[whatwg] Allowed characters in attribute names (was: Re: Stepsfor finding one or two numbers in a string)

[Křištof Želechovski]

If the IDE can shield the element name, it can shield the attribute name as
well.  Alternatively, it can be automatically before publishing to the Web.
Making everybody else learn Chinese or Arabic is not a good idea right now.
As I already stated, the judgment day may be near, but then we will have to
start translating from html and WWW-or rather accept the translations as
they are proposed by the other party-and not from some exotic attribute
name.
You can tell the day is coming when you see ideograms on your BIOS welcome
screen.  Until that happens, the machines out there are expected to
understand US English only.
Cheers
Chris

-----Original Message-----
From: whatwg-bounces at lists.whatwg.org
[mailto:whatwg-bounces at lists.whatwg.org] On Behalf Of Charles McCathieNevile
Sent: Monday, June 18, 2007 1:11 AM
To: Křištof Želechovski; whatwg at whatwg.org
Subject: Re: [whatwg] Allowed characters in attribute names (was: Re:
Stepsfor finding one or two numbers in a string)

On Fri, 15 Jun 2007 00:51:46 +0200, Křištof Želechovski  
<giecrilj at stegny.2a.pl> wrote:

> Your hypothetical author is unable to insert an embed element because  
> embed is all English to him.  Being able to use a Mandarin attribute
> name will not help him much because he cannot produce the element to
> use it with.

In my example I had the author using an IDE that is localised already for  
known elements. People really do this. The ones I am aware of actually use  
japanese or arabic, and do it for convenience - although in principle they  
can normally work with some kind of latin script, they are not very  
familiar with english, and would be as happy to use "inbed" as "embed",  
and happier with something that they recognise more instinctively. China,  
in particular, is quite keen on localising everything. India seems more  
inclined to do so, as well, as time goes on and they become a more  
important player in technology.

The scenario does require that the language is extensible - this can be  
done with XML, presumably including the XML version HTML 5, but not HTML 5  
as currently proposed.