Static builds and why they map onto S3
Understanding the build artefact before deploying it, so the deployment step teaches deployment rather than refactoring.
npm run build produces a dist/ directory: plain HTML, CSS and a little JS,
with no server needed to run it. That maps exactly onto what S3 serves and
CloudFront caches — a bucket of files behind a CDN.
The point of learning this now, before deploying, is that it makes the eventual deployment teach deployment rather than refactoring. If the site needed a runtime, the first AWS step would be fighting the architecture. Because it does not, the first AWS step is purely about S3, CloudFront, DNS and certificates — the actual cloud concepts.
It also keeps the platform portable: the same dist/ folder would serve from
other providers, so I am not locked in by this choice.