On Jun 10, 2011, at 9:46 AM, Cameron Heavon-Jones <[email protected]> wrote:
> To bring it back to the current case, i'm not sure that it is appropriate to > specify a language extension for the document. This would imply that there > MAY be new words, spellings or meanings in the resulting text and that just > isn't the case. Private use tags do indeed imply that there may be differences in language and style, but because the tags are private, the differences are _only_ meaningful to the people who have agreed on their use -- in this case, the W3C authors. For everyone outside that agreement, the tags provide no meaningful new information. One might well wonder _why_ the W3C would want to distinguish documents written in Ian Hickson's inimitable style from others (phrases like "coming apocalypse" and "robot war" jump to mind for some reason), but the reasons are by definition unimportant, at least to us mere mortals. For what it's worth, I think the private use tag provides a rather clever test for user agent conformance. UAs should treat en-US-x-anything _exactly_ like en-US. There are enough browsers out there that seem to want to conform only if and when they want to (I'm looking at you, Microsoft, Freedom Scientific and tattooed teenagers) that such a test is well worthwhile. Chris
