{
  "title": "Can a capture carry its links, with their positions?",
  "measured": "2026-08-06",
  "question": "Before building a second product for mapping pages: does the data even exist, how much of it is there, and where does the approach break?",
  "method": "Three pages of deliberately different kinds, loaded in headless Chromium 1208, every visible a[href] measured with getBoundingClientRect in document coordinates. One run each.",
  "pages": {
    "Enzyklopädie": {
      "seite": {
        "w": 765,
        "h": 24784,
        "titel": "PDF - Wikipedia"
      },
      "anzahl": 1528,
      "intern": 1297,
      "anker": 273,
      "bild": 10,
      "ohne_text": 0,
      "beispiele": [
        {
          "x": 64,
          "y": 14,
          "w": 140,
          "h": 38,
          "href": "https://en.wikipedia.org/wiki/Main_Page",
          "text": "",
          "intern": true,
          "anker": false,
          "bild": true
        },
        {
          "x": 443,
          "y": 11,
          "w": 44,
          "h": 44,
          "href": "https://en.wikipedia.org/wiki/Special:Search",
          "text": "Search",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 543,
          "y": 25,
          "w": 45,
          "h": 16,
          "href": "https://donate.wikimedia.org/?wmf_source=donate&wmf_medium=sidebar&wmf_campaign=en.wikipedia.org&uselang=en",
          "text": "Donate",
          "intern": false,
          "anker": false,
          "bild": false
        },
        {
          "x": 596,
          "y": 25,
          "w": 95,
          "h": 16,
          "href": "https://en.wikipedia.org/w/index.php?title=Special%3ACreateAccount&returnto=PDF&experiments%5B0%5D=we-1-8-account-creation-no-desktop-benefits%3Aunsampled",
          "text": "Create account",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 699,
          "y": 25,
          "w": 38,
          "h": 16,
          "href": "https://en.wikipedia.org/w/index.php?title=Special%3AUserLogin&returnto=PDF&experiments%5B0%5D=we-1-8-account-creation-no-desktop-benefits%3Aunsampled",
          "text": "Log in",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 24,
          "y": 134,
          "w": 39,
          "h": 32,
          "href": "https://en.wikipedia.org/wiki/PDF",
          "text": "Article",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 79,
          "y": 134,
          "w": 25,
          "h": 32,
          "href": "https://en.wikipedia.org/wiki/Talk:PDF",
          "text": "Talk",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 544,
          "y": 134,
          "w": 33,
          "h": 32,
          "href": "https://en.wikipedia.org/wiki/PDF",
          "text": "Read",
          "intern": true,
          "anker": false,
          "bild": false
        }
      ]
    },
    "Verlag": {
      "seite": {
        "w": 780,
        "h": 437,
        "titel": "403"
      },
      "anzahl": 0,
      "intern": 0,
      "anker": 0,
      "bild": 0,
      "ohne_text": 0,
      "beispiele": []
    },
    "Behörde": {
      "seite": {
        "w": 768,
        "h": 1126,
        "titel": "RIS Informationsangebote"
      },
      "anzahl": 28,
      "intern": 23,
      "anker": 1,
      "bild": 1,
      "ohne_text": 0,
      "beispiele": [
        {
          "x": 139,
          "y": 8,
          "w": 74,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/UI/Kontakt.aspx",
          "text": "Kontakt (Accesskey 4)",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 212,
          "y": 8,
          "w": 93,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/UI/Impressum.aspx",
          "text": "Impressum (Accesskey 5)",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 306,
          "y": 8,
          "w": 155,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/UI/Datenschutz.aspx",
          "text": "Datenschutzerklärung (Accesskey 6)",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 460,
          "y": 8,
          "w": 174,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/UI/Barrierefreiheit.aspx",
          "text": "Barrierefreiheitserklärung (Accesskey 7)",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 634,
          "y": 8,
          "w": 77,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/UI/SiteMap.aspx",
          "text": "Sitemap (Accesskey 8)",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 711,
          "y": 8,
          "w": 57,
          "h": 15,
          "href": "https://www.ris.bka.gv.at/defaultEn.aspx",
          "text": "English",
          "intern": true,
          "anker": false,
          "bild": false
        },
        {
          "x": 3,
          "y": 89,
          "w": 762,
          "h": 17,
          "href": "https://www.ris.bka.gv.at/default.aspx",
          "text": "",
          "intern": true,
          "anker": false,
          "bild": true
        },
        {
          "x": 0,
          "y": 103,
          "w": 63,
          "h": 54,
          "href": "https://www.ris.bka.gv.at/Bund/",
          "text": "Bund",
          "intern": true,
          "anker": false,
          "bild": false
        }
      ]
    }
  },
  "finding": "Wikipedia yields 1,528 links with positions, a government portal 28. PubMed yields nothing — the page did not render headless at all, 437 px tall instead of several thousand. The sources that are hard for a crawler are hard for a headless link map for the same reason, which is the argument for collecting inside the browser where the capture already happens.",
  "context": {
    "print_to_pdf_link_annotations": 1450,
    "capture_link_annotations": 0,
    "note": "Measured 5 August on the same Wikipedia article. Chrome's print export already writes complete link annotations; the capture writes none. That gap, not the mapping idea, is the concrete shortfall."
  },
  "limitations": [
    "Three pages, one run each, no averages.",
    "Headless rendering is the wrong instrument for the hard cases and was used here only to answer the machbarkeit question cheaply.",
    "Whether an agent would use such a map is unmeasured. Nobody has asked for it."
  ],
  "license": "CC BY 4.0",
  "plan": "https://github.com/Bubu89/full-page-pdf-snap/blob/main/PLAN-LINKKARTE.md"
}