Minutes, ARIA Mapping Subteam, 11 May

Minutes from today's teleconference of the ARIA Mapping Subteam are
available as text below, and as HTML at:
http://www.w3.org/2010/05/11-aria-minutes.html

   W3C

                                                           - DRAFT -

                                               ARIA mapping group - HTML A11Y TF

11 May 2010

   Agenda

   See also: IRC log

Attendees

   Present
          Janina, Rich, Steve_Faulkner, Michael_Cooper, Cynthia_Shelly

   Regrets
   Chair
          Steve_Faulkner

   Scribe
          janina

Contents

     * Topics
     * Summary of Action Items
     __________________________________________________________________________________________________________________

   <MichaelC> Last week's minutes

   <MichaelC> ARIA mapping issues and actions

   <cshelly> url for actions?

   <MichaelC> action-39: duplicate of action-38

   <MichaelC> close action-39

   <MichaelC> action-2: superseded by the creation of the ARIA mapping sub-group

   <MichaelC> close action-2

   <MichaelC> scribe: janina

   <scribe> scribe: janina

   general discussion trying to achieve clarity on our actions

   http://www.w3.org/2010/05/04-aria-minutes

   sf: the current (modified) mapping doc is roughly this
   ... understand we need a rationale

   cs: let's look at the role categories -- where's the table?

   <cshelly> http://www.w3.org/WAI/PF/aria/roles#roles_categorization

   cs: looking at this link is descendent of widget, should it be overwritable by any widget?

   sf: somewhat confused ... command is also subclass of widget

   <davidb> regrets, I am IRC only for now

   cs: pieces of a slider often made of links, but not the slider itself ...
   ... is it inputs that are not also composites?

   sf: spinbutton is subclass of range ... trying to follow the text because can't follow the chart

   cs: buttons, list items, and menu items, and tree items most link users
   ... does turning a link into a checkbox make sense?

   sf: seen it being done
   ... looking for an example
   ... aren't we just going to define by mapping

   cs: so it's buttons and options?

   rs: anything activatable -- so not grid

   mc: i updated taxonomy last week, some where hard to graph

   <cshelly> http://www.w3.org/WAI/PF/aria/rdf_model

   <richardschwerdtfe> http://www.w3.org/WAI/PF/aria/rdf_model.svg

   cs: link is a command--so we can say all commands can replace each other

   agreement

   cs: so, links can be turned into commands and ranges

   agreement

   sf: thumb displaying the range isn't marked with a role, is it?
   ... usually slider is a container element

   cs: it's clearer that commands are the same
   ... range does seem very different
   ... can't say i've seen a lot of this

   rs: not difficult

   cs: can menu items?

   rs: suppose

   cs: doesn't seem to make sense

   rs: perhaps a spinbutton

   cs: but ht5 already has

   rs: but as a range item
   ... could have a menu item with a slider in it

   <cshelly> Commands can all replace each other

   <cshelly> buttons and links are special cased. they can be replaced by ranges which are not composites

   <cshelly> commands=link, button, menuitem, menuitemcheckbox, treeitem, menuitemradio

   above from cs appears to be group consensus

   further checking via the diagram

   mc: explaining uml convention where arrow points to parent

   <cshelly> ranges that are not composites = progressbar, slider, scrollbar

   mc: another internal convention is that all roles are in rows
   ... anything below is descendent

   cs: option is a command in ht5,

   mc: but not here ...

   cs: suggest: any role can be overwritten by a descendent

   mc: should one of our best design patterns

   cs: a bp or this doc

   mc: both

   cs: suggest: div and span can be overwritten by anything

   sf: take that for granted

   janina: never take rocks for granite

   general agreement

   cs; inputs are trickier ...

   cs: options often overwritten by menu item
   ... expect ht5 to do that automagically

   sf: inputs in ref to command?

   cs: looking at taxonomy
   ... not reasonable to say all inputs can overwrite each other

   sf: what about tree item

   cs: not a command
   ... it's a list item
   ... so three decisions -- and if we see this doesn't work, probably a taxonomy error

   sf: sounds good

   cs: so next week i'm due for a draft

   sf: yes, all of us
   ... rich, cyns, and steve

   rs: i'm out next week

   cs: should discuss input or my deliverable
   ... which is best?
   ... maybe inputs, ranges and some coposite for next week

   <cyns> http://www.w3.org/Bugs/Public/show_bug.cgi?id=9326

Summary of Action Items

   [End of minutes]
     __________________________________________________________________________________________________________________

Found Scribe: janina
Present: Janina Rich Steve_Faulkner Michael_Cooper Cynthia_Shelly

-- 

Janina Sajka,	Phone:	+1.443.300.2200
		sip:janina@asterisk.rednote.net

Chair, Open Accessibility	janina@a11y.org	
Linux Foundation		http://a11y.org

Chair, Protocols & Formats
Web Accessibility Initiative	http://www.w3.org/wai/pf
World Wide Web Consortium (W3C)

Received on Tuesday, 11 May 2010 20:13:35 UTC