<jimal> https://www.w3.org/2002/09/wbs/35422/Oct_17th_agwg/

defending 4.5:1 https://www.w3.org/2002/09/wbs/35422/Oct_17th_agwg/

<laura> Results:

<laura> Survey https://www.w3.org/2002/09/wbs/35422/Oct_17th_agwg/results#xnew3

<jimal> techniques - outline-offset

<JohnRochford> Jim: We have to discuss defending 4.5 to 1.

<jimal> push the outline-offset 3 px away from the object with focus, then you measure the contrast only between the outline and the background color

<JohnRochford> Jim: That should not be that difficult to do.

<JohnRochford> Jim: I wonder if the outline contrast was one of the two, could you really even see it?

<JohnRochford> Steve: This is just one technique, and there are so many others, for 3-way contrast.

<JohnRochford> Steve: If we are talking about the focus ring around something, you could change the color of the button on focus.

<JohnRochford> Steve: It gets confusing in the SC.

<JohnRochford> Steve: Yes, Jim, if you had a 3-pixel outline in a black color, it would be difficult to discriminate from a black button.

<jimal> dashes are easier to see than dotted

<JohnRochford> Jim: Dashes are easier to discern because there is more space between dashes (than there is between dots).

<JohnRochford> Jim: I don't want this to become a CSS thing where users are expected to do this themselves.

<JohnRochford> Glenda: Testing for 3 CSS pixels is not simple.

<JohnRochford> Glenda: I suggest we go forward with 3:1, but reserve the right to go with 4.5:1.

<JohnRochford> Glenda: The difficult with pixels is that they can't be measured on new monitors.

<JohnRochford> Glenda: We would have to come up with a measuring tool.

<JohnRochford> Glenda: Developers would have to measure all kinds of things with such a tool.

<JohnRochford> Steve: If it's something that could be done with CSS or another technology, I could see how 3 pixels would be easy.

<JohnRochford> Steve: Otherwise, it would be near-impossible to do with any accuracy.

<JohnRochford> Jim: I use a TPG tool, an eye dropper, for measurement.

<laura> Colour Contrast Analyser: https://developer.paciellogroup.com/resources/contrastanalyser/

<JohnRochford> Glenda: It already takes my employees too much time to test a page for WCAG 2.0. If we had to use an eye dropper too, the burden would be too much.

<JohnRochford> Glenda: One difference between 4.5:1 and 3:1 is that the color pallete is much larger in 4.5:1.

<JohnRochford> Glenda: When you are reading a page, and are glancing across words, the difference between an I and a T is very important. When you are looking through a focus indicator, you are not speed reading.

<JohnRochford> Glenda: There's a higher bar for color contrast.

<JohnRochford> Glenda: We get research between now and Silver that explores 4.5:1. My instinct says that the cost of implementing 4:5.1 is too great a burden for the effect it would have for people with disabilities.

<JohnRochford> Laura: If we pull back on this one, will we have to pull back on color contrast too?

<JohnRochford> Glenda: I think we would have to pull them back in unison.

<scott> We need good tools to test

<scott> manual testing won’t scale

<JohnRochford> Glenda: If it proves difficult to implement, we could discuss at the next call.

<JohnRochford> Scott: Any manual testing is not going to scale. Tools could be a solution, but they don't exist yet.

<JohnRochford> Scott: It's very difficult for me to see mouse pointers and focus indicators.

<jimal> scott - has difficulty finding focus indicators, if you make them move like marching ants,

<JohnRochford> Scott: The size/thickness of the focus indicators does help. Color can make a difference. Noticability from a distance is important.

<JohnRochford> Scott: If you are using a border with an animation, that would be more noticable than a thicker border.

<jimal> August Colenbrander - What is adequate under normal circumstances, may not be adequate under stress.  **Low vision is a form of permanent stress.**  

<JohnRochford> Scott: If there are black borders, they may appear as light-blue on LED montiors. That is difficult to see unless you get close to the monitor.

<JohnRochford> Scott: How do you test for that?

<Zakim> steverep, you wanted to say focus indication has many more aspects than "identification"

<jimal> steve - no proof of text being different from text

<JohnRochford> Steve: I don't buy that it's harder to read text in graphics until I see usability studies.

<Glenda> no proof of text being different that non-text (need for contrast)

<JohnRochford> Steve: Identifying is different from focus indication.

<jimal> +1 to form controls are very different from focus indicator

<Glenda> Hey Jim, have we filed bugs against the browsers for poor color contrast (by default) on focus indicators?

<jimal> different tools available for changing focus indicators

<JohnRochford> Steve: Finding a focus indicator, and finding it again, is a significant issue.

<jimal> I believe so. I filed bugs about form controls. don't recall focus. will check

<Glenda> but but but…there is no requirement for a text box to have a “box” at all

<JohnRochford> Steve: I don't want to drop 3:1 for things like text being harder to read.

<JohnRochford> Steve: Selection state is much more what I am worried about.

<JohnRochford> Glenda: I am uncomfortable throwing focus indication under the bus.

<JohnRochford> uncomfortable with

<JohnRochford> Scott: I'm trying to draw a distinction, not throw focus indication under the bus.

<JohnRochford> Glenda: I think that, for keyboard-only users, focus indication is very important.

<JohnRochford> Scott: It all gets back to what you are trying to identify: state, contrast, focus.

<JohnRochford> Glenda: If we are struggling with seeing it, and even knowing what to look for, people with cognitive disabilities would have even more trouble.

<JohnRochford> Glenda: The COGA TF is working on a related AAA SC.

<scott> I am not surprised

<jimal> User agent control: The color(s) of the user interface component and any adjacent color(s) are determined by the user agent and are not modified by the author.

<jimal> if the author never touches form controls or focus they are off the hook. only if they do a custom border on the form or focus do they have to make it right

If you touch it, then you need to do it right. If you don't, then it's a browser issue and you're off the hook.

(and that will be addressed through Silver)

<scribe> scribe: Shawn

<scott> you break it you buy it :)

Jim: Have techniques for how to deal with focus. Have if you don't touch it, no foul. Issue is bear of testing.
... Put up to larger community and see how much pushback. Although seeming want to get it perfect now (despite enemy of good).
... Ducks in row for now. Take a break before the next call.

<scott> bye

<laura> bye.

<jimal> join #ag

trackbot, end meeting

This is scribe.perl Revision: 1.152  of Date: 2017/02/06 11:04:15  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/Jim, please let me know when I should resume scribing.//
Succeeded: s/for things like testing/for things like text being harder to read/
Succeeded: s/present=//
Succeeded: s/zakim. clear agenda//
Succeeded: s/Results://
Default Present: Jim, Scott, Shawn, SteveRep, JohnR, Laura, Glenda, JohnRochford
Present: Jim Scott Shawn SteveRep Laura Glenda JohnRochford
Found Scribe: Shawn
Inferring ScribeNick: shawn
Found Date: 19 Oct 2017
Guessing minutes URL: http://www.w3.org/2017/10/19-lvtf-minutes.html
People with action items: