Workbench Tags

Browse research experiments by subject.

All 3D printing 1 AI-assisted coding 1 agentic AI 3 archives 1 big data 1 citations 2 hallucinations 1 maps 1 material culture 1 model tiers 2 paleography 1 peer review 1 prompting 4 writing 1

Showing all research experiments with tags.

Date Aug 2026 Status tested Type writing feedback Time one prompt per role; an afternoon to work through the responses Level researcher Topics writingpeer reviewprompting
Asking AI to Criticize a Draft

Asking AI to Criticize a Draft

ExperimentFour prompts asked for objections to the same draft, each emphasizing a different part of its intended audience.

Results
  • four sets of comments on the same draft, using different role descriptions
  • insights from every perspective that I had not considered
  • substantive revisions to the article
Date Sep 2026 Status tested Type LLM orientation Time less than 10 minutes Level any Topics promptingmodel tiershallucinations
Comparing Two Models on a Question About Huizinga

Comparing Two Models on a Question About Huizinga

ExperimentThe same question about a passage in Huizinga produced different answers from Gemini Flash Lite and Claude Opus 5.

Results
  • Gemini returned a passage that did not address court ceremony
  • Claude found relevant material without inventing page numbers
Date Apr 2026 Status tested Type data work Time 30–60 min Level any Topics 3D printingmaterial culture
Generate 3D Prints from 2D Drawings

Generate 3D Prints from 2D Drawings

ExperimentWe used historical drawings of IUDs to try making 3D-printable replicas with Meshy.

Results
  • generated printable models from historical drawings
  • found that Meshy could smooth away asymmetries in a source
Date Sep 2026 Status tested Type LLM orientation Time less than 10 minutes Level any Topics promptingmodel tierscitations
Initiate Research with an Annotated Bibliography

Initiate Research with an Annotated Bibliography

ExperimentA bibliography request about McCarthyism produced sources and an objection to the way the question was framed.

Results
  • an initial bibliography and a longer research report
  • a challenge to the assumption that pre-McCarthy norms were restored
Date Apr 2026 Status tested Type data work Time less than 1 hour Level any Topics AI-assisted codingmapsagentic AI
Photos to Map Pins

Photos to Map Pins

ExperimentA folder of street-cat photographs became an interactive map using the location data recorded by a phone.

Results
  • extracted GPS coordinates from more than 200 photographs
  • published a map with a clickable photograph at each pin
Date Apr 2026 Status tested Type processing sources Time downloaded document images; two hours to set up; automated run of ~12hr/register Level researcher Topics archivesbig datapaleography
Pipelines for Medieval Handwriting Recognition

Pipelines for Medieval Handwriting Recognition

ExperimentGemini and Claude transcribe 300-page archival registers for full-text searching, with readings checked against the images.

Results
  • processed entire handwritten registers into searchable text
  • kept model outputs for comparison and error checking
Date Aug 2026 Status tested Type citation work Time less than 10 minutes Level any Topics citationspromptingagentic AI
Table of Contents to Zotero

Table of Contents to Zotero

ExperimentClaude turned a publisher’s table of contents into chapter entries that could be imported into Zotero.

Results
  • extracted data from the Table of Contents
  • built a file in BibTeX format
  • imported to Zotero