16633 2012-04-04 10:10:35 +0000 [AAPI]: "aside" mapping firefox 2013-03-24 00:12:06 +0000 1 1 1 Unclassified HTML WG HTML a11y APIs (editor: Steve Faulkner, Cynthia Shelly) unspecified PC Windows NT RESOLVED FIXED http://lists.w3.org/Archives/Public/wai-xtech/2012Apr/0007.html a11y P2 normal --- 1 faulkner.steve jason hans.hillen jason mike public-html-admin public-html-wg-issue-tracking surkov.alexander sideshowbarker+html-a11y-api oldest_to_newest 66416 0 faulkner.steve 2012-04-04 10:10:35 +0000 alex wrote: aside. Firefox exposes VARIANT aside role on MSAA and IA2_ROLE_NOTE on IA2, ATK_ROLE_SECTION on ATK. 68603 1 jason 2012-06-04 22:55:40 +0000 MSAA roles are an enumerated list. Firefox's implementation using VARIANT is unique. For IA2, IA2_ROLE_NOTE is defined as "An embedded note which is not visible until activated."[1] Is this appropriate for aside? Why not IA2_ROLE_SECTION? [1] http://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_role_8idl.html 68612 2 surkov.alexander 2012-06-05 03:00:11 +0000 Good point. The same we do for ARIA role="note" which is defined as: A section whose content is parenthetic or ancillary to the main content of the resource. http://www.w3.org/WAI/PF/aria-implementation/#mapping_role_table http://www.w3.org/WAI/PF/aria/roles#note So we'd need to get feedback from ARIA guys 69702 3 jason 2012-07-08 04:11:48 +0000 MSAA mapping for <aside> updated to indicate Firefox's use of VARIANT to pass string value "aside" for accRole. Proposed IA2 mapping for <aside> changed from IA2_ROLE_NOTE to IA2_ROLE_SECTION, which is more appropriate and consistent with ATK/AT-SPI mapping. Related bug filed against mapping of ARIA role "note" to IA2_ROLE_NOTE. See Bug 17716 - Mapping role "note" to IA2_ROLE_NOTE doesn't seem appropriate [1] [1] https://www.w3.org/Bugs/Public/show_bug.cgi?id=17716 73161 4 surkov.alexander 2012-09-03 07:36:36 +0000 Should this bug be reopen since ARIA's one was reopen: https://www.w3.org/Bugs/Public/show_bug.cgi?id=17716. 73162 5 jason 2012-09-03 08:45:02 +0000 Yes, probably. I find the current situation with the mappings for the aside element a little confusing, and think it needs some clearing up: 1. the current default implied role for aside is note [1] 2. the definition of aside [2] seems more closely aligned to the definition of the complementary role [3] than it does to the definition of the note role [4], particularly where the aside still being meaningful when separated from the surrounding content is concerned. 3. the HTML to Platform Accessibility API guide associates aside with the complementary role, although this may be have been done by accident given current browser mappings of the aside element [5] 4. Firefox and Safari both map the aside element to the complementary role, Firefox using an object attribute and Safari a new AXSubrole of AXLandmarkComplementary. 5. The latest IA2 spec has a new constant IA2_ROLE_COMPLEMENTARY_CONTENT that reflects the definitions of the aside element and complementary role [6] 6. The latest IA2 spec has not updated its definition of the IA2_ROLE_NOTE constant, which still refers to "An embedded note which is not visible until activated" [7], which is not appropropriate for the aside element, or for the note role for that matter. Given that the browsers that natively map the aside element to an accessible role are using the complementary role, and not the note role, shouldn't the IA2 mapping for aside be IA2_ROLE_COMPLEMENTARY_CONTENT? Also, although this would need to be filed with the ARIA/HTML folk, shouldn't the default implied role for aside be complementary and not note? [1] http://dev.w3.org/html5/spec/wai-aria.html#sec-implicit-aria-semantics [2] http://dev.w3.org/html5/spec/the-aside-element.html#the-aside-element [3] http://www.w3.org/WAI/PF/aria/roles#complementary [4] http://www.w3.org/WAI/PF/aria/roles#note [5] http://dvcs.w3.org/hg/html-api-map/raw-file/tip/Overview.html#el-8 [6] http://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_role_8idl.html#e37ff81431ee3762a5d41a2cb909108d7723b5d0895f8a53121b5282b15bd502 [7] http://accessibility.linuxfoundation.org/a11yspecs/ia2/docs/html/_accessible_role_8idl.html#e37ff81431ee3762a5d41a2cb909108daec490b192bd57535e3310282bb7ba01 73168 6 surkov.alexander 2012-09-03 14:53:16 +0000 You're right. HTML aside should use IA2 complimentary role. That was decision after long discussion at https://lists.linux-foundation.org/pipermail/accessibility-ia2/2010-November/001233.html. 73169 7 surkov.alexander 2012-09-03 15:04:42 +0000 (In reply to comment #6) > You're right. HTML aside should use IA2 complimentary role. That was decision > after long discussion at > https://lists.linux-foundation.org/pipermail/accessibility-ia2/2010-November/001233.html. btw, we have a bug in Firefox about this issue: https://bugzilla.mozilla.org/show_bug.cgi?id=614572 84891 8 jason 2013-03-24 00:12:06 +0000 Closing as <aside> now mapped [1] to IA2_ROLE_COMPLEMENTARY_CONTENT, which is consistent with recent change [2] to default implicit ARIA semantic of complementary role for <aside> [3]. [1] https://dvcs.w3.org/hg/html-api-map/raw-file/tip/Overview.html#el-8 [2] https://www.w3.org/Bugs/Public/show_bug.cgi?id=19145 [3] http://www.w3.org/html/wg/drafts/html/master/dom.html#sec-implicit-aria-semantics