Questions in My Dissertation
I asked Claude Code to go through my dissertation and pull out all the questions. There are 109 of them?
The extraction script parses the HTML (long ago created from the markdown used to draft my dissertation) of each chapter, finds lines with question marks (filtering out URLs and footnotes), splits multi-sentence lines into individual questions, and classifies each one as author, interview, or cited. We went back and forth on the heuristics over about 14 rounds — tweaking participant name detection, handling epigraphs, dropping orphaned fragments, fixing sentence-boundary splitting that was grabbing preamble text before the actual question, deciding what counts as a "real question."
My dissertation about search on the web is built around questions. My own questions driving the research. Questions from interview participants. Questions from cited works. And cascades of future research questions in the conclusion. My work now at Hypandra is deeply engaged with questions — how we will continue to develop our ability to ask curious questions effectively amidst and about AI.
A few favorites:
How will you look for it, Socrates, when you do not know at all what it is? How will you aim to search for something you do not know at all? If you should meet with it, how will you know that this is the thing that you did not know?
That's Meno's paradox — the epigraph for the chapter on extending searching. It captures something essential about what I found: data engineers don't succeed at search because they know what they're looking for. They succeed because their work practices structure the searching.
Sometimes I just wonder, like, who taught them how to search?
Victor asked this about his colleagues. It was one of those moments in an interview where the participant articulates the research question better than you did.
The conclusion is where the questions really cascade. Each of the five takeaways opens into a set of provocations for future research — questions about legitimation, extensions, knowledge, responsibility, and technocratization of search. They're the questions I carried out of the dissertation and into the work I'm doing now.