Jacob Polay

Projects

Pipelines, catalogues and series, and how they connect.

Project: AI and CollaborationAI and CollaborationProject: EarlyModernNEREarlyModernNERProject: Footnotes on FilmFootnotes on FilmProject: Reading the Archive by Machine: An OCR Benchmark for Historians, 1612–1921Reading the Archive by MachineProject: The Digital Historian's ToolboxThe Digital Historian's ToolboxPublication: The AI-Enabled Boom in Document TranscriptionThe AI-Enabled Boom in Document TranscriptionPublication: The Archive is Not a Toy: The Hidden Problems of a "Vintage" AIThe Archive is Not a ToyPublication: Can You Read Two Million Pages? An LLM Can Help.Can You Read Two Million Pages? An LLMPublication: Circumventing the GarblerCircumventing the GarblerPublication: Not Just Farming: How an LLM Helped Me Study Saskatchewan’s Smallest SettlementsNot Just FarmingPublication: Stop Choosing Tools. Start Building Them.Stop Choosing Tools. Start Building Them.Talk: Fitting Machine to PageFitting Machine to PageTalk: Reading the Unreadable: Optical Character Recognition of Early Modern PrintReading the UnreadableTalk: Solving OCR: Using olmOCR to Follow Commodities across the British WorldSolving OCRPerson: Chloë FarrChloë FarrPerson: Jessica JackJessica JackPerson: Jim CliffordJim CliffordPerson: Keith PembertonKeith PembertonPerson: Lianne C. LeddyLianne C. LeddyPerson: Mark HumphriesMark HumphriesMethod: Critical AICritical AIMethod: Knowledge graphsKnowledge graphsMethod: Named-entity recognitionNamed-entity recognitionMethod: OCROCR

Projects, writing and talks People and methodsHover or tab to a node to see its connections; select it to open.

  1. The Digital Historian's Toolbox

    A curated catalogue of the tools and public datasets historians have actually used, including newer AI tools, with every entry linked to the studies that used it.

    in development · Creator and editor

  2. EarlyModernNER

    Named-entity recognition for early modern English, and the open pipeline around it for transcription and knowledge-graph retrieval, built for my MA thesis.

    complete · Designer and developer (MA thesis)

  3. Reading the Archive by Machine: An OCR Benchmark for Historians, 1612–1921

    A benchmark of how well current OCR reads historical documents from 1612 to 1921, and what its errors mean for historical research.

    complete · Co-author

  4. AI and Collaboration

    A collaborative series on Active History exploring how AI can support collaboration in historical research, and the technical and ethical questions it raises.

    ongoing · Co-initiator; lead author of two posts

  5. Footnotes on Film

    A recurring film-and-lecture series at Saskatoon's Broadway Theatre that pairs a film with a history graduate student's research.

    ongoing · Founder