Concept: Machine Verifiable Citations

CSL Evolution
1

This is an early concept about adding capability in CSL for machine verifiable citations.

{
  "type": "webpage",
  "URL": "https://en.wikipedia.org/wiki/MySociety",
  "title": "mySociety",
  "accessed": {"date-parts": [[2026, 8, 12]]},
  "archive": "Internet Archive Wayback Machine",
  "archive_location": "https://web.archive.org/web/20260812123456/https://en.wikipedia.org/wiki/MySociety",
  "evidence": [
    {
      "type": "quote-match",
      "quote": "mySociety was founded by Tom Steinberg in September 2003",
      "last-verified": "2026-08-12",
      "verified-by": "DOD-Bot/1.0 (+https://www.designingopendemocracy.com/bot/)",
      "status": "MATCH",
      "context": {
        "text": "mySociety was founded by Tom Steinberg in September 2003 and started with TheyWorkForYou, a parliamentary monitoring website that makes it easy for people to keep tabs on their elected representatives.",
        "prefix": "mySociety is a UK-based not-for-profit social enterprise,",
        "suffix": "and started with TheyWorkForYou, a parliamentary monitoring website.",
        "sha256": "abc789..."
      }
    }
  ]
}

The concept is that we add two capabilities to the citation

  • archive location/pinning
    • Assist with using internet archive way back machine if the url is missing
  • citation evidence
    • Structured evidence/claim that if formatted correctly can be machine verified

For now, this concept is being tested in the website http://DesigningOpenDemocracy.com to check that any quotes or references in the website actually exists in the other website.

This concept can be extended further to stuff like pdfs… but for now just currently testing against url as that’s easy for a script to auto check.

Currently tracking the concept as it’s being developed in this page: