Blog

Notes on the understanding problem: sharing agent-built work, review culture, and making comprehension as cheap as generation.

• October 22, 2026Agreement is the least useful thing a review producesA plan everyone approves can still be several plans. Research on hidden profiles and explanation says to ask each person for their assumption, privately, before anyone talks. Inputs is now open to everyone on thing.
• October 8, 2026thing Inputs: A step toward shared understandingOur vision for thing is to help teams build shared understanding. Inputs is an early step toward making assumptions, questions, and differences visible.
• August 29, 2026The artifact is the right idea. It should outlive the chat that made it.Claude Artifacts got the unit right: a page you can hand someone, not a transcript. The weakness is not quality, it is that the page belongs to the session that produced it. Here is what portable would mean, and what it costs.
• August 29, 2026Vibe coding works until someone else looks at itThe definition of vibe coding was always honest about the tradeoff: you stop reading the code. What nobody says out loud is that the bill arrives as a question from another person, not as a bug.
• August 24, 2026The real work is collective understandingPeter Naur argued in 1985 that a program is a theory held by the people who built it. Forty years on, agents write the code and the theory quietly stops forming, not in one head but between them.
• August 24, 2026Agents got faster at writing code. Nobody got faster at reading it.Review is where 2026 backs up. What teams are reporting, why more agent output can make a team slower, and the case that comprehension is now the constraint.