First Draft Feedback: 3 Editors Give Honest Writing Analysis - Part 1

In Part 3 of Art, Automation, and AI, Tim Grahl and Shawn Coyne ask what it means that language can be turned into math. Shawn compares a developmental editor to a CPA who reads only the evidence and always tells the truth, and explains the difference between an LLM’s probability inferences and a deterministic calculation. Tim lays out why writing is one of the most wicked learning environments there is, and why knowledge never turns into skill without fast, accurate feedback. Shawn then describes the software he built: a calculator, not an AI, that grades a writer’s sentences against 35 years of editorial judgment and returns feedback within 24 hours.

Can you do math on a sentence? Picking up from last week, Shawn Coyne argues that an LLM produces mimicked artifacts, not original ones, and that confusing the two is where the "ghost in the machine" fear comes from. Then he goes somewhere most writers won't like: writing can be evaluated objectively, all the way down to the sentence. Drawing on Gödel, Church, Turing, Deutsch, and Shannon, he explains why a finished piece of writing is a fully computable artifact, walks Tim Grahl through the exact process he uses to give feedback on a scene, and lays out what he's actually been working toward for 35 years: formalizing what he knows well enough that a machine can analyze a writer's work the way he would, and get them feedback in seconds instead of weeks.

Is AI the devil, the savior, or something nobody has defined yet? Shawn Coyne spent three years writing Mentoring the Machines with cognitive scientist John Vervaeke to answer that question, and in this episode he walks Tim Grahl through what he found. He lays out a framework for understanding every technology from the shovel to ChatGPT, explains why the real distinction isn't fast versus slow but computational versus conscious, and makes the case that what large language models fundamentally lack is myth. Along the way, he explains what the word myth actually means, why it is not a synonym for lie, and why both the "AI will destroy us" and "AI will save us" stories should make you suspicious.

Tim Grahl surveyed 506 unpublished writers and found they'd been working on their craft for an average of 14.9 years. In any other creative field that would be a sign something is broken. In this episode, Shawn Coyne explains why studying plot, structure, and character development doesn't get most writers there, and what does: learning to write a single sentence that conforms to the mythic pattern of everyday experience. He and Tim get into why Story Grid starts students at the sentence level, why feedback matters more than theory, why a flow state can fool you into thinking bad writing is good, and the difference between journaling and actually writing for a reader.

This will be the final episode of the Story Grid Podcast for the foreseeable future. Tim gives some background on why this decision was made along with an update on his current writing project and other happenings around Story Grid.