a while back on hn there was talk of an archive project or webscraper that "got it right"[sic] It didn't rewrite the html but served files by the original path

a while back on hn there was talk of an archive project or webscraper that "got it right"[sic] It didn't rewrite the html but served files by the original path. It sounds like the right approach. Maybe someone remembers what it was called?