Re: CfC: return HTML 5.0 to Last Call

following up on - see notes added inline

HTML 5 CR Implicit ARIA Semantics at risk:


   1. output element  = status role (note: spec bug
   https://www.w3.org/Bugs/Public/show_bug.cgi?id=26135 discussing with
   implementers)
   2. link element that creates a hyperlink = link role (note: spec bug
   https://www.w3.org/Bugs/Public/show_bug.cgi?id=26099 discussing with
   implementers)
   3. input element ) with a type attribute in the Text, Search, Telephone,
   URL, or E-mail states with a suggestions source element) = combobox role
   (note: need further checking and bug filing)
   4. input element with a type attribute in the Number state = spinbutton
   role (note: after checking Safari - PASSES with 2 implementations)
   5. img element whose alt attribute's value is empty = presentation role
   (note: safari passes, bugs filed on chrome/FF/IE. expect to be fixed in
   firefox at least in short time.)
   6. hr element = separator role (note: after checking Safari - PASSES
   with 2 implementations)



--

Regards

SteveF
HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/>


On 16 June 2014 19:29, Steve Faulkner <faulkner.steve@gmail.com> wrote:

>
> From the test results http://stevefaulkner.github.io/html-mapping-tests/
>
> 55 passing requirements
>  6 fail
>
> HTML 5 CR Implicit ARIA Semantics at risk:
>
>
>    1. output element  = status role
>    2. link element that creates a hyperlink = link role
>    3. input element )with a type attribute in the Text, Search,
>    Telephone, URL, or E-mail states with a suggestions source element) =
>    combobox role
>    4. input element with a type attribute in the Number state =
>    spinbutton role
>    5. img element whose alt attribute's value is empty = presentation
>    role
>    6. hr element = separator role
>
>
> --
>
> Regards
>
> SteveF
> HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/>
>
>
> On 16 June 2014 19:19, Janina Sajka <janina@rednote.net> wrote:
>
>> Hi Robin, Steve, PLH, All:
>>
>> An update on the PF request ...
>>
>> The relevant CfC in the PFWG[1] asks EITHER for testing or an RFC2119
>> redesignation to "informative."
>>
>> As Robin notes, Steve responded to the PF concern by creating and
>> running tests whose results are clearly impressive.  Today's ARIA
>> telecon discussed these developments with Steve.[2]
>>
>> We came to the following conclusions:
>>
>> *       Marking the entire section "At Risk" would be excessive, as much
>> *       of it is correct.
>>
>> *       There are a small number of assertions, perhaps a dozen, which
>> *       need to be corrected. Steve will be providing a list of these
>> *       shortly. He is also intending to file bugs on these items during
>> *       LC. Following today's discussion in the ARIA Task Force,
>> *       these are the only items we would consider "At Risk."
>>
>> What is unclear to us is whether correcting these items, which would
>> involve edits to the normative text of the 5.0 specification during Last
>> Call, is practical. We do not seek to force HTML back to CR--that is not
>> our intent here.
>>
>> I'm looking for suggestions for resolving the failures identified by
>> Steve's testing while keeping HTML moving toward PR and TR.
>>
>> Janina
>>
>> [1] http://lists.w3.org/Archives/Public/public-pfwg/2014Jun/0067.html
>> [2] http://www.w3.org/2014/06/16-aria-minutes.html
>>
>> Robin Berjon writes:
>> > Hi Janina,
>> >
>> > On 13/06/2014 16:13 , Janina Sajka wrote:
>> > >However, we do believe there is one section of the candidate LC
>> document
>> > >whose RFC2119 status may need to be changed. Therefore, we are
>> > >requesting that the following section be marked "At Risk" for the LC:
>> > >
>> > >
>> http://htmlwg.org/heartbeat/WD-html5-20140617/dom.html#sec-strong-native-semantics
>> > >
>> > >PF notes that the normative requirements on user agents in this section
>> > >are not tested and believes that they cannot be appropriately
>> > >implemented from the specification alone.
>> >
>> > I understand where you are coming from but I feel uncomfortable making
>> this
>> > at risk (which does slate for removal) when it is in fact implemented,
>> and
>> > implemented relatively well to boot!
>> >
>> > Looking at these tests:
>> >
>> >     http://stevefaulkner.github.io/html-mapping-tests/
>> >
>> > If you forget about elements that are only in 5.1 (details, dialog) it
>> > actually looks pretty good. Sure enough, there are a few failures, but
>> they
>> > don't look like showstoppers to me. The tests leave me rather optimistic
>> > overall.
>> >
>> > Based on this, can you please clarify your concerns regarding
>> > implementability?
>> >
>> > --
>> > Robin Berjon - http://berjon.com/ - @robinberjon
>>
>> --
>>
>> Janina Sajka,   Phone:  +1.443.300.2200
>>                         sip:janina@asterisk.rednote.net
>>                 Email:  janina@rednote.net
>>
>> Linux Foundation Fellow
>> Executive Chair, Accessibility Workgroup:       http://a11y.org
>>
>> The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
>> Chair,  Protocols & Formats     http://www.w3.org/wai/pf
>>         Indie UI                        http://www.w3.org/WAI/IndieUI/
>>
>>
>>
>