- From: Steve Faulkner <faulkner.steve@gmail.com>
- Date: Wed, 13 Mar 2013 14:45:47 +0000
- To: Sailesh Panchang <sailesh.panchang@deque.com>
- Cc: Leif Halvard Silli <xn--mlform-iua.no@xn--mlform-iua.no>, James Craig <jcraig@apple.com>, Joseph Scheuhammer <clown@utoronto.ca>, wai-xtech@w3.org
- Message-ID: <CA+ri+V=Yb12KgPd+zVNfGRsyJjdo2bkY1abhp2=bja8YSuVCUw@mail.gmail.com>
Hi Sailesh, i think you are getting the wrong end of the stick I am looking at it from an API implementation point of view. IN HTML5 we have an element <hgroup> that is meant to convey a subheading/subtitle semantic, there is no role for subheading in APIs. If we want to be able to convey something is a subheading we will either need a new role or we could use an existing role such as heading but with no level set. 'heading' is just a token its the AT/APIs that provide human understandable output for the roles/states and properties. with regards SteveF <http://www.paciellogroup.com/resources/wat-ie-about.html> -- HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/> <http://www.paciellogroup.com/resources/wat-ie-about.html> On 13 March 2013 14:29, Sailesh Panchang <sailesh.panchang@deque.com> wrote: > A role=header with no level does not make sense. > Headings are meant to convey hierarchy and structure. Authors should > use them consistently across the site or within a collection of pages. > It is alright if a page has a single heading. But it should have a level. > Without a level, I'd assume maybe the author simply wanted to > emphasize that text and that content is not really a section heading. > So role=heading without a level should be flagged as a violation. > But really when one can style headings as required or use image of > text within an h-tag, the need to use role=header should be limited. > Simply complicates matters for user agents and AT and accessibility > eventually. > Sailesh Panchang > Tel 571-344-1765 > > > On 3/13/13, Steve Faulkner <faulkner.steve@gmail.com> wrote: > >> Regarding heading without level as indicator for generic subheading: It > >> doesn't seem to me that 'heading level zero', can be interpreted as > >> subtitle. > >> > > > > don't see why not > > > > > > But there is also some situations where heading level perhaps is > >> irrelevant. For instance, for an independent page with a single > >> heading. I don't hover know if AT user agree ... > >> > > > > there is no <h0> in HTML > > > > with regards > > > > SteveF <http://www.paciellogroup.com/resources/wat-ie-about.html> > > -- > > HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/> > > <http://www.paciellogroup.com/resources/wat-ie-about.html> > > > > > > On 13 March 2013 11:50, Leif Halvard Silli < > > xn--mlform-iua.no@xn--mlform-iua.no.no> wrote: > > > >> Regarding heading without level as indicator for generic subheading: It > >> doesn't seem to me that 'heading level zero', can be interpreted as > >> subtitle. Butif it gets announced as subtitle/subheading, then I guess > >> it could work. > >> > >> But there is also some situations where heading level perhaps is > >> irrelevant. For instance, for an independent page with a single > >> heading. I don't hover know if AT user agree ... > >> > >> Leif H Silli > >> > >> Steve Faulkner, Wed, 13 Mar 2013 11:21:31 +0000: > >> > Hi all, > >> > > >> > I have been mulling over whether a heading role without an explicit > >> > level > >> > set could be used as an indicator of a generic "subheading" semantic. > >> > > >> > thoughts? > >> > > >> > with regards > >> > > >> > SteveF <http://www.paciellogroup.com/resources/wat-ie-about.html> > >> > -- > >> > HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/> > >> > <http://www.paciellogroup.com/resources/wat-ie-about.html> > >> > > >> > > >> > On 13 March 2013 00:04, James Craig <jcraig@apple.com> wrote: > >> > > >> >> I don't have a strong preference one way or the other, but it seems > >> >> like > >> >> this should be raised as an issue on the UAIG. > >> >> > >> >> On Mar 12, 2013, at 2:57 PM, Leif Halvard Silli < > >> >> xn--mlform-iua@xn--mlform-iua.no> wrote: > >> >> > >> >>> WAI-ARIA 1.0 says about aria-level: [1] "The value for aria-level is > >> >>> an > >> >>> integer greater than or equal to 1." And about role=heading, it > says: > >> >>> [2] "If headings are organized into a logical outline, the > aria-level > >> >>> attribute may be used to indicate the nesting level." > >> >>> > >> >>> However, which level does a role="heading" element without an > >> >>> aria-level="*" attribute have? According Firefox+VoiceOver, it has > >> >>> level zero. But according to Safari+VoiceOver, it has level one. I > >> >>> have > >> >>> not checked other ATs. Should a level be announced at all? > >> >>> > >> >>> Tested on this example: > >> >>> > >> >>> <foo role="heading"> > >> >>> Lorem Ipsum. > >> >>> </foo> > >> >>> > >> >>> [1] http://www.w3.org/TR/wai-aria/states_and_properties#aria-level > >> >>> [2] http://www.w3.org/TR/wai-aria/roles#heading > >> >>> -- > >> >>> leif halvard silli > >> >>> > >> >> > >> >> > >> >> > >> > > >
Received on Wednesday, 13 March 2013 14:47:03 UTC