If we currently have correct, pure-US spelling, then changing it piecemeal would seem to be a regression. Maybe we can explain that this is a considerable effort and we'll do it for C++20?

The big one is "undefined behavior". I think "program" is OK in UK English, even if traditionally it would be programme.

There's an enum color in an example in [dcl.enum], which is OK for an example, although the text after it says "colors". The second half of the example could use enum class Colour instead of enum class Col. There's another enum color in [diff.dcl].

We can't change the errc::operation_canceled enumerator.

[category.messages] talks about "catalogs", but we can't change the messages_base::catalog type.

[re.traits] says "honor".

This is not an exhaustive list.

I would expect all program-level things (e.g. class names in the standard library) to be spelled with US spelling, out of tradition. It seems odd to convert the rest of the text to UK spelling. In any case, that's some serious effort at the last minute to get right throughout. Do we have a channel to ISO to ask whether they will reject publication if we don't fix this for C++17? After all, all previous standards were published with US spelling.

Note that among the dictionaries suggested by the drafting directives is the US English Websters. Perhaps we could arrive at a permanent settlement that we use US English, because many of the constructions in the language itself do, and it would be less confusing to use a single variant everywhere.

Response: REJECTED. The document defines many terms that must be spelled the same way by conforming implementations and by users of those implementations, and these terms use US English spelling. It would be confusing for the document text to use UK English spelling when describing these terms.