Re: XML Blueberry (non-ASCII name characters in Japan)

[John Cowan <jcowan@reutershealth.com>]

Thomas B. Passin wrote:

> Who would be able
> to remember the exclusion rules for thousands and tens of thousands of
> characters?


The rules aren't changing much:

	If it's a letter or : or _, it's in.
	If it's a digit or diacritic or dot or -, it's in
		(except initially).
	If it's a compatibility character, it's out.
	If it's one of about 25 backward compatibility exceptions,
		it's in anyway.

It's the list of characters to which these rules are applied that is
growing in Blueberry.

> To me, then, the question reduces to just this:  will these characters show
> up as printable characters in anyone's everyday, normal text editor-like
> application?


Well, provided the application understands the encoding used,
and provided your system has the font.

> No

> problem with keeping the current disallowed non-name characters, I suppose.
> They amount to only a few special cases, not all of them even visible.


Thousands of non-letter, non-digit characters (symbols, punctuation
marks, etc.) are currently excluded, and rightly so IMHO.

 
> Conversely, I don't want my EditPlus-generated markup to suddenly be
> rejected because of a change in the markup character set.


I don't think there's any question of ever shrinking the set.

-- 
There is / one art             || John Cowan <jcowan@reutershealth.com>
no more / no less              || http://www.reutershealth.com
to do / all things             || http://www.ccil.org/~cowan
with art- / lessness           \\ -- Piet Hein