18:44:45 RRSAgent has joined #aapi 18:44:45 logging to http://www.w3.org/2015/06/30-aapi-irc 18:44:47 RRSAgent, make logs public 18:44:47 Zakim has joined #aapi 18:44:49 Zakim, this will be WAI_PF 18:44:49 ok, trackbot; I see WAI_PFWG(AAPI)3:00PM scheduled to start in 16 minutes 18:44:50 Meeting: Protocols and Formats Working Group Teleconference 18:44:50 Date: 30 June 2015 18:45:03 chair: Joseph_Scheuhammer 18:45:14 agenda: this 18:45:24 agenda+ Bugzilla-28816 (All) Some ARIA roles lacking explicit a11y API role mappings. 18:45:29 agenda+ ACTION-1541: (James) AXAPI mapping for aria-modal. 18:45:35 agenda+ ACTION-1535: (James) AXAPI mapping for aria-current. 18:45:42 agenda+ ACTION-1608: (Cynthia) Bring up issue-708 with html-a11y taskforce (conditions for inclusion within a11y tree). 18:45:52 agenda+ ACTION-1660/ISSUE-714 (Joanie) ATK/AT-SPI mapping of aria-roledescription. 18:45:57 agenda+ ACTION-1661/ISSUE-714 (Cynthia) UIA mapping of aria-roledescription (UIA Express?) 18:46:03 agenda+ ACTION-1662/ISSUE-714 (Alex) IA2 mapping of aria-roledescription. 18:46:12 agenda+ ACTION-1622: (Alex) MSAA+IA2 mapping of aria-placeholder. 18:46:18 agenda+ Action/issue triage (https://www.w3.org/WAI/PF/Group/track/products/23) 18:46:25 agenda+ be done. 18:46:35 I have made the request to generate http://www.w3.org/2015/06/30-aapi-minutes.html clown 18:55:51 present+ Joseph_Scheuhammer 18:58:18 present+ Joanmarie_Diggs 19:13:58 issue=-708? 19:14:04 issue-708? 19:14:04 issue-708 -- The phrase "If not already excluded from the accessibility tree per the rules above" implies that the listed conditions do not override. Is that correct? -- open 19:14:04 https://www.w3.org/WAI/PF/Group/track/issues/708 19:20:53 richardschwerdtfeger has joined #aapi 19:23:34 scribe: joanie 19:23:39 Zakim, take up item 1 19:23:39 agendum 1. "Bugzilla-28816 (All) Some ARIA roles lacking explicit a11y API role mappings." taken up [from clown] 19:23:48 https://www.w3.org/Bugs/Public/show_bug.cgi?id=28816 19:23:53 JS: This is a bug that Jason Kiss raised a while ago. 19:24:26 JS: Summary: He's noting that certain roles (banner, navigation, complementary, contentinfo, article and form) don't have any role information in the core mapping spec. 19:24:43 JS: I replied and said that information is in the xml-roles object attribute. 19:24:53 JS: To which he said we should give them a real role. 19:25:08 JS: Then I thought, how does an AT know to look at the object attribute? 19:25:19 RS: They look at role="region". 19:25:42 RS: Landmark is abstract, so what you'd need to do is use role="region". Then they'd look for more information. 19:25:55 JS: I did have a conversation with Joanie about this. 19:27:05 JD: Orca only looks at xml-roles when it "has to." 19:27:18 http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-banner 19:27:41 JS: It turns out ATK/AT-SPI2 already handles all of these cases. 19:27:51 JS: I put the URL in the room (above) 19:28:40 RS: Looking at banner... What does AT-SPI2? ROLE_LANDMARK. 19:28:55 JS: So Orca sees it's a landmark and then looks at the object attributes to see which one. 19:29:17 JS: For UIA Cynthia said to expose role grouping with a localized role name. 19:29:42 JS: I predict that in IA2, the role you get is based on the tag you use. 19:30:24 RS: We could add a ROLE_LANDMARK for IA2. 19:31:20 RS: (Tries test case in Firefox) 19:31:58 s/localized role name/localized control type matching the AriaRole/ 19:37:06 BG: What if you put a role="banner" on an iframe? 19:37:17 (Group ponders) 19:37:25 (Scribe shudders) 19:38:32 RS: AccRole is banner. 19:38:58 JS: In MSAA there is an AccRole field for accessible role. It's a string. 19:39:17 JS: According to the MSAA documentation, only known MSAA roles should go in that value. 19:39:34 JS: But Gecko puts "banner" as the value for AccRole. 19:39:57 RS: The accessible role is section for an actual section. For div it's also a section. For paragraph, it's a role of paragraph. 19:40:07 JS: What is the IA2 role? 19:40:21 RS: The AccRole in all those cases is banner. 19:40:33 RS: The IA2 roles are as I described earlier. 19:41:14 RS: And the xml-roles object attribute has banner. 19:41:23 RS: Do we want to create an IA2 role of landmark? 19:41:34 JS: Yes, but will it get done in the 1.1 time frame? 19:41:44 RS: It's a constant. 19:42:12 RS: I think it's an enum. Let me look. 19:43:06 RS: We could add a ROLE_LANDMARK at the end. 19:43:09 http://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_role_8idl.html#ae37ff81431ee3762a5d41a2cb909108d 19:43:20 RS: The URL above is the documentation for IA2 roles. 19:44:34 RS: I'm looking at the mappings to see how we can do this consistently. 19:45:11 Cynthia says: "For UIA, these should be mapped to the group control type with a localized control type that matches the aria role." 19:48:18 http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-region 19:49:08 JS: (Reads the mapping of region) 19:49:40 JD: Given that as Bryan and Rich stated, a named region is a landmark, we probably want to update the mapping Joseph just read. 19:49:51 http://w3c.github.io/aria/aria/aria.html#region 19:49:55 https://rawgit.com/w3c/aria/master/aria/aria.html#region 19:50:34 RS: I think there should be an action here to map region to landmark and then get Alex to add a ROLE_LANDMARK to IA2. 19:50:52 RS: Let's give Alex an action to add that to IA2. 19:51:00 JS: But that's not a W3C issue. 19:51:35 JS: That takes care of landmark, but there are other roles that Jason pointed out, like article. 19:51:52 JS: Article and form lack UIA roles. 19:52:15 http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-article 19:52:38 see the UIA entry. All it says is, "Expose as text string in AriaRole". 19:53:41 action: Joseph to update the mapping spec for aria role region to reflect its landmark inheritance as defined in the aria 1.1 spec. 19:53:42 Created ACTION-1667 - Update the mapping spec for aria role region to reflect its landmark inheritance as defined in the aria 1.1 spec. [on Joseph Scheuhammer - due 2015-07-07]. 19:54:07 action-1667? 19:54:07 action-1667 -- Joseph Scheuhammer to Update the mapping spec for aria role region to reflect its landmark inheritance as defined in the aria 1.1 spec. -- due 2015-07-07 -- OPEN 19:54:07 https://www.w3.org/WAI/PF/Group/track/actions/1667 19:54:17 (Group discusses some IE issues) 19:54:50 JS: I've given myself an action to do something about role region with respect to its landmark status. 19:56:01 Action: Action Add a IA2_ROLE_LANDMARK to IA2. This will tell ATS to look at the xml-roles value for landmarks 19:56:01 Error finding 'Action'. You can review and register nicknames at . 19:56:26 Action: Alex Add a IA2_ROLE_LANDMARK to IA2. This will tell ATS to look at the xml-roles value for landmarks 19:56:26 Error finding 'Alex'. You can review and register nicknames at . 19:56:38 Action: Surkov Add a IA2_ROLE_LANDMARK to IA2. This will tell ATS to look at the xml-roles value for landmarks 19:56:38 Created ACTION-1668 - Add a ia2_role_landmark to ia2. this will tell ats to look at the xml-roles value for landmarks [on Alexander Surkov - due 2015-07-07]. 19:56:58 action-1668? 19:56:58 action-1668 -- Alexander Surkov to Add a ia2_role_landmark to ia2. this will tell ats to look at the xml-roles value for landmarks -- due 2015-07-07 -- OPEN 19:56:58 https://www.w3.org/WAI/PF/Group/track/actions/1668 19:57:39 http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-article 19:57:44 JS: The issue about article and form are just UIA questions. 19:57:58 http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-form 19:58:02 JS: Links above are for the mappings. 19:58:25 JS: How do you know it's a form? How do you know to go look at the string? 20:02:34 present+ Rich_Schwerdtfeger, Bryan_Garaventa 20:03:01 regrets: Cynthia_Shelly 20:03:07 Zakim, part 20:03:07 Zakim has left #aapi 20:03:13 RRSAgent, stop