Skip to content

calculationInterpretationAPI.md

Latest commit

 

History

History
100 lines (81 loc) · 1.79 KB

calculationInterpretationAPI.md

File metadata and controls

100 lines (81 loc) · 1.79 KB

the Calculation / Interpretation API

The Calculation Engine sends a request to the nostr Interpretation Engine, which returns a response.

Request

The request is an object with three properties: source, universalInterpretationProtocolID and parameters.

The request must validate against the nostr interpretation engine request json schema:

{
  required: [ "source", "universalInterpretationProtocolID", "parameters" ],
  properties: {
    source: {
      type: string,
      enum: [ " nostr" ]
    },
    universalInterpretationProtocolID: {
      type: string
    },
    parameters: {
      type: object
    },
  }
}

where:

  • universalInterpretationProtocolID must be recognized, i.e. must be in the Interpretation Engine's database of supported protocols
  • parameters is a stringified version of an object, the schema of which is protocol-specific.

An example request is the following:

{
  source: nostr
  universalInterpretationProtocolID: basicFollowsInterpretationProtocol
  parameters: ... (see below)
}

where parameters is the object:

{
  score: 1.0,
  confidence: 0.05,
  depth: 5,
  pubkeys: [...],
  context: notSpam
}

Response

The Interpretation Engine sends a response back to the Calculation Engine.

success

An example success response:

{
  success: true,
  ratingsTable: ...
}

The ratingsTable is a stringified (or should we leave this as an array?) array of individual ratings. Example:

[
  {
    rater: pk_Bob,
    ratee: pk_Alice,
    context: notSpam,
    score: 1.0,
    confidence: 0.05
  },
  {
    rater: pk_Bob,
    ratee: pk_Charlie,
    context: notSpam,
    score: 1.0,
    confidence: 0.05
  }
]

Error responses

{
  success: false,
  message: ...
}

(work in progress to specify the error messages)