W3C home > Mailing lists > Public > public-i18n-core@w3.org > July to September 2006

[Fwd: Re: a request for comment from the i18n WG]

From: François Yergeau <francois@yergeau.com>
Date: Tue, 26 Sep 2006 06:47:54 -0700
To: public-i18n-core@w3.org
Message-id: <45192F8A.7050207@yergeau.com>

-------- Message original --------
Sujet: Re: a request for comment from the i18n WG
Date: Mon, 11 Sep 2006 11:36:56 -0700
De: François Yergeau <francois@yergeau.com>
Pour: C. M. Sperberg-McQueen <cmsmcq@acm.org>
Copie: Felix Sasaki <fsasaki@w3.org>
Références: <3984A1C2-9E23-46A8-A49C-E3D8566DF012@acm.org>

C. M. Sperberg-McQueen a écrit :
> Francois, Felix,
> it occurs to me that it would be useful to know i18n's reaction to the 
> proposal
> made by the XML Query and XSL WGs about the type xsd:string, namely
> that for certain purposes strings should be treated as equal (not 
> necessarily
> identical) if they have the same Unicode-normalized form.

That sounds like a good idea.  I note, however, that looking at the
latest public datatype spec (2006/02/17) it doesn't seem that it would
have any effect!  Unless I missed something of course, I'm not a schema

> http://www.w3.org/Bugs/Public/show_bug.cgi?id=3245

Ah!  "assuming, of course, that equality and not identity is used for
things such as enumerations"  There comes some effect!

> I can imagine either that i18n would be happy with the idea, since it
> would give some prominence to questions of normalization, or
> alternatively that you might be very unhappy with it, as making it
> too easy for generators of XML to evade the consequences of failure
> to perform Unicode normalization.

It depends if one thinks that early-normalization is still a viable
proposition.  You are correct that 3245 would help generators get away
without normalizing early, in the same way that lenient browsers help
(or even encourage) authors get away with lousy HTML.  But for
normalization (or more properly Unicode canonical equivalence) there is
the possibility of late checking to save the day.

Personnally I've given up on early-normalization, not because I think
it's not a good idea, but because I'm now convinced that the world won't
embrace it.  But that's only my personal opinion, the WG will need to
discuss your idea before we react to it officially.

> I'd send this to the WG myself, as an inquiry, but at the moment I'm
> not entirely certain which i18n list I should address myself to.  Feel
> free to forward this mail to the appropriate group(s).

Just did.


> thanks.
> Michael
Received on Tuesday, 26 September 2006 13:48:06 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:23:01 UTC