toLower/toUpper are tricky to define

[drathier]

String.toUpper only handles a-z. This breaks Swedish, for example. åsa gets uppercased to åSA instead of ÅSA, assuming the å character is a single code point.

Unfortunately, which lowercase characters correspond to what uppercase characters is locale-dependent, so implementing a non-ascii version is tricky. https://stackoverflow.com/questions/12537377/in-haskell-how-can-i-uppercase-a-unicode-character-with-respect-to-current-local

I'd suggest dropping these functions and using css (text-transform: uppercase) to uppercase/lowercase written strings, hoping that the browser locale is correct most of the time. For case-insensitive string comparisons, there's unicode normalization algorithms which can be used.

[BrianHicks]

🤔 this is something we've inherited by making a façade over String, and in fact we're passing directly through to String.

I see two ways forward:

1. remove the functions, as you suggest (and probably roll these fixes into the next major version)
2. add a note about this to the docstring

[BrianHicks]

@drathier would you mind moving this issue to elm/core? I don't think we're going to fix it here.