{
  "handle": "ComputationalChannel",
  "ref": "ComputationalChannel#4dbe",
  "stub": "4dbe",
  "semaId": "sema:ComputationalChannel#mh:SHA-256:4dbe50ab8b0edb70fa56a2db1c32fd83298f6813dc8972290cadd678e6c69336",
  "layer": "Infrastructure",
  "category": "Primitives",
  "gloss": "A channel formed by a generator language and an acceptor language, where rejection is computation, not noise",
  "mechanism": "Model a channel as two formal languages: a generator language G, the inputs a source tends to produce, and an acceptor language A, the inputs a receiver or validator accepts. Inputs in the intersection of G and A pass; inputs in G but not A are rejected. Unlike a Shannon channel, where lost information is noise, rejection here is meaningful and part of the channel's design, as when a password validator rejects memorable but weak passwords, or an environment rejects unviable genomes. The rejected set G minus A holds proximate options: already generated, carrying the generator's information, and made valid by a small change to A or G. If G is a subset of A, the channel has no friction and stores no options.",
  "invariants": [
    "Rejection is decided by the acceptor language, not by random corruption",
    "If the generator language is a subset of the acceptor language, nothing is rejected and no options are stored",
    "Either language can be treated as the reference for a valid input"
  ],
  "failureModes": [
    "Treating meaningful rejection as noise and engineering it away",
    "Treating the acceptor as the only reference language and ignoring what the generator encodes",
    "Assuming all rejected inputs are equally far from validity"
  ],
  "ring": 2,
  "tier": 2,
  "aliases": {},
  "buildsOn": [],
  "usedBy": [
    {
      "handle": "ContentAddressedMinting",
      "type": "references",
      "alias": "computational_channel",
      "url": "/knowledge-graph/ContentAddressedMinting/",
      "json": "/knowledge-graph/ContentAddressedMinting.json"
    },
    {
      "handle": "FractionalMetaModel",
      "type": "references",
      "alias": "computational_channel",
      "url": "/knowledge-graph/FractionalMetaModel/",
      "json": "/knowledge-graph/FractionalMetaModel.json"
    },
    {
      "handle": "ComputationalEntropyRate",
      "type": "accepts",
      "alias": "computational_channel",
      "url": "/knowledge-graph/ComputationalEntropyRate/",
      "json": "/knowledge-graph/ComputationalEntropyRate.json"
    },
    {
      "handle": "CrackpotProbe",
      "type": "references",
      "alias": "computational_channel",
      "url": "/knowledge-graph/CrackpotProbe/",
      "json": "/knowledge-graph/CrackpotProbe.json"
    }
  ],
  "sources": {
    "research": [],
    "works": [
      {
        "kind": "book",
        "label": "On the Origin of Risk",
        "note": "(Timlett 2025), ch. 16-17 (computational channels, password generator and acceptor)",
        "url": "/research/on_the_origin_of_risk/"
      }
    ],
    "homePage": false
  },
  "url": "/knowledge-graph/ComputationalChannel/",
  "graphUrl": "/knowledge-graph#ComputationalChannel",
  "directory": "/knowledge-graph/patterns.json"
}