W3C home > Mailing lists > Public > www-html@w3.org > December 2004

Re: XHTML attributes and namespace, clarification needed

From: Karl Dubost <karl@w3.org>
Date: Tue, 14 Dec 2004 10:40:20 -0500
Message-Id: <760FA6E4-4DE6-11D9-AC70-0003934BEBF0@w3.org>
Cc: www-style@w3.org
To: W3C HTML <www-html@w3.org>


Le 14 déc. 2004, à 09:04, Daniel Glazman a écrit :
> Robin Berjon wrote:
>
>> No it doesn't. Attributes are things that float in space, they're 
>> attributes of something (hence the name).
>
> Absolutely. That's why they should acquire the something's namespace...

which means as Bjoern showed
Le 14 déc. 2004, à 08:14, Bjoern Hoehrmann a écrit :
>   <ul     type = '...'>
>   <style  type = '...'>
>   <input  type = '...'>
>   <button type = '...'>

that it's not that easy to do now depending on the way it's defined.

If you put a default namespace on attribute, we would break the old 
specification like XHTML 1.0

What is the default semantics of xhtml:type? or we have to redefine it 
locally which goes back to the initial problem. We have to understand 
the local context (element name) to understand the semantics of the 
attributes.

Which raises another question: Should attributes have unique names in a 
specification? I think there are pros and cons on both sides, then no 
easy solutions. :)))

Unique names for attributes will drive us to strange names like
>   <ul     type = '...'>
>   <style  type = '...'>
>   <input  type = '...'>
>   <button type = '...'>

	===> ltype, stype, itype, btype :) and zillions of other terms :)
Received on Tuesday, 14 December 2004 15:40:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:01 GMT