This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 21412 - Extensibility of @role should be better defined
Summary: Extensibility of @role should be better defined
Status: RESOLVED WORKSFORME
Alias: None
Product: HTML WG
Classification: Unclassified
Component: HTML5 spec (show other bugs)
Version: unspecified
Hardware: PC All
: P2 normal
Target Milestone: ---
Assignee: Robin Berjon
QA Contact: HTML WG Bugzilla archive list
URL:
Whiteboard:
Keywords: a11y, aria
Depends on:
Blocks:
 
Reported: 2013-03-27 12:24 UTC by Robin Berjon
Modified: 2015-06-05 14:33 UTC (History)
8 users (show)

See Also:

Attachments

Description Robin Berjon 2013-03-27 12:24:48 UTC 
The use case is adding values that match new needs, such as those found in epub.

The ARIA role attribute document has a variety of indications about CURIEs and XHTML, but these are hardly applicable.

A clearer description of the extensibility would help.
Comment 1 Henri Sivonen 2013-03-28 13:21:42 UTC 
(In reply to comment #0)
> The use case is adding values that match new needs, such as those found in
> epub.

Do you mean that the stuff that lives in epub:type should live in role? Or something else?

Is there any Reading System implementor interest in epub:type beyond the footnote stuff? Frankly, to me, it looks like someone interested in semantics for the sake of semantics got carried away without being backed by true implementor interest. (I'm aware of BlueGriffon. I'm asking specifically about Reading System implementor interest.)

> The ARIA role attribute document has a variety of indications about CURIEs
> and XHTML, but these are hardly applicable.

The role attribute doc is old XHTML WG stuff. Doesn't apply to HTML5.

> A clearer description of the extensibility would help.

How about: "If you need new roles and you have secured implementor interest from two independent browser engine implementors, mint some and let the PFWG know so that collisions can be avoided."? I don't see a need for a mechanism beyond collision avoidance.
Comment 2 Henri Sivonen 2013-03-28 13:23:12 UTC 
(IIRC, I already suggested this at TPAC 2010.)
Comment 3 Edward O'Connor 2013-05-02 21:21:27 UTC 
I don't think we should use role="" for anything but ARIA roles, so I don't think we need to do anything here.