Just as additional context, since this seems to come up a lot - that does not mean that UTF-8 is bad. The main alternative is UTF-32, which provides random acc

Just as additional context, since this seems to come up a lot - that does not mean that UTF-8 is bad. The main alternative is UTF-32, which provides random access to codepoints rather than bytes, but since Unicode graphemes consist of multiple codepoints, UTF-32 is effectively a variable width encoding as well for most purposes, just one that uses much more memory than UTF-8 most of the time and never uses less.