W3C

- DRAFT -

HTML techniques TF

22 Oct 2012

See also: IRC log

Attendees

Present
Regrets
Chair
Josh
Scribe
Josh

Contents


<Joshue108> scribe: Josh

https://www.w3.org/2002/09/wbs/35422/22_10_2012/

<Joshue108> <discussion on Jons technique>

<Joshue108> LGR: This could be another tech suitable for 1.3.1

<Joshue108> Jon: Do you want more detail? Event handlers etc?

<Joshue108> Loretta: We should be telling people to look at what they have implemented, and what the tech is describing.

<Joshue108> LGR: The functional tests can be problematic in use.

<Joshue108> JG: So you're more interested in how you add keyboard support to any kind of widget.

<Joshue108> LGR: Thats what I am wondering.

<Joshue108> LGR: You can add tabindex="0", you may or may not use activedescendent etc.

<Joshue108> JG: Do you really want to promote people doing tabindex="0" on everything?

<Joshue108> LGR: We want to indicate best practice. In some cases, this may be perfectly fine.

<Joshue108> JG: So something about adding event handlers to partciular widgets.

<Joshue108> JOC: We should have something somehwere.

<Joshue108> JOC: We should have examples of where using tabindex is best practice and other where using event handlers or whatever is preferable..

<Joshue108> DF: We do have people who may to expect to reach things by tabbing thru, which is the method in most manu type cases.

<Joshue108> DF: So this tech may be fine for some browser AT combos, but in some cases in can lead to a loss of context etc.

<Joshue108> DF: There should be a disclaimer about use where the user base is known where proper AT and browser combinations are used.

<Joshue108> LGR: A technique would be a good place to have that discussion.

JG: I've seen companies use tab technique and felt that was a good technique - but let to thousands of tab stops
... Other models for keyboard - but not an ARIA technique at that point
... Big part of ARIA / ARIA techs would be to focus on authoring practices - because then when folks get to a menu, they know what behavior to expect
... Otherwise they're left wondering how a widget may act

<Loretta> http://www.w3.org/TR/2012/NOTE-UNDERSTANDING-WCAG20-20120103/keyboard-operation-keyboard-operable.html

DF: I wonder about those folks in install base that don't have latest browser / screenreader. They would be badly served by a widget that don't tell them. Perhaps tell users that will miss out on this that they need to use menu keys etc - how to operate

<Loretta> HTML non-ARIA technique: http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/SCR35

JG: Menu bar and menu navigation are two separate things
... In a menu instead of links you'll have menu items - you have to add event handlers. You have to look at that because something looks like a menu, you may not want to use ARIA to make it a "menu". If it's just a bunch of links there is already a bunch of built in AT support to navigate
... In general I see users overusing ARIA where they shouldn't use it.
... Can often add to overcomplicated designs. Menu is one of those items

DF: Sounds like a clear case to make a differentiation between the techniques

LGR: May be a 1.3.1 technique for menus where description tries to capture what instances you would use this markup for.

<Joshue108> +q

LGR: Issue where people are using where it's not appropriate is 1.3.1 - where people are trying to expose semantics w/ ARIA
... There is an existing technique - pre-ARIA - for making actions keyboard accessible

<Joshue108> Here it is again:

<Joshue108> -q

http://www.w3.org/TR/2012/NOTE-WCAG20-TECHS-20120103/SCR35

JOC: Sounds like Jon has hit on several techniques that we need

LGR: Willing to try to take existing keyboard technique and create a version that is an ARIA/HTML5 technique

<Joshue108> JOC: I'd like to help Loretta on that technique

<jongund> Uses tab or arrow keys for pull down lists of links: http://www.dhs.state.il.us/page.aspx?

<Detlev> sounds good!

JG: ARIA Authoring Practices outlines keyboard navigation - what different keystrokes should do for different widgets

LGR: Would you consider writing up a 1.3.1 technique for menus where you capture your understanding under what circumstances ARIA techniques could be used for menus?

JG: Yes. I think most of the navigation is just listed links, then just landmark navigation role is fine.

<Joshue108> JOC: Thats great, thanks Jon.

JG: I'll try writing it up and sending it. Something about the different types of menus

<Joshue108> http://www.w3.org/TR/WCAG20/#content-structure-separation-programmatic

JG: So it would be a technique on when it's good to use an ARIA menu? When it's good to use one - when not to use one?

LGR: I'm trying to think of the parallel for the technique of when it's appropriate to use Headings

JOC: Traditionally menus over the years have been Lists - but there is a disconnect on what these things may be and how you expect them to act

Agreement to write up draft and then discuss further when we have that draft

LGR: Two techniques for Jon - 4.1.2 using specific menu roles and properties and 1.3.1 when to use menu, menubar roles

ARIA Expandable

<Loretta> https://www.w3.org/2002/09/wbs/35422/htmltechs20121005/results#xtree

JG: Don't believe ARIA has the concept of a static tree

LGR: I pulled that example out of the ARIA spec

JG: To me it's just a nested list - not sure why we should put ARIA on it. I'll ask Rich S what the use case would be

LGR: Certainly not interactive in the sense of expand or collapse

<jongund> I need to go

JOC: Will look at this expanded state in a tree widget technique next week

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.137 (CVS log)
$Date: 2012/10/22 16:59:04 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: marcjohlic
Found Scribe: Josh

WARNING: No "Present: ... " found!
Possibly Present: DF Detlev IPcaller JG JOC Jon Jon_Gunderson Joshue Joshue108 LGR Loretta Marc_Johlic P0 aaaa https inserted jongund
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Got date from IRC log name: 22 Oct 2012
Guessing minutes URL: http://www.w3.org/2012/10/22-html-techs-tf-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


[End of scribe.perl diagnostic output]