One of the design criteria for Unicode was to guarantee lossless round-trip conversion through Unicode for a bunch of encodings that Unicode superceded. If one

One of the design criteria for Unicode was to guarantee lossless round-trip conversion through Unicode for a bunch of encodings that Unicode superceded. If one of those encodings contained both a K and a K-for-Kelvin, then Unicode also had to allocate two codepoints for it. I don't know if that's the case for these characters in particular, but it is one possible explanation. Supporting round-trip conversion may sound pointless now that Unicode is ubiquitous, but it was probably necessary to make gradual adoption practical. Ref: https://en.m.wikipedia.org/wiki/Unicode_compatibility_characters https://en.m.wikipedia.org/wiki/Unicode_compatibility_charac...