See also: IRC log
<trackbot> Date: 14 May 2009
<jeanne> instructions for scribing http://www.w3.org/WAI/UA/scribing.html
<KFord> Scribe: JALLAN
<jeanne> scribe: jeanne
trackbot, start meeting
<trackbot> Meeting: User Agent Accessibility Guidelines Working Group Teleconference
<trackbot> Date: 14 May 2009
<scribe> chair: KFord and AllanJ
KF: We finished discussion of #53 with an action item for KFord that he will have for next week's survey.
KF: Last week we agreed that it needed more work but there are no proposals. Would anyone take an action item to draft a proposal.
<AllanJ> ACTION: Jim to clean up Greg#55 (4.1.10 & 11) make proposal [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action01]
<trackbot> Created ACTION-183 - Clean up Greg#55 (4.1.10 & 11) make proposal [on Jim Allan - due 2009-05-21].
KP: For speech input users, it is important to have a central listing of all keyboard commands.
JS: We have a requirement for a central listing of all keyboard commands.
KF: Do we think an option to surface the keyboard abilities is a level 2.
<AllanJ> +1 to KS have level 2 option to reveal keys
<Henny> + 1
GL: It would be useful to
separate out the keyboard shortcuts in the user interface from
the keyboard shortcuts i nthe content.
... I think it is useful to have a level A that shows all the recognized shortcuts in the content, and then a AA requirement for the user interface, because it is a greater cognitive load to expect the user to remember the keyboard shortcuts for every site.
... by splitting it up, we get the advantages of the content, but without the load on the developer
<Zakim> AllanJ, you wanted to say keystroke to reveal, like word ribbon
<scribe> ACTION: Greg to take #63 and turn it into a proposal for next week. [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action02]
<trackbot> Sorry, couldn't find user - Greg
KF: As a group we need to commit to taking the action items and turn it around as a proposal within a week, otherwise we are getting too bogged down and not accomplishing what we need to.
<scribe> ACTION: JS to work with Greg to take #63 and turn it into a proposal for next week. [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action03]
<AllanJ> Proposed: Standard text area keyboard conventions: Views that render text support the standard text area conventions for the platform including, but not necessarily limited to: character keys, backspace/delete, insert, "arrow" key navigation (e.g., "caret" browsing), cut, copy, paste, delete, select all, undo, navigate to start/end, navigate by paragraph, shift-to-select mechanism, etc. (Level A)
<AllanJ> +1 to proposed wording
<Greg> Re Simon's propsed wording of 4.1.6, to make the title make sense out of context, I'd modify slightly to "Standard Text Area Keyboard Conventions".
KP: I had comments that I put in about standard wordings, and I would like to expand the examples.
<KFord> Kim's answers/comments are at http://www.w3.org/2002/09/wbs/36791/20090513/results
JA: 4.1.6 applies to text area boxes
KF: We have a guideline that tells the agent that the user gets to specify the precedence of the processing order of the keys.
KP: Only Undo seems to apply.
GL: I recommend that we don't try
to make it an exhaustive list of examples, but rather give an
example of categories.
... a lot of platforms aren't going to have standards for these.
JA: but that is covered by "avavilable on the platform"
GL: because it is a long list, it appears to be exhaustive.
<Greg> That's a very good point: if there is keyboard access to static, rendered text, then the UA should support platform keyboard conventions there, too. Not just text input fields.
JS: I would like to standardize our wording, either "such as" or "including, but not limited to,"
KF: we need to get the gestalt of what we want to say and look for the small wording changes later.
<Greg> I'm fine either way on standardizing the wording of "including but not limited to" or "such as".
<AllanJ> Proposed: Standard text area keyboard conventions: Views that render text support the standard text area conventions for the platform including, but not necessarily limited to: character keys, backspace/delete, insert, "arrow" key navigation (e.g., "caret" browsing), cut, copy, paste, delete, select all, undo/redo, navigate to start/end, navigate by paragraph, shift-to-select mechanism,...
<AllanJ> ...etc. (Level A)
GL: Does this only apply to text input or interactive fields, or does it apply to all text fields? Do we need to make it more specific?
<Greg> It seems that there are two basic types of text field: interactive and non-interactive, and interactive can be read-write or read-only.
<Greg> For all interactive fields, the functionality is already required to be available via the keyboard, but we need to make sure it follows standard keyboard conventions.
KF: Interactive text areas also apply to rendered content if you support carat browsing to that text. But it does not apply to the title bar of the user interface.
GL: THere is an example of a web application that is emulating MS WOrd, and using the Word keyboard conventions, and is not supporting the user platform standards.
JA: the author can make the user
interface do whatever it wants, but the browser is not aware of
what the author has remapped how the keys are supposed to
... if the author alters that, the user agent can't be responsible for what the author has done.
<Greg> However...what about content like a text editor that intentionally provides its own keyboard UI. For example, EmacSpeak is UA that doesn't support platform standard keyboard UI, does it fail level A?
GL: Use EMAC-speak as an example. It uses a different keyboard shortcuts than Mac or PC or Gnome.
KF: THis is only referring to editable text fields.
<AllanJ> scribe: allanj
<Greg> It is not currently written so as to apply only to editable text fields.
KF: good understanding of what
... can we defer and say technical note will clarify
... concerned we are going to deep
JS: change views thta render text support the rendered editable text areas
views that render text, support the editable text area conventions
KF: If you have a text box, do what is expected to support as keyboard accessible.
<Greg> Perhaps "rendered text that can be selected or modified...follow platform keyboard conventions" such as editing, navigation, copy and paste, etc."
GL: Emacs speak uses different UI form the platforms on which it works. does it fail this SC
GL: at least put in ability to not use standard keyboard
<Greg> Need to at least provide escape clause for cases where the user wants to use keyboard UI that differs from the platform standard.
<Greg> E.g. VIM using Ctrl+V to start selection vs. Windows using Ctrl+V for paste.
<scribe> ACTION: Greg to rework 4.1.6 [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action04]
<trackbot> Sorry, couldn't find user - Greg
<jeanne> ACTION: JS to assist GL to write a proposal for 4.1.6 to address concerns of this survey [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action05]
<trackbot> Created ACTION-184 - Assist GL to write a proposal for 4.1.6 to address concerns of this survey [on Jeanne Spellman - due 2009-05-21].
JA: 4.1.6 is supposed to apply to
only editable areas in content - iniput, textarea, etc.
... does not apply to entire viewport
JS: one way to address...platform or platform emulation
<Greg> On Jim's 4.1.8 wording, I'd remove "unmodified keystrokes" to "a single or sequence of keystrokes or key combinations"
proposed: 4.1.8 Important Command
Functions: Important command functions (e.g. related to
navigation, display, content, information management, etc.) are
available using a single keystroke, key combination, or a
single or sequence of keystrokes or key combinations. (Level
... 4.1.8 Important Command Functions: Important command functions (e.g. related to navigation, display, content, information management, etc.) are available using a single or sequence of keystrokes or key combinations. (Level AA)
<Greg> That looks good to me (except of course "Important" is vague, but that's another issue.)
Resolved: 4.1.8 Important Command Functions: Important command functions (e.g. related to navigation, display, content, information management, etc.) are available using a single or sequence of keystrokes or key combinations. (Level AA)
<Greg> Are we going to drop the issue of key sequences being memorizable?
<jeanne> ACTION: JS to update the draft with the updated text for 4.1.8: 4.1.8 Important Command Functions: Important command functions (e.g. related to navigation, display, content, information management, etc.) are available using a single or sequence of keystrokes or key combinations. (Level AA) [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action06]
<trackbot> Created ACTION-185 - Update the draft with the updated text for 4.1.8: 4.1.8 Important Command Functions: Important command functions (e.g. related to navigation, display, content, information management, etc.) are available using a single or sequence of keystrokes or key combinations. (Level AA) [on Jeanne Spellman - due 2009-05-21].
GL: memorized, user can enter
without having to review or read to see the next command,
... huge cognitive issue
... may not belong in 4.1.8, but we don't want it to be unusable
... select files from a list, rather than enter whole file.
... need to find wording. If take away wording specifics. Do we need something to address - efficiency.
KP: applies to speech as well
Issue: keyboard efficiency, for screenreaders and speech users. cognitive load from GL comments (micorsoft wordding)
<trackbot> Created ISSUE-36 - Keyboard efficiency, for screenreaders and speech users. cognitive load from GL comments (micorsoft wordding) ; please complete additional details at http://www.w3.org/WAI/UA/tracker/issues/36/edit .
<mhakkinen> lurking comment: use of mnemonics where possible, taking into consideration localization issues
The functions provided to operate a user interface using only *keyboard
commands*, without any need for pointer actions. Examples include:
sequential keyboard navigation through a GUI, keyboard shortcuts, and
command line interfaces.
The set of signals that a user interface will accept from a keyboard or
keyboard emulator in a given context (e.g., with focus in a document vs.
with focus in the menus). Signals may be composed of one keyboard event
(e.g., the "Tab" key") or multiple keyboard events that occur either
simultaneously (e.g., "ctrl"+"S") or sequentially (e.g. "alt","F","S").
For the purposes of UAAG 2.0, several types of keyboard commands are
(a) *Sequential Commands* are those that are not tied to any particular
UI controls or application functions, but rather support traversal of
sets of controls (e.g., repeating "Tab" to move between all active
controls, "arrow" keys to move focus through items in a list).
Sequential commands help users explore what is available.
(b) *Direct Commands* (also called "keyboard shortcuts" or "accelerator
keys") are those tied to particular UI controls or application
functions, allowing the user to navigate-to or activate them without
traversing any intervening controls (e.g., "ctrl"+"S" to save a
document). It is sometimes useful to distinguish direct commands that
are associated with controls that are rendered in the current context
(e.g., "alt"+"D" to move focus to the address bar) from those that may
be able to activate program functionality that is not associated with
any currently rendered controls (e.g., "F1" to open the Help system).
Direct commands help users accelerate their selections.
(c) *Spatial Commands* are those in which the keyboard is used to
control the position of controls in space (e.g., using the arrow keys to
move a mouse pointer by set numbers of pixels). ACCESSIBILITY NOTE:
Spatial commands do not typically enhance exploration or acceleration of
selection of selection and should not be considered an alternative to
direct or sequential commands.
GL: def of Direct commands looks
... replace in document "direct keyboard command' with 'direct commands'
JS: important to keep 'keyboard'
KF: proposal is to use this as definition for 'keyboard command"
<Greg> I recommend tha t the glossary entry explicitly include any synonyms used in the document (e.g. include "Direct KEYBOARD Command" along with "Keyboard Commands" so that a user doing a text search for the string used in the guidelines will find the glossary entry.
all: discussion of glossary structure
JS: have a protocol for covering this - definitions, synonyms, etc.
Resolved: use Keyboard Commands defintion at http://lists.w3.org/Archives/Public/w3c-wai-ua/2008AprJun/0004.html
<jeanne> ACTION: JS to update the glossary entry for Keyboard commands to include the synonyms. [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action07]
<trackbot> Created ACTION-186 - Update the glossary entry for Keyboard commands to include the synonyms. [on Jeanne Spellman - due 2009-05-21].
HS: question about definitions
and incorporating new defintions
... still some missing. see http://lists.w3.org/Archives/Public/w3c-wai-ua/2009AprJun/0052.html
KF: reviews survey comments
GL: this is not a Level A priority in general software guidelines
KF: as I read 4.1.9, does not allow rebinding of platform keybindings (alt, etc.)
<Greg> Sounds like there are two issues: (1) should this be level A, and (2) confusion about which keyboard UI the UA is required to let the user redefine, and what if any it prohibits the user from redefining?
KF: what to we want UA to do with this requirement.
GL: UA, minimally accessible. ??? thinking...
KF: in IE* add hot key for
private browsing - ctrl shhift p
... but if conflict, user should be alble to change that key
<Greg> Why allow someone to remap keyboard commands? (a) make it something easier for them to enter, (b) make it something that doesn't conflict with assistive technology.
KF: but not allow user to change
cntl+o which opens page.
... if UA wants ... fine
KP: speech user, trouble speaking, certain key strokes may be difficult
KF: should the UA allow more
accessibility that the platform on which it is running
... are we asking too much?
<Greg> I feel it shouldn't be level A because (a) most software doesn't allow this, thus pretty much all UA would fail to meet Level A (b) because most assistive technology allows (or should allow) their commands to be remapped, (c) ISO 9241-141/ANSI 200.2 make this lower priority than the highest priority, and I don't think it makes sense for the requirements for UA to be higher than general...
<Greg> ...software in this regard.
KP: having the option is good. not sure if it is too much.
SH: problem conflating UA and platform accessibility.
GL: rationale, make it easier to use, remove conflictions.
KF: 2 questions - 1. which
... 2. override ALL, or not allowed to override 'platform
... perhaps needs to be rewritten.
... lower priority, leave workding
agree: SH, HS, JA, JS, KF
GL: concern: "except" - not prohibiting the overriding of Ctrl+o
<KimPatch> agree as well
<scribe> ACTION: KF to minor edit 4.1.9 "does not prohibit' overriding of platform keys [recorded in http://www.w3.org/2009/05/14-ua-minutes.html#action08]
<trackbot> Created ACTION-187 - Minor edit 4.1.9 "does not prohibit' overriding of platform keys [on Kelly Ford - due 2009-05-21].
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found Scribe: JALLAN Found Scribe: jeanne Inferring ScribeNick: jeanne Found Scribe: allanj Inferring ScribeNick: AllanJ Scribes: JALLAN, jeanne, allanj ScribeNicks: jeanne, AllanJ Default Present: KFord, Jeanne, Greg_Lowney, Allanj, Henny, Kim_Patch, sharper Present: KFord Jeanne Greg_Lowney Allanj Henny Kim_Patch sharper Regrets: Jan MH Mark_Hakkinen Jan_Richards Found Date: 14 May 2009 Guessing minutes URL: http://www.w3.org/2009/05/14-ua-minutes.html People with action items: greg jim js kf[End of scribe.perl diagnostic output]