Türker Yıldırım
A trial on one suite, honestly reported. Nothing was wrong with the tests we had.
v1, v2 and v3 in production, and v1 has four clients. Every change is made three times.
5.2 in July, and custom properties throughout the components. A Sass fork that had to be rebased on every patch release.
A polyglot system needs more than one person able to read each part, and that is a lower bar than being able to write it. Setting the bar…
July, and the Composition API backported to the 2.x line. 214 components and no appetite for a big-bang rewrite.
July’s release moves component styling onto CSS variables, which makes runtime theming possible without a Sass fork. A fork that had to be rebased on every patch release…
July, and a major whose changes are mostly defaults. The dev server port moved and every script assumed 3000.
6.0, and locking that is per block rather than per template. templateLock all was too much and nothing was too little.
6.0 in May, and styles/ as a directory of theme.json fragments. Four client sites that differ only in colour and type.
7.0 in April, and a script registry that survives a restart. EVAL with a SHA nobody can map back to source.