Skip to Content

Wiki source code of Specification

Version 8.1 by Robert Schaub on 2025/12/12 08:32
Show last authors
1 {{warning title="Version 0.8x (Delta Update)"}}
2 This document describes the **Specification** of FactHarbor. It is a working draft.
3 {{/warning}}
4
5 = Specification =
6
7 This section defines the technical architecture, data models, and functional requirements of FactHarbor.
8
9 == Core Concepts ==
10
11 FactHarbor structures reasoning about claims into transparent, inspectable steps:
12
13 * **Claims and Claim Clusters** – people submit real-world statements; similar phrasings are grouped.
14 * **Scenarios** – each claim is evaluated under clearly defined contexts (assumptions, definitions, boundaries).
15 * **Evidence** – sources and data are collected and linked to specific scenarios.
16 * **Verdicts** – **There is no single verdict for a claim.** Instead, there is a distinct verdict for **each scenario** against which the claim is evaluated.
17 * **Truth Landscape** – the aggregation of these multiple scenario-dependent verdicts shows where (under which assumptions) a claim is plausible and where it is not.
18 * **Time Evolution** – all important entities (claims, scenarios, evidence links, verdicts) are versioned.
19
20 == Functional Lifecycle ==
21
22 The system follows a six-step lifecycle:
23
24 1. **Claim submission**: Automatic extraction and normalisation; Cluster detection.
25 2. **Scenario building**: Clarifying definitions and assumptions; AI proposals with human approval.
26 3. **Evidence handling**: AI-assisted retrieval; Human assessment of reliability; Explicit scenario linking.
27 4. **Verdict creation**: AI-generated draft verdicts **per scenario**; Human refinement; Reasoning explanations.
28 5. **Public presentation**: Concise summaries; Truth Landscape comparison; Deep dives.
29 6. **Time evolution**: Versioning of all entities; Re-evaluation triggers when evidence changes.
30
31 == Chapters ==
32
33 {{children sort="title"/}}