A last missing link for interoperable representation

wjgo_10009@btinternet.com via Unicode unicode at unicode.org
Wed Jan 9 04:29:36 CST 2019

I suggest that a solution to the problem would be to encode a COMBINING 
ITALICIZER character, such that it only applies to the character that it 
immediately follows. So, for example, to make the word apricot become 
displayed in italics one would use seven COMBINING ITALICIZER 
characters, one after each letter of the word apricot. The display could 
be sorted out using an OpenType font by treating each pair of a letter 
and a COMBINING ITALICIZER as a ligature. If, say, the glyph name of 
COMBINING ITALICIZER were italic then the glyph for c italic could be 
c_italic and so plain text might well be copyable from a PDF (Portable 
Document Format) document and pasted to WordPad as plain text retaining 
the COMBINING ITALICIZER character, depending upon which application 
program is used to produce the PDF document and which PDF reader is in 

This would seem a workable solution. Many years ago I suggested having 
characters that would have been comparable in use in plain text as to 
how italics is switched on and off in HTML (Hypertext Markup Language) 
yet was advised that such an encoding would make plain text stateful and 
thus would not be agreed for encoding. That objection might well still 
be the case today. So using a COMBINING ITALICIZER character would avoid 
that objection and would also provide a solution that could be 
straightforwardly implemented using existing OpenType technology.

William Overington
Wednesday 9 January 2019

More information about the Unicode mailing list