I think for a plaintext format to be "complete", it needs some mechanism of associating the language with some segment of text. Plaintext formats that don't acknowledge unified characters are just Latin-biased.
that's basically point - you can open an ascii file now because utf-8 is ascii oriented, but a utf-8 first editor will struggle with an old french text file for example. plain text has inbuilt biases which have changed over time, it's not as pure as simple as people say.
That's something I've always thought; plain text is pure and wonderful for me, an Anglo-writing American, because most of these formats were written for people like me.
I suspect for nearly every other language (or at least any language that doesn't use the ~100 characters/symbols used in the English alphabet), old ASCII text isn't terribly useful.