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

Re: Six Reasons Not to use XML Attributes

• From: Liam R E Quin <liam@w3.org>
• To: "Christopher R. Maden" <crism@maden.org>
• Date: Wed, 07 Mar 2012 17:59:28 -0500

On Fri, 2012-03-02 at 12:51 -0500, Christopher R. Maden wrote:

> Consider the default processing rule for HTML browsers: [unknown
> attributes ignores, unknown element content displayed]

There are plenty of ways in which a system could be designed to be
extensible without such a flawed design as HTML.

Forcing title and alt displayed content into attributes means you can't
associate a language with them, can't include Japanese ruby or other
markup.

Conversely, when inline CSS and JavaScript were added there was no good
way to stop older browsers displaying the content, so HTML comments are
used.

A better way might have been to distinguish elements syntactically if
they could contain text to be displayed. Example - start non-displayed
element named with an underscore.

Attributes have a place, for machine-readable metadata. The distinction
is useful. So we agree on that part! :-)

Liam

-- 
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
The barefoot typographer

• References:
  • Six Reasons Not to use XML Attributes
    • From: "Costello, Roger L." <costello@mitre.org>
  • Re: Six Reasons Not to use XML Attributes
    • From: "Christopher R. Maden" <crism@maden.org>