	+String Email
		21	+DateTime RegisteredAt
		22	+submitEvidence() Attach sources
		23	+proposeScenario() Draft scenarios
		24	+comment() Participate in discussions
		25	+requestReview() Request human review
		26	}
		27
		28	class Reviewer {
		29	+String ReviewerID
		30	+String[] Domains
		31	+DateTime AppointedAt
		32	+review() Review contributions and AI drafts
		33	+validate() Validate AI-generated content Mode 2→3
		34	+edit() Edit claims, scenarios, evidence
		35	+approve() Approve Tier B/C for Human-Reviewed
		36	+flagForExpert() Escalate to expert review
		37	+audit() Participate in sampling audits
		38	}
		39
		40	class Auditor {
		41	+String AuditorID
		42	+String[] AuditDomains
		43	+Float AuditAccuracy
		44	+reviewSample() Review sampled AI content
		45	+validateQualityGates() Check gate enforcement
		46	+identifyPatterns() Find AI error patterns
		47	+provideFeedback() Improve system quality
		48	}
		49
		50	class Expert {
		51	+String ExpertID
		52	+String ExpertiseArea
		53	+String[] Certifications
		54	+DateTime VerifiedAt
		55	+authoritativeApproval() Final authority Tier A
		56	+validateComplex() Complex domain validation
		57	+defineStandards() Set domain quality standards
		58	+overrideAKEL() Override AI suggestions with docs
		59	}
		60
		61	class Moderator {
		62	+String ModeratorID
		63	+String[] Responsibilities
		64	+handleAbuse() Manage abuse reports
		65	+manageUsers() User permissions
		66	+enforceGuidelines() Community guidelines
		67	+adjustRiskTiers() Review tier assignments
		68	+overseeAudits() Audit system oversight
		69	}
		70
		71	class Maintainer {
		72	+String MaintainerID
		73	+String[] SystemAccess
		74	+configureSystem() Technical configuration
		75	+manageRoles() Grant and revoke roles
		76	+configureAKEL() Quality gates and parameters
		77	+deployUpdates() System deployment
		78	+setPolicy() Risk tier policies
		79	}
		80
		81	class AKEL {
		82	+String InstanceID
		83	+Enum AuthorType "AI"
		84	+extractClaims() Claim extraction
		85	+classifyRisk() Risk tier assignment
		86	+generateScenarios() Draft scenarios
		87	+searchContradictions() Mandatory counter-evidence search
		88	+validateQualityGates() Run 4 quality gates
		89	+proposeVerdicts() Draft verdicts
		90	}
		91
		92	Reader <\|-- Contributor : extends
		93	Contributor <\|-- Reviewer : is-a
		94	Reviewer <\|-- Auditor : specialized-role
		95	Reviewer <\|-- Expert : is-a
		96	Expert <\|-- Moderator : is-a
		97	Moderator <\|-- Maintainer : is-a
		98
		99	AKEL ..> Contributor : creates-drafts-for
		100	AKEL ..> Reviewer : submits-to
		101	AKEL ..> Auditor : audited-by
		102
		103	{{/mermaid}}
		104
		105	== Role Hierarchy Explanation ==
		106
		107	Base: Reader
		108	* Anyone, no login required
		109	* Foundation for all other roles
		110	* Can browse, search, compare, flag, and automatically submit claims
		111
		112	Contributor (extends Reader)
		113	* Registered users
		114	* Can submit evidence, propose scenarios, participate in discussions
		115	* All Reader capabilities plus contribution rights
		116
		117	Reviewer (is-a Contributor)
		118	* Trusted community members
		119	* Can review and approve content
		120	* Can validate AI-generated content (Mode 2 → Mode 3 for Tier B/C)
		121	* Participates in audit sampling
		122
		123	Auditor (specialized Reviewer)
		124	* Dedicated audit role
		125	* Reviews sampled AI-generated content
		126	* Validates quality gate enforcement
		127	* Provides feedback for system improvement
		128
		129	Expert (is-a Reviewer)
		130	* Domain specialists
		131	* Final authority for Tier A content
		132	* Can define domain-specific standards
		133	* Required for Tier A "Human-Reviewed" status
		134
		135	Moderator (is-a Expert)
		136	* Cross-domain authority
		137	* Handles abuse and disputes
		138	* Manages users and permissions
		139	* Oversees audit system
		140
		141	Maintainer (is-a Moderator)
		142	* Core team members
		143	* Technical system access
		144	* Configure AKEL and quality gates
		145	* Set platform-wide policies
		146
		147	AKEL (AI System)
		148	* Separate from user hierarchy
		149	* Creates drafts for Contributor review
		150	* Subject to audit oversight
		151	* Never bypasses human authority
		152
		153	== Key Design Principles ==
		154
		155	Progressive Trust:
		156	* Each level inherits capabilities from below
		157	* Additional responsibilities require appointment
		158	* Trust earned through contribution quality
		159
		160	Human Authority:
		161	* AKEL assists but doesn't decide
		162	* Human reviewers validate AI outputs
		163	* Experts have final say on Tier A content
		164
		165	Audit Oversight:
		166	* Dedicated Auditor role for quality assurance
		167	* Sampling-based rather than reviewing everything
		168	* Feedback loop improves AI performance
		169
		170	Separation of Concerns:
		171	* Technical (Maintainer) vs Editorial (Expert/Reviewer)
		172	* Domain expertise (Expert) vs Cross-domain moderation (Moderator)
		173	* Content creation (Contributor) vs Validation (Reviewer)