A Brain That Publishes Itself

One source of truth in a database, with the website derived from it at build time — no CMS, no runtime database, no drift.

The usual arrangement has it backwards. Content lives in a CMS, which is also the thing that serves it, which means the published surface is the source — and the moment you want it somewhere else, you're exporting, copying, and hoping the copies agree. Two places that both claim to be true is the definition of drift.

Invert it and the problem disappears. The files are the source; the site is a projection of them, rebuilt from scratch every time. Nothing is authored in the site, so nothing can be lost when it's regenerated. The website becomes disposable in the best sense: delete it, rebuild it in seconds, and it's identical — because everything that mattered was upstream.

Once publishing is a build step rather than an act, it stops being work. The same source can project into a site, a PDF, a newsletter, a slide deck — one pass, many faces. That's the difference between a brain you publish from and a brain that publishes itself.