email@example.com via Unicode
unicode at unicode.org
Fri Feb 8 11:16:09 CST 2019
Andrew West wrote:
> Just reminding you that "The initial character in a variation sequence
is never a nonspacing combining mark (gc=Mn) or a canonical
decomposable character" (The Unicode Standard 11.0 §23.4). This means
that a variation sequence cannot be defined for any precomposed
letters and diacritics, so for example you could not italicize the
word "fête" by simply adding VS14 after each letter because "ê" (in
NFC form) cannot act as the base for a variation sequence. You would
have to first convert any text to be italicized to NFD, then apply
VS14 to each non-combining character. This alone would make a VS
solution unacceptable in my opinion.
As it happens I was not aware of that before, and in fact I had already
produced a PDF document for submission to the Unicode Technical
Committee when I read your post.
So, it is an issue that needs to be resolved.
I am a researcher and I am looking for the best way to do this so as to
get a good result that people can use, I am not trying to assert that my
suggestion is necessarily the best way to do it. For example, I accepted
the suggestion that James made. The meeting of the Unicode Technical
Committee is not due until April and hopefully some other people will
send in documents and comments on the topic.
Hopefully the issue that Andrew mentions can be resolved in some way.
Friday 8 February 2019
More information about the Unicode