You are blind; and you want to read this blog. Do you have to wade through the huge left hand side menu of crap before you can find out any information. Wouldn't it be great if your screenreader could guess that a menu is a menu and content is content? (ignoring one unless you ask for it).
- Author creates site, marks up css with classes and names like menu and content and comments
- Author or third party describes in generic terms what a menu is, what content is in an ontology. For site specific things, you can extend basic concepts like UnimportantInformation to create a meaningful description of your site. There are editors for this.
- This is made available with a LINK tag to the RDF.
- A semantic web enabled screenreader comes along. It knows what Content is; and can map the CSS styles for content to the Content object. It doesn't know what CloCkWeRXAboutPage is; but it knows CloCkWeRXAboutPage extends UnimportantInformation.
- It reorganises the layout of the content to provide a better user experience for the end user. Unimportant stuff gets left behind, important stuff gets upped in priority
My only thought is the affect of spam / advertising if this took off.