You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The simple styles that use roman letters or international digits inside an em-square have fullwidth in their name.
fullwidth-lower-alpha
fullwidth-upper-alpha
fullwidth-decimal
fullwidth-lower-roman
fullwidth-upper-roman
Counter styles with precomposed circles, parentheses or dots without using suffix: do not have the fullwidth- prefix.
circled-decimal
circled-lower-latin
circled-upper-latin
dotted-decimal
double-circled-decimal
filled-circled-decimal
parenthesized-decimal
As far as I know, Unicode inherited theses characters from legacy East Asian character sets/encodings primarily for round-trip compatibility with existing content. For authors who are not familiar with the history of character encodings, fullwidth is completely opaque and meaningless. That’s most of them, at least outside China, Japan and Korea. Yet, Europeans, Africans or Americans used only to the roman script might still feel tempted to use these counter styles.
Should they be specified at all?
Should we find a better prefix (or reuse cjk-) and apply it consistently?
They are used for CJK lists in places where users specifically want the fullwidth versions rather than the non-fullwidth. One example where they are useful is for vertically set lists, since fullwidth characters sit upright. Dotted decimal and parenthesised are particularly useful for vertical lists because the reduce the complexity of authoring if you want to use those styles.
My belief is that the fullwidth prefix is useful to distinguish from the non-fullwidth.
I don't want to add to just one (or to all) of CJK. Note that they may be used for Mongolian too. The current section division helps you find things quickly - ie. look under Latin for alpha, and under Digits.. for numeric templates.
If I understand correctly, we do agree that most (if not all) of the styles I mentioned are only supposed to be used within “fullwidth” typography, i.e. with sinographic character grid. They should not be used within documents that are primarily using a different script, the Latin/Roman one in particular. The document should reflect these recommendations.
I still believe it’s harmful then to group them with Digit and Latin styles, respectively, because none of them is either. My point is, they’re Eastern (compatibility) characters that just happen to look like Western letters and digits (with some adornment).
The simple styles that use roman letters or international digits inside an em-square have
fullwidth
in their name.fullwidth-lower-alpha
fullwidth-upper-alpha
fullwidth-decimal
fullwidth-lower-roman
fullwidth-upper-roman
Counter styles with precomposed circles, parentheses or dots without using
suffix:
do not have thefullwidth-
prefix.circled-decimal
circled-lower-latin
circled-upper-latin
dotted-decimal
double-circled-decimal
filled-circled-decimal
parenthesized-decimal
As far as I know, Unicode inherited theses characters from legacy East Asian character sets/encodings primarily for round-trip compatibility with existing content. For authors who are not familiar with the history of character encodings,
fullwidth
is completely opaque and meaningless. That’s most of them, at least outside China, Japan and Korea. Yet, Europeans, Africans or Americans used only to the roman script might still feel tempted to use these counter styles.cjk-
) and apply it consistently?I believe the circled ones are okay, the rest should not be used, at least not in Western texts.
The text was updated successfully, but these errors were encountered: