Wiki source code of Implementation Roadmap
Last modified by Robert Schaub on 2026/02/08 08:13
Hide last authors
| author | version | line-number | content |
|---|---|---|---|
 |
1.1 | 1 | = Implementation Roadmap = |
| 2 | |||
| 3 | **FactHarbor development follows a phased approach from Proof of Concept to production launch.** | ||
| 4 | |||
| 5 | == Development Phases == | ||
| 6 | |||
| 7 | {{info}} | ||
| |
1.11 | 8 | **Detailed Requirement Mapping:** See [[Requirements Roadmap Matrix>>Archive.FactHarbor 2026\.01\.20.Roadmap.Requirements-Roadmap-Matrix.WebHome]] for complete phase-to-requirement mapping, implementation levels, and User Needs fulfillment by phase. |
| |
1.1 | 9 | {{/info}} |
| 10 | |||
| |
1.9 | 11 | * [[POC1: Core Workflow with Quality Gates>>Archive.FactHarbor 2026\.01\.20.Roadmap.POC1.WebHome]] |
| |
1.10 | 12 | * [[POC2: Robust Quality & Reliability>>Archive.FactHarbor 2026\.01\.20.Roadmap.POC2.WebHome]] |
| |
1.8 | 13 | * [[Beta 0: User Testing & Production Readiness>>Archive.FactHarbor 2026\.01\.20.Roadmap.Beta0.WebHome]] |
| |
1.12 | 14 | * [[V1.0: Public Launch>>Archive.FactHarbor 2026\.01\.20.Roadmap.V10.WebHome]] |
| |
1.1 | 15 | * [[Future Milestones (POC2, Beta0, V1.0)>>FactHarbor.Roadmap.Future Milestones]] |
| 16 | |||
| 17 | == Philosophy == | ||
| 18 | |||
| 19 | **Validate AI automation quality BEFORE building production features** | ||
| 20 | |||
| 21 | Each phase builds on proven capabilities from the previous phase. We never skip ahead - if AI quality is insufficient, we improve prompts and algorithms rather than adding manual workarounds. | ||
| 22 | |||
| 23 | == Development Phases Overview == | ||
| 24 | |||
| 25 | **POC1: Core Workflow with Quality Gates** | ||
| 26 | |||
| 27 | Build: | ||
| |
1.2 | 28 | |
| |
1.1 | 29 | * Automated claim analysis |
| 30 | * Confidence scoring | ||
| 31 | * Source evaluation | ||
| 32 | * Browse/search interface | ||
| 33 | * User flagging system | ||
| 34 | |||
| 35 | * **LLM Provider Abstraction (NFR-14):** Provider-agnostic AI access | ||
| |
1.2 | 36 | * POC1: Anthropic Claude API primary (Haiku Stage 1, Sonnet Stages 2 & 3) |
| 37 | * Architecture: Abstract LLMProvider interface, environment config, basic failover | ||
| 38 | * Future: OpenAI/Google secondary providers, admin API, cost comparison (POC2/Beta) | ||
| |
1.1 | 39 | |
| 40 | **Goal**: Prove AI quality before adding user editing | ||
| 41 | |||
| 42 | **User Needs fulfilled in POC1**: UN-1, UN-2, UN-3, UN-4, UN-5, UN-6, UN-7, UN-8, UN-9, UN-12 | ||
| 43 | |||
| 44 | **POC2/Beta 0: User Contributions** | ||
| 45 | |||
| 46 | Add only if needed: | ||
| |
1.2 | 47 | |
| |
1.1 | 48 | * Simple editing (Wikipedia-style) |
| 49 | * Reputation system | ||
| 50 | * Basic moderation | ||
| 51 | * In-article claim highlighting (FR13) | ||
| 52 | |||
| 53 | **Additional User Needs fulfilled**: UN-13, UN-17 | ||
| 54 | |||
| 55 | **V1.0+: Continuous Improvement** | ||
| 56 | |||
| 57 | * Continuous quality improvement | ||
| 58 | * Feature additions based on real usage | ||
| 59 | * Scale infrastructure | ||
| 60 | |||
| 61 | **Additional User Needs fulfilled**: UN-14 (API access) | ||
| 62 | |||
| 63 | **Deferred Beyond V1.0**: | ||
| |
1.2 | 64 | |
| |
1.1 | 65 | * Federation (until multiple successful instances exist) |
| 66 | * Complex contribution workflows (focus on automation) | ||
| 67 | * Extensive role hierarchy (keep simple) | ||
| 68 | |||
| 69 | == Success Metrics Framework == | ||
| 70 | |||
| 71 | **Note**: Each phase defines specific success criteria in detail. This section provides the overall metrics framework. | ||
| 72 | |||
| 73 | **System Quality**: | ||
| |
1.2 | 74 | |
| |
1.1 | 75 | * Error rate by category (target: continuous improvement) |
| 76 | * Average confidence score (target: increase) | ||
| 77 | * Source quality distribution (target: more high-quality) | ||
| 78 | * Contradiction detection rate (target: increase) | ||
| 79 | |||
| 80 | **Efficiency**: | ||
| |
1.2 | 81 | |
| |
1.1 | 82 | * Claims processed per hour (target: increase) |
| 83 | * Human hours per claim (target: decrease) | ||
| 84 | * Automation coverage (target: >90%) | ||
| 85 | * Re-work rate (target: <5%) | ||
| 86 | |||
| 87 | **User Satisfaction**: | ||
| |
1.2 | 88 | |
| |
1.1 | 89 | * User flag rate (issues found) |
| 90 | * Correction acceptance rate (flags valid) | ||
| 91 | * Return user rate | ||
| 92 | * Trust indicators (surveys) | ||
| 93 | |||
| 94 | **User Needs Metrics**: | ||
| |
1.2 | 95 | |
| |
1.1 | 96 | * UN-1: % users who understand trust scores |
| 97 | * UN-4: Time to verify social media claim | ||
| 98 | * UN-7: % users who access evidence details | ||
| 99 | * UN-8: % users who view multiple scenarios | ||
| 100 | * UN-15: % users who check evolution timeline | ||
| 101 | * UN-17: % users who enable in-article highlighting; avg. time spent on highlighted vs. non-highlighted articles | ||
| 102 | |||
| 103 | == Status == | ||
| 104 | |||
| 105 | **Current Phase:** POC1 Specification Complete | ||
| 106 | **Next Action:** Begin POC1 Implementation | ||
| 107 | **Version:** V0.9.70 |