- From: klensin via GitHub <sysbot+gh@w3.org>
- Date: Thu, 04 Feb 2016 03:37:08 +0000
- To: public-i18n-archive@w3.org
Note that "locale-dependent" as used above, is different from "language sensitive" (or "language dependent") as used further on in the text... at least unless one defines languages very precisely by place and/or time. The Turkic example, which is cited so often that many people believe it is the only case (and is cited in TUS as "locale-dependent" when it is really language-dependent), is ultimately not a lot different from the cases in which, while "ß" can always be transformed into "ss" (even if the result is seen as a serious spelling error on some locales), "ss" cannot be transformed in some cases to "ß" without loss of meaning and/or silly states. The Unicode Standard seems to say that the main application of case folding is for caseless matching (subsection of that name in TUS 7.0, Section 5.18) and that using case folding for mapping and replacement operations is unwise. It seems to me that point is important; it is hard to find or deduce from this section. -- GitHub Notification of comment by klensin Please view or discuss this issue at https://github.com/w3c/charmod-norm/issues/68#issuecomment-179601456 using your GitHub account
Received on Thursday, 4 February 2016 03:37:11 UTC