Re: Rolling the personalization SC into 4.1.2

[Jason] Consider the following two conditions:

  1.  Vocabulary from a taxonomy is used on the site to declare the purpose or function of user interface components (substitute here your preferred substantive requirement).
  2.  The chosen taxonomy is accessibility-supported (or perhaps something slightly stronger than accessibility-supported – meaning, in this case, actively used by user agents or assistive technologies to enhance accessibility for users).
If these conditions are met, so is the proposal. I think this is plainly and reliably testable.

[AC] I think it would need quite an involved declartation mechanism. For example, are you using the whole taxonomy?

There are parts of coga-personalisation that are fairly obvious how to do, and they are concrete. There are other parts that are more open and could be implemented in hugely different ways, but it's part of the same spec.
Also, as an author or tester, how do you know which bits of a spec/taxonomy are accessibility supported?

This openness & flexibility is really screaming triple-A to me. As a first step, I think a focused double-A SC on key elements is the best starting point.

Cheers,

-Alastair