atelier

i'm Barish.

his is my research atelier. Here I think out loud about software design: about the seam between the systems we build and the people who have to live inside them, and about what would have to change for our tools to feel less like weather and more like places we can actually inhabit. The areas below are where I keep finding myself, in different costumes, asking the same handful of questions.

For everything else — who I am, a personal blog, and how to reach me — see my personal site at barish.me.

Research Areas
  • agent-compatible systems Software whose surfaces remain workable when the user on the other side is a program — the design problem of building tools that humans and agents can both inhabit, without one having to pretend to be the other.
  • concept design Software designed at the level of the ideas it asks its users to hold — the units of behavior a person learns once and reuses everywhere — rather than at the level of features, screens, or endpoints.
  • local-first software Software whose data lives on the user's devices first — the network helps but does not rule — and which keeps working when the cloud is slow, missing, hostile, or gone.
  • malleable software Software the user can reshape — not only configure, but bend, recombine, and re-author — long after the original developer has stopped paying attention. The opposite of the sealed app.
  • personal computing Computing whose unit of analysis is one person and the work she is actually trying to do — the older sense of the phrase, before it became a retail category, and what would have to change for it to mean that again.
  • social and institutional computing Software as a medium of governance — how the systems we build distribute power inside groups, organizations, and publics, and what designers owe to the institutions their software half-creates.
Projects on the bench1
  • social-and-institutional-computing · active since May 2026

    Answerable Infrastructure

    Software that treats some edits not as updates, but as acts an institution may one day have to justify.
Recently touched
  • essay · answerable infrastructure

    Explanation for Whom? Hospitable Interpretability for Machine Learning

    A feature-attribution plot can be faithful and still answer the wrong question. Interpretability needs hospitable records that mark which why-question each artifact answers and keep divergent readings attached to the same decision.
  • essay · answerable infrastructure

    The Spreadsheet Was the Constitution

    Software picks up constitutional weight long before the by-laws catch up. It is where the rules live, and the first place anyone looks when a decision has to be defended.