KENSAI Research: Per-Language JSON Is a Publishing Boundary

A multilingual blog stops being credible the moment English starts pretending to be the source of truth for every language. Real parity comes from per-language HTML, per-language JSON, and listings that rebuild from the right files.

Why the boundary matters

A multilingual publishing system only stays trustworthy when each language owns its own source of truth. English can be the reference for editorial intent, but it should never silently absorb localized posts into its own JSON index or pretend that translation happened just because the English article exists.

What breaks when English absorbs locale posts

Once non-English entries leak into the English index, the public surface becomes ambiguous. Counts drift, language overviews stop matching what users actually read, and same-day freshness turns into a bookkeeping illusion instead of a route-level fact.

What same-day parity actually requires

The real loop is simple: publish the localized HTML, rebuild the matching language JSON from those files, and regenerate the overview so the listing reflects the language-specific source tree. That is the only version of parity that survives verification.

The KENSAI takeaway

Per-language JSON is not cleanup. It is a publishing boundary. When each language keeps its own HTML, metadata, slug, and index entry, the blog stays precise, searchable, and honest across every locale.