If you heard me discussing Dahlia Doctor on a podcast, this page collects the articles, projects, and other resources mentioned in those conversations. The newest appearance will always be listed first.
An upcoming conversation about how I use artificial intelligence as a research, writing, and editorial tool in my work on dahlias.
I joined Dig on Dahlias to talk about how I use AI tools in research and writing, including two very different Dahlia Doctor projects: a focused investigation into the genetics of dahlia flower color and the much larger reconstruction of the writings of early American dahlia breeder J. J. Broomall.
The episode is forthcoming. In the meantime, the projects discussed in the interview are available below.
These projects show two different ways I have used AI-assisted research and editorial tools: one to investigate a tightly defined scientific question, and another to organize and reconstruct a large body of historical material.
This article began with a deceptively simple breeding question: can repeated selection make dahlia flower color reliably stable? I used it as a test case for a more rigorous way of working with AI, in which evidence was assembled, challenged, corrected, and kept within clearly defined limits rather than pushed toward a predetermined answer.
Read the articleThis companion piece tells the story behind the research article: how I worked with multiple AI systems in different roles, how errors were caught and corrected, and why the process ultimately mattered as much as the conclusion. If the first article is the finished research product, this is the account of how it was made.
Read the companion articleJ. J. Broomall was an important early American dahlia breeder and writer whose work had largely disappeared from view. Over an extended research project, I reconstructed his scattered writings into a 29-part online work combining surviving articles, essays, catalogs, and historical records with editorial introductions, notes, and extensive internal linking.
The Broomall project shows AI being used at a very different scale: locating and comparing sources, reconstructing publication history, resolving contradictions, organizing a large body of material, and maintaining consistency across dozens of interconnected pieces.
Explore the Broomall project