▦ silicon·eras

Problem

Half a century of consumer hardware — CPUs, memory, storage, displays — is a story of overlapping exponential curves and the eras they define. But the raw history lives in scattered spec sheets and wiki tables: impossible to feel the trajectories, the plateaus, or where one era handed off to the next. A static chart flattens it; an interactive one usually means a backend and a data fetch.

Approach

silicon·eras is a single-page explorable written in pure Rust (Dioxus 0.7), compiled to WebAssembly. A long-format dataset.csv is parsed and typed at build time by build.rs and embedded straight into the wasm — so there is no backend, no API, and no network after the module loads; a data edit flows through CI with no extra step. The centerpiece is an era-band trajectory chart: envelope or three-line tiers, log/linear and absolute/indexed axes, multiple era schemes, pins and plateaus, a hover tooltip, an era brush, full keyboard control, and URL-shareable state so any view can be linked. All colors are var(--viz-*) tokens that inherit the host page's data-theme, so it recolors instantly with the site's light/dark toggle.

Outcome

A fast, offline-capable way to see hardware history move — pivot the axes, brush an era, share the exact view — with the whole dataset reproducible from one CSV and the entire app running in your tab.

Explore it live at /silicon-eras — real Rust compiled to WebAssembly, the full dataset embedded, nothing leaves your browser.