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

Re: Using the 4 Great XML Namespace Attributes

• From: Andrew Welch <andrew.j.welch@gmail.com>
• To: "Costello, Roger L." <costello@mitre.org>
• Date: Wed, 1 Sep 2010 21:02:11 +0100

> --------------------------------
> The Great XML Namespace Attributes
> --------------------------------
> The XML namespace contains 4 great attributes:
>
> 1.      The lang attribute: use this on an element to state the language of its content.
> 2.      The id attribute: use this on an element to uniquely identify it.
> 3.      The space attribute: use this on an element to state that you want its whitespace nodes preserved during processing.
> 4.      The base attribute: use this on an element to create a base URL; URL's within the element can be relative to the base URL.

Are they really great?  Should they be used at all?

Take xml:lang as an example, and it's use in XHTML.  Should you define
the language of the content at the XHTML level, or at the XML level?

To me it only makes sense to define it in XHTML, but if you read this:

http://www.w3.org/TR/i18n-html-tech-lang/#ri20040429.092928424

...it's a bit of mess, ending with it has to be defined it at the xml
level.  I don't know why - is the intention the xml parser filters the
non-applicable languages, so the receiving application doesn't know
they exist?  What if you want to present a choice of available
languages?

A similar thing applies for id - do you want your parse to fail with a
message from the xml parser, or let your application manage uniqueness
and the duplicate id failure messages.


-- 
Andrew Welch
http://andrewjwelch.com

• References:
  • Using the 4 Great XML Namespace Attributes
    • From: "Costello, Roger L." <costello@mitre.org>