Treatment of U+2028

Dear XML editors,

One of the changes proposed for XML 1.1 is to include character U+2028 
(LINE SEPARATOR) into the list of characters subject to end-of-line
normalization. In Section 1.3, the rationale is simply:

XML 1.1>   For completeness, the Unicode line separator character, 
XML 1.1>   #x2028, is also supported [=converted to #xA].

The semantic of the character in Unicode is defined as follows:

Unicode>   2028  LINE SEPARATOR may be used to represent 
Unicode>    this semantic unambiguously

In an application that deals with typographical presentation, 
it may be desirable to have a character that indicates a line break - 
a character equivalent of <BR/>. U+2028 is a natural candidate
for this; but if it is suppressed in normalization, the application will 
not be able to reach it anymore.  

I believe that U+2028 should be exempt from the normalization,
for XML applications to be free to employ its Unicode semantics.
After all, the whole range of special Unicode space characters 
(U+00A0, U+2000 to U+200B) is preserved; the line separator
character is no different.
 
Best regards,
Nikolai Grigoriev
RenderX

Received on Thursday, 13 November 2003 16:26:52 UTC