- From: Alexander Surkov <surkov.alexander@gmail.com>
- Date: Thu, 31 Oct 2013 16:19:05 -0400
- To: Joseph Scheuhammer <clown@alum.mit.edu>
- Cc: Richard Schwerdtfeger <schwer@us.ibm.com>, Cynthia Shelly <cyns@microsoft.com>, David Bolter <dbolter@mozilla.com>, "W3C WAI Protocols & Formats" <public-pfwg@w3.org>
Comments inline: On Thu, Oct 31, 2013 at 3:59 PM, Joseph Scheuhammer <clown@alum.mit.edu> wrote: > Comments inline: > > > On 2013-10-31 3:11 PM, Alexander Surkov wrote: > > Hi, Rich. Few notes: > > Row1. Phrase: "Menubar must a DOM parent of the menuitems or one defined by > aria-owns" not sound are part of specific scenario. It relates to all > scenarios so probably it makes sense to move it out of table. > > > Yes and no. There are situations that involve menus without a menubar. Two > examples are a menu that is invoked from a toolbar button, and a context > menu that is triggered from "white space". > - Rows 1, 2, and 5 describe a scenario involving a menubar. > - Toolbar button and context menu popup/dismiss are handled by rows 3 and 4. > - Rows 3 and 4 also handle a menu associated with a menubar. > - Row 2 currently only speaks about focus on a menuitem in a menubar, but > the same occurs when focus is on a on a menuitem within a menu. The text > could be modified to reflect both scenarios. > > That said, the idea of moving the statement regarding the parent/child > relationship with menubar/menuitems is okay with me. Here is the suggested > text: > > "Frequently, a menubar is used organize a hierarchy of menus. In those > cases, the menubar MUST be a DOM parent of the associated menuitems, or one > defined by aria-owns. In other cases, no menubar is involved as the menu > may be associated with a toolbar button, or may be a context menu triggered > from white space. Nonetheless relevant menu events are fired as described in > the following table". Good > Row3. "Should only be fired once until the menu is closed and opened again" > sounds excess because a menu can't be open twice without closing. > > > Sigh. Sometimes readers complain that the spec. language isn't complete and > leaves too much for the reader to decide what it's saying. Other readers > complain that the language is "too excessive". IMO, if the excess > reinforces the point, I'm for keeping it as leaves less room for > (mis)intrepretation. > > > Row4. "EVENT_SYSTEM_MENUPOPUPEND once only for accessible menu object and > only if EVENT_SYSTEM_MENUPOPUPSTART was fired for it. Second part sounds > unnecessary because menupopup can't be hidden without being opening. > > > See my previous comment. I'd agree if I had an example how these words can be read ambiguously. Anyway, if theoretically I've got an open menu that never received a menupopup_start event then why would AT care it must not receive menupopup_end on its close? > Row5. "All menus closed, and user moves focus away from menubar; menubar is > deactivated." What is propose of "all menus closed"? What if menus were > never be open? > > > Maybe that's why there was both a row 5 and row 6 in the previous version? > There are two possibilities: > 1. User navigates to a menubar, navigates among its menuitems never opening > a menu, and finally navigates away from the menubar. > 2. User navigates to a menubar, navigates among its menuitems, opens at > least one menu, and possibly more than one, and finally navigates away from > the menubar. > > I believe the events are the same for these two scenarios in terms of what > events are fired when the user finally navigates away from the menubar. I > could modify the text describing the scenario such that is covers these two > cases. I think the key point here is menubar is deactivated, i.e. focus moves out of menubar tree and it doesn't really matter how the user achieved that. Having said that I think it might be good to cite those two examples as clarification. > > > Thank you. > Alex. > > > Thank you. > > > -- > ;;;;joseph. > > > 'A: After all, it isn't rocket science.' > 'K: Right. It's merely computer science.' > - J. D. Klaun -
Received on Thursday, 31 October 2013 20:19:32 UTC