Stencil · an open journal
Stencil is an open journal — a place for paradigm notes, design proposals, and contrarian takes on the architectures, numerical methods, and operating models I think are worth exploring or disseminating across scientific computing, AI, and the chip-design industry.
It is a personal platform, not a publication of record. Some essays speculate; some respond directly to what I'm building. All are anchored in physics simulation. Source files, diagrams, and data are open by default and tracked in the site's public repository.
In numerical methods, a stencil is the small, deliberate pattern of grid points a solver uses to approximate a derivative — local, geometric, with the whole continuum hanging from a handful of decisions. Every essay here is an attempt at one.
Manifesto
Stencil is free. It has no paywall, no editorial board, no submission queue, no impact-factor anxiety, and no advertisement. It will stay that way.
The institutions that mediate scientific discourse have done good work in their day. They have also drifted somewhere that no longer serves the writer or the reader. Commercial publishers extract billions in subscription fees from publicly funded science. "Open access" routes charge authors thousands of dollars per paper for the privilege of being read. Editorial boards decide which ideas are worth circulating; the safe, the consensus-friendly, and the easy-to-cite win, and contrarian work is filtered out before it reaches the people who would build on it. Labs use journal venues as marketing for themselves and their cohorts. The middleman has become the bottleneck.
I don't think this calls for overthrowing anything. I think it calls for walking around it.
Stencil is the smallest possible version of that walk-around. One person, one website, one essay at a time — source files open, diagrams open, data open, and a permanent DOI for citation. No editor approved it. No board of insiders endorsed it. It exists because I wrote it and put it on the internet under my own name. That is the entire infrastructure required.
In a free market of ideas, the right correction for a bad essay is a better essay. The right response to a flawed paradigm is an alternative paradigm — published openly, named openly, and tested in the open. No gatekeeper required.
So Stencil is also a blueprint. If you have an opinion you can defend, a paradigm you want to test, a contrarian read of your field — register a domain, write the essay, mint a DOI on Zenodo, and publish. Send me the link. Let's build a network of personal journals — a thousand Stencils — and let ideas compete on merit, not on which committee they passed.
The middleman has had a long run. Time for the rest of us to write.
Essays
A paradigm note. AI as a co-designer of the solver — embedded in the building blocks of an exascale simulation, learning iron laws of design from comparison against benchmarks and experiment, accumulating skill across the pillars of classical physics.