Context Edge expresses “in which worldview/context is this Claim true.”
It corresponds to the concept of possible worlds in Modal Logic, where different facts can exist about the same Subject depending on the worldview.
Context "Reality": (Earth, age, 4.6 billion years)
Context "Young Earth": (Earth, age, 6000 years)
Context "Harry Potter": (magic, exists, true)
Packet Structure (3 words, 48 bits)
1st WORD (16 bits):
┌─────────────────────┬─────────────────┐
│ Prefix │ Context Type │
│ 10 bits │ 6 bits │
└─────────────────────┴─────────────────┘
[1100 000 100] [TTTTTT]
2nd WORD: Context TID (16 bits)
3rd WORD: Target TID (16 bits)
| Field | Bits | Description |
|---|
| Prefix | 10 | 1100 000 100 |
| Context Type | 6 | 0=unspecified, 1~62=type, 63=extended (reserved) |
| Context TID | 16 | Unique identifier for this Context |
| Target TID | 16 | Target Claim (Triple/Verb/Event6/Clause TID) |
Context Type (6 bits = 64)
Source – Code 1~20
| Code | Type | Description | Example |
|---|
| 1 | SYSTEM | System-generated | Wikidata sync |
| 2 | USER | User-entered | Manual creation |
| 3 | DOCUMENT | General document | PDF, Word |
| 4 | NEWS | News article | Reuters, AP |
| 5 | ACADEMIC | Academic paper | arXiv, Nature |
| 6 | GOVERNMENT | Government/public agency | SEC, census bureau |
| 7 | WIKI | Wikipedia/Wikidata | Q42, P31 |
| 8 | API | External API | Finance, weather |
| 9 | ORG | Organization announcement | Corporate IR |
| 10 | BOOK | Book | ISBN-based |
| 11 | INTERVIEW | Interview/testimony | Direct quote |
| 12 | DATASET | Dataset | Kaggle |
| 13 | SOCIAL | Social media | Twitter |
| 14 | LEGAL | Law/case law | Court ruling |
| 15 | ARCHIVE | Archive | archive.org |
| 16 | MULTIMEDIA | Video/audio | YouTube |
| 17 | DATABASE | Database | IMDB, Freebase |
| 18 | ENCYCLOPEDIA | Encyclopedia | Britannica |
| 19 | MANUAL | Manual/guide | Technical docs |
| 20 | STANDARD | Standard document | ISO, RFC |
Derived/Inference – Code 21~30
| Code | Type | Description | Example |
|---|
| 21 | MODEL | AI model generation | GPT, Claude |
| 22 | INFERENCE | Logical inference | Rule-based |
| 23 | AGGREGATION | Aggregation/integration | Multi-source synthesis |
| 24 | CALCULATION | Calculation result | Formula application |
| 25 | TRANSLATION | Translation | Source → translation |
| 26 | EXTRACTION | Extraction | NER, RE |
| 27 | CORRECTION | Correction/amendment | Error correction |
| 28 | HEARSAY | Hearsay/rumor | Unverified |
| 29 | ESTIMATION | Estimation | Approximate value |
| 30 | PREDICTION | Prediction | Future forecast |
Worldview/Belief – Code 31~45
| Code | Type | Description | Example |
|---|
| 31 | RELIGION | Religious worldview | Protestantism, Buddhism |
| 32 | PHILOSOPHY | Philosophical perspective | Existentialism |
| 33 | SCIENCE | Scientific consensus | Modern physics |
| 34 | POLITICS | Political perspective | Conservative, progressive |
| 35 | CULTURE | Cultural perspective | Eastern, Western |
| 36 | MYTHOLOGY | Mythological system | Greek mythology |
| 37 | FOLKLORE | Folklore/tradition | Local tales |
| 38 | IDEOLOGY | Ideological system | Capitalism |
| 39 | THEORY | Theory | Theory of relativity |
| 40 | HYPOTHESIS | Hypothesis | Pre-verification |
| 41 | TRADITION | Tradition/custom | Confucian tradition |
| 42 | CONSENSUS | Consensus/accepted view | Academic consensus |
| 43 | MAINSTREAM | Mainstream view | Majority opinion |
| 44 | ALTERNATIVE | Alternative view | Minority opinion |
| 45 | FRINGE | Fringe/heterodox | Pseudoscience |
Fiction/Creative – Code 46~55
| Code | Type | Description | Example |
|---|
| 46 | NOVEL | Novel universe | Lord of the Rings |
| 47 | FILM | Film universe | MCU |
| 48 | GAME | Game universe | Zelda |
| 49 | COMICS | Comics universe | DC Universe |
| 50 | ANIMATION | Animation universe | Studio Ghibli |
| 51 | DRAMA | Drama universe | Game of Thrones |
| 52 | THEATER | Theater universe | Hamlet |
| 53 | FANFIC | Fan creation | Fan fiction |
| 54 | LEGEND | Legend | King Arthur |
| 55 | FAIRYTALE | Fairy tale | Cinderella |
Perspective/Speaker – Code 56~62
| Code | Type | Description | Example |
|---|
| 56 | NARRATOR | Narrator perspective | Omniscient narrator |
| 57 | PROTAGONIST | Protagonist perspective | Hero’s viewpoint |
| 58 | ANTAGONIST | Antagonist perspective | Villain’s viewpoint |
| 59 | AUTHOR | Author intent | Author’s commentary |
| 60 | EXPERT | Expert opinion | Scholar’s opinion |
| 61 | LAYMAN | Layperson perception | Public perception |
| 62 | SATIRICAL | Satire/irony | Ironic expression |
Code 0 is UNSPECIFIED, and Code 63 is EXTENDED (reserved).
Additional information about the Context itself (source, confidence, worldview name) is expressed via Triple Edge.
(Context TID, P:source_entity, Reuters_Entity) - Source organization
(Context TID, P:confidence, 0.95) - Confidence
(Context TID, P:universe_name, "Harry Potter") - Universe name
(Context TID, P:perspective_holder, Villain_Entity) - Perspective holder
Examples
Source: “Reuters report”
Context Edge:
1st: [1100 000 100] + [000100] - NEWS (4)
2nd: [0x0300] - Context TID
3rd: [0x0001] - Target: Triple "Apple acquired Tesla"
Additional Triples:
(0x0300, P:source_entity, Reuters)
(0x0300, P:date, 2026-01-29)
Fiction: “Harry Potter universe”
Context Edge:
1st: [1100 000 100] + [101110] - NOVEL (46)
2nd: [0x0302] - Context TID
3rd: [0x0003] - Target: Triple "Hogwarts is_a school"
Additional Triples:
(0x0302, P:universe_name, "Harry Potter")
(0x0302, P:author, J.K. Rowling)
AI Inference: “Claude’s inference”
Context Edge:
1st: [1100 000 100] + [010101] - MODEL (21)
2nd: [0x0304] - Context TID
3rd: [0x0005] - Target: Triple "X causes Y"
Additional Triples:
(0x0304, P:model, Claude_Entity)
(0x0304, P:confidence, 0.75)
Design Rationale
- Context Edge as a separate type: Worldviews are a meta-layer distinct from Triple/Clause. Corresponds to the G (Graph) in RDF Quads.
- 6-bit Context Type: Enables immediate classification without separate Triples. 62 types cover most use cases.
- 3-word lightweight structure: Context links occur in large volumes, so minimal size ensures storage efficiency.