For those interested in Chinese-language computing, there's a fascinating rabbit hole.
In too-brief summary, character-based Chinese has almost died twice in modern history.
First, with the typewriter, where a mechanically unique typewriter had to be invented to span the Chinese character alphabet.
Second, with early low-bit computers, where the address space was insufficient to span the character alphabet, and what's essentially a multi-character-to-single-character encoding had to be invented.
In too-brief summary, character-based Chinese has almost died twice in modern history.
First, with the typewriter, where a mechanically unique typewriter had to be invented to span the Chinese character alphabet.
Second, with early low-bit computers, where the address space was insufficient to span the character alphabet, and what's essentially a multi-character-to-single-character encoding had to be invented.
More detail: https://www.radiolab.org/podcast/wubi-effect-2308
It'll be curious to see how the LLMs evolve, given have alphabetically distinct Chinese is.