[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: XML versus Unicode ... here are the facts abouttheir diffe

• From: Liam R E Quin <liam@w3.org>
• To: "Costello, Roger L." <costello@mitre.org>
• Date: Thu, 31 Jan 2013 16:51:15 -0500

On Thu, 2013-01-31 at 20:59 +0000, Costello, Roger L. wrote:

> Assume: This XML tag uses the precomposed ñ character:
> 
> 	<Martiñez>
> 
> Assume: This XML tag uses 'n' plus the "combining tilde" character:
> 
> 	<Martiñez>
> 
> Fact: The two tags are visually IDENTICAL. More precisely, the glyphs on display screens are IDENTICAL.
This may or may not be true depending on the font and the font rendering
system. But it is likely.


> Fact: Below are two representations of the SAME CHARACTER:
> 
> 	a. precomposed ñ character
> 
> 	b. 'n' plus the "combining tilde" character 
> 
> Fact: According to the Unicode standard, the two representations ARE EQUIVALENT.
> Fact: According to the XML standard, the two representations ARE NOT EQUIVALENT.

Correct.
> 
> Fact: According to the Unicode standard, applications must treat the
> two representations exactly the SAME. Applications must compare the
> two representations as EQUAL.

"MUST" or "SHOULD"? Do you have a reference?
> 
The XML choice of Unicode was controversial at the time - Unicode was
not so well established as today. It seemed overly onerous to require
normalization. In addition, lower layers of software shouldn't usually
change user's data without telling higher layers, although XML violates
that in so many other ways I'm not sure it's really an issue. It was
certainly not a design principle.

For good or ill it's not something that could easily be changed today,
because existing systems may depend on one or other representation. 

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
Ankh: irc.sorcery.net irc.gnome.org freenode/#xml

• References:
  • XML versus Unicode ... here are the facts about their differences
    • From: "Costello, Roger L." <costello@mitre.org>