Accessibility Guidelines Working Group Teleconference

01 May 2018


Laura, gowerm, alastairc, kirkwood, Greg_Lowney, Glenda, marcjohlic, AWK, Brooks, JakeAbma, Joshue108, Makoto, Chuck, Bruce, Kim_D, Kathy
Brooks, gowerm


<AWK_> zakim clear agenda

<Brooks> scribe: Brooks

<alastairc> zakaim, take up item

<alastairc> zakaim, take up item 1

WCAG 2.2 and Silver Survey: https://www.w3.org/2002/09/wbs/35422/wcag_next/results

<Zakim> AWK_, you wanted to speak to the current charter

awk: Current charter allows for 2 normative specifications - WCAG 2.1 and the testing framework our taskforce is working on. If we want to publish working drafts of 2.2 or Silver before October, 2019, we need to re-charter.
... we are trying to stay on a 18-24 month update cycle.
... If we want to stay on that cycle, we need to re-charter - and, we need to determine what we want to re-charter: WCAG 2.2 only, WCAG 2.2 and Silver, or Silver only.

Chuck: Can't get to survey link.

Gowerm: Normal W3C credentials didn't work.

<Zakim> Joshue, you wanted to ask what the group feels is required

<Joshue108> webex

Chuck: I'm into the page now.

<Chuck> I see it now, it was the quote in the url.

David: I'm concerned about Silver - there are serious redesign issues with Silver that make it so that it likely will not be able to be developed within the 18-24 month timeframe.

<kirkwood> _1 to David

<kirkwood> +1 to David I meant

David: I'd like for both WCAG 2.2 and Silver development to take place concurrently.

<Joshue108> I'll comment in the survey.

Gowerm: It is prudent for us to continue on the 18-24 month cycle for the work that is derivitive of WCAG 2.0 - takes the pressure off of the Silver folks.

<AWK_> brooks: Yes, I indicated a preference to go to silver

<AWK_> ... we are locked into the 2.0 framework

<AWK_> ... and we need to consider big changes to address many of these issues

<AWK_> AC: What if it took longer?

<AWK_> brooks: would feel the same way

<AWK_> ... that would make me more agreeable to also doing 2.2

alastair: plan is for a Silver plan to be reviewed by TPAC later this year.

awk: we would be looking at working on a charter later this summer - in July.

<Glenda> David, isn’t in 12 new SC at A/AA (for WCAG 2.1)?

David: We keep adding things, but we aren't removing anything from the standard. There's a large increase in the number of requirements. This makes it more and more difficult and expensive to test WCAG compliance.

<Zakim> AWK_, you wanted to speak to backward compatibility

alastair: we will address process improvements to the 2.x work, assuming we undertake that

awk: this is a question that's been raised before - we need to consider a variety of factors. There are some backward compatibility concerns, proliferation of requirements, and other concerns related to conformance. On top of all that, we have an 18 month mandate to consider.
... I think we have to have timeline that is in between 18 months and the 36 month timeline for Silver that peope have been estimating.
... Do people have the availability to serve both WCAG 2.2 and Silver concurrently?

Joshue: I think we need to be conscious of what the market need, in terms of support from us in achieving the mission of enhanced inclusion.

<bruce_bailey> I do not think we are ready (in the next 24 months) to start “publishing drafts of Silver”

<Zakim> bruce_bailey, you wanted to ask about MarcJ survey comment to begin work

Bruce: I think we should enable the work on Silver to move forward - charter should be general in nature.

<Zakim> MichaelC, you wanted to talk about silver in charter even with long timeline

alastair: I think there may be a law of diminishing returns for subsequent iterations of WCAG 2.x. There likely be a reduction in the number of Success Criteria as move forward with 2.x.

<MichaelC> Even on long timeline, need charter to support silver

<MichaelC> since need charter allowance for FPWD

<MichaelC> part of question is whether we can live with ¨just¨ WCAG 2.1 in the meantime?

David: If we want to be involved in Silver, we need to have it in the charter. Will the people who are in the community group for Silver be able to continue?

<MichaelC> re CG, that´s under discussion but expect to find a way to keep working with it

alastair: I expect there will be a transition period.

Progress on Understanding

alastair: Please answer the survey and provide comments, if available.
... Are there questions and/or comments on Understanding documents?

<AWK_> https://www.w3.org/2002/09/wbs/35422/Clearing_Understanding/results

Gowerm: Are the non-text contrast conflicts resolved?

Alastair: I will work through that to make sure those issues are addressed.

Gowerm: I've got questions about the process of pulling a split branch back to the Understanding branch.

awk: we are going to be able change this documents easily and rapidly. if we can't get in touch with Steve immediately, we'll have the capacity to redress his comments as they come in at a later date.

alastair: can anyone commit to review the motion and orientation SCs?

Gowerm: I can help.

David: I can help, as well.
... What is the decision about having to test SCs in a cumulative manner?

Gowerm: My recollection is that they don't have to be tested in combination.

<david-macdonald_> https://github.com/w3c/wcag21/issues/850

Gowerm: for any that have to be cumulative, we need to provide that information in the test case documentation.

David: What do we think, in terms of what will be required to comply with WCAG 2.1? - Cumulative testing or not?

Kathy: The part that we added to the conformance areas addresses this.

<david-macdonald_> Increase text spacing + Flip to landscape + minimize screen to 320 px

alastair: it's about where a site has preventing scrolling

Jon: The reflow requirement is that the content work in a 320 pixel width

<Glenda> I think each SC must be tested on it’s own. Otherwise it becomes exponentially more difficult to test for WCAG

David: On a 1280 screen, if you magnify to 400%, it is the same as viewing content that is formatted in a 320 px width.

Jon: what did you mean by flipping your device?

David: The viewport changes required additional testing for SCs.

Gowerm: It is going to be tricky on where to put this guidance on testing for multiple SCs at once.

<alastairc> https://github.com/w3c/wcag21/issues/883#issuecomment-384800487

alastair: I've also wondered about where to put this information within the standard.

<alastairc> https://www.w3.org/TR/WCAG21/#cc2

<Zakim> gowerm, you wanted to say the Test Procedure is actually at the techniques stage, so it's not as simple as I said

Glenda: I think each SC stands on its own. If we want to make it a requirement that they have to be combined, we need to specify that explicitly. Combining SCs in testing makes accessibility testing exponentially more difficult.

<AWK_> https://www.w3.org/TR/WCAG21/#cc2

<jon_avila> I will object to WCAG 2.1 if contrast 1.4.3 and zoom 1.4.4 are not tested together.

David: I don't mind testing at each of the breakpoints, but testing SCs cumulatively will be difficult.

<AWK_> Worth pointing out that 1.4.3 and 1.4.4 are in WCAG 2.0

Jon: When you zoom in, the contrast changes. Text that used to have sufficient contrast, no longer has sufficient contrast. Foreground and background combos change on the reflow.

<kirkwood> is that not a reflow issue?

<AWK_> It happens all the time, especially with text over photographs.

Jon: Contrast and reflow should be tested together.

<jon_avila> I run into the issue with page zoom

David: I'd like to learn more about what you are experiencing. Are you saying that the layout is changing because of the reflow and causes different combinations of foreground and background?

Jon: Yes, that can happen. There are all sorts of problems that come with zoom. Contrast is one of those things.

Alastair: Backgrounds and foregrounds can move around during zoom.

<Zakim> bruce_bailey, you wanted to say that bad behavior fails 1.4.4 but not 1.4.3

Bruce: I'd like to see us write this up as failure against 1.4.4, not 1.4.3. It gets very complicated to do cumulative SC testing.

<jon_avila> http://www3.hilton.com/en/hotels/florida/hilton-grand-vacations-at-seaworld-ORLSWGV/attractions/seaworld.html

<jon_avila> Look at content at top right when you zoom in -- you may have to scroll to the right to see it.

Gowerm: To measure loss of content or functionality, we are going to have to definition around this

<jon_avila> I have brought the issue up weeks ago. I presented on it at CSUN.

David: This is huge issue. I'm totally fine with us coming to a unified position, but we do need to get consensus and document it somewhere.

<jon_avila> as it is I believe you have to test SC 1.4.4 at intervals anyway such as 110%, 120%, 130%, etc.

Chuck: The success and failures that presently exist, remain. We haven't introduced anything different with WCAG 2.1.

<jon_avila> We introduced a WCAG conformance requirement that variations on a page must conform.

Chuck: If you deemed that there was failure before, it fails now. Same with success - what's there before is there now, with WCAG 2.1.

<AWK_> https://www.w3.org/TR/WCAG20-TECHS/G178.html

awk: 1.4.4 and 1.4.12 are the SCs that talk about loss of content or functionality. It sounds like Jon has clearly included loss of contrast as part of his definition of loss of content or functionality on zoom.
... one way to clarify this is to add specific language to understanding documents that talks about loss of contrast on zoom. If we were talking about cumulative testing for all SC, it would be very difficult. With just 1.4.4 and 1.4.12, its managable.

Jon: I'm worried that we've introduced a WCAG conformance requirement for all sorts of variations of the page.

<Zakim> gowerm, you wanted to say this seems to involve 1.4.3 Resize Text, 1.4.10 Reflow and 1.4.12 Text Spacing

Jon: People with disabilities have to live with this changes in how content is presented in variations of the page.

<AWK_> AWK notes that alistair is right that reflow 1.4.10 is also using similar language

Gowerm: I think that all three of these can be tested together. However, it may not be feasible to test all variations of the page using the three concurrently.

<AWK_> Orientation: Content does not restrict its view and operation to a single display orientation, such as portrait or landscape, unless a specific display orientation is essential.

<jon_avila> Thank you Mike.

Gowerm: Orientation is just defined as measuring orientation.

<gowerm> scribe: gowerm

David: The biggest one for me is orientation. I'm interested in Mike's comment.
... Can I share my content to demonstrate what i was finding?

<jon_avila> Where does David get 200%

<jon_avila> 200 pixels

Alastair: David is sharing and I will walk through

David: There is a sticky header here. You open and close it.
... If you were to turn the orientation, the menu goes off the screen

<jon_avila> doesn't SC 1.4.10 say 256 vertical?

David: Do we need to require people that it meets in both ways?

Kathy: Back to the Orientation SC. We went back and forth a lot.
... This was changed to say that we just don't restrict a user's ability. You should be able to switch from portait to landscape mode.

<AWK_> David, if it didn't work in landscape would it impact everyone or just PWD?

Kathy: It doesn't cover functionality. it just says you can't restrict.

<david-macdonald_> Content does not restrict its view and operation to a single display orientation, such as portrait or landscape, unless a specific display orientation is essential.

Alastair: In terms of how Reflow works, if it doesn't cause horizontal scrolling, we don't have SC text that says anything about height.
... What I was complaining about on your site was the sticky header, which isn't a failure per se
... We don't require it technically, but people should be encourage to address.

David: We arent' requiring but using it as a best practice.
... Headings in 2.0 was similar. We had a huge discussion about whether headings need to be properly nested.
... Are we suggesting that testing at both orientations is recommended.
... Do we want to require testing in both orientations?
... What do you think Jon?

Jon: My understanding was a bit murky.
... I agree that once you start adding them cumulatively together. But contrast + zoom is one thing. Or contrast and reflow.

<laura> Draft Allowing for Reflow technique: https://www.w3.org/WAI/GL/wiki/Allowing_for_Reflow

<laura> Draft Allowing for Spacing Override technique: https://www.w3.org/WAI/GL/wiki/Allowing_for_Spacing_Override

Jon: "operation" is a bit unclear in the SC.
... The contrast should be available in both views.

Kathy: David, for me, looking at the actual requirements, it's more about what happens at the different breakpoints. The landscape view vs portait mode just goes into a different view. It just falls under more of the comformance.

David: How would you test the orientation SC?

Kathy: I would make sure it can be displayed in both portait and landscape mode.
... We allow the content and menus to change.

David: I think I'm getting some clarity...

<jon_avila> The understanding doc for orientation does not align as it says content and functionality

Kathy: There is a failure for using specific CSS that restricts re-orienation.

<jon_avila> https://www.w3.org/WAI/WCAG21/Understanding/orientation.html

David: You flip the device, and make sure it displays in both portait and landscape

The Understanding document will be updated to reflect that.

Kathy: There was some discussion about what "operation" meant.

<alastairc> for more on this discussion: https://github.com/w3c/wcag21/issues/850

Alastair: The link is about the cumulative question.

<jon_avila> I'd second discussion operation offline

Alastair: I think the conformance information we've added, Reflow in terms of different media queries, should be tested across for contrast and others. But when it comes to Text Spacing I don't think it's covered by the Conformance mode. So if you thought it failed before, it fails now. If you thought it passes, it passes.

David: Should we come to a conclusion about the "operation" wording in the SC for orientation?

Kathy: The text that we have in the Understanding was due to the fact the SC says "doesn't restrict ...view and operation".

<david-macdonald_> Operation: The fact or condition of functioning or being active. "the construction and operation of power stations" synonyms: functioning, working, running, performance, action "the slide bars ensure smooth operation"

Techniques overview https://www.w3.org/WAI/GL/wiki/Wcag21-techniques

AWK: "Operation" is in 4 SC in the normative text.

Alastair: I would like to come back to that.
... There are very few techniques

Kathy: the Mobile TF have a wiki page that lists all the techniques being worked on

<Kathy> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/WCAG_2.1_Understanding_and_Techniques_Development

Kathy: Ours are mostly completed.

Alastair: Laura, are there others from the LVTF?

Laura: Not that I'm aware of

AWK: Some tehcniques are easier to do than Understanding documents, some aren't.

<laura> A previous wiki page with 2.1 Techniques

Alastair: Do you ahve any advice?

<laura> https://www.w3.org/WAI/GL/wiki/Proposed_WCAG_2.1_SC_Techniques

<Kathy> https://www.w3.org/WAI/GL/wiki/Techniques

AWK: In the github repository, there is a section on editing techniques, which we need to expand.

<AWK_> README has info about drafting techniques https://github.com/w3c/wcag21

AWK: From a writing perspective, that would be adequate. People are more than welcome to write techniques where it makes sense to them.
... IF I was starting a new technique, I would probably start in MS Word or a text editor and then paste into a github file... I might start with github now, but not everyone is comfortable with that.

Alastair: If anyone would like to sign up for other techniques, please sign up on the page. And feel free to add techniques.
... I didn't include all the wcag 2.0 techniques listed, as I assume they're there.

Any other business

Combining SCs

David: I think we have to have a discussion about Orientation.
... Do we have consensus about whether reflow and text spacing are cumulative

Alastair: Start at 1280, zoom into 400%, turn on text spacing (using stylish), look for overlaps, at maximum checking that text size is at least twice as big. go back down zoom scale to standard

David: And does text resizing have to be on?
... The conformance requirements say it has to work at each breakpoint. Do we require that all SCs have to work at 320?

Kathy: We need to be careful about what we say about how to test breakpoints.
... It doesn't mean they have to have 200 text size at all points

<jon_avila> You could have a non-responsive site that supports 320 CSS pixels with fluid design.

Alastair: Your text could get smaller as you zoom in.
... We do have wording that at some point it must be twice as big.

<jon_avila> I see that a lot too Kathy

Kathy: It is a common scenario that text size will be reduced at some point.

<alastairc> Text from the understanding: If text is reduced in size when people zoom in (or for small-screen usage), it should still be possible to get to 200% bigger. For example, if text is set at 20px when the window is 1280px wide, at 200% zoom it should be at least 20px (so 200% larger), but at 400% zoom it could be set to 10px (therefore still 200% larger than the 100% view. It is not required to be 200% larger at every breakpoint, but it should be possible to get

<alastairc> 200% large text in some way.

<Glenda> I disagree with Jon.

<alastairc> That was from the heading "What is the relation to Resize Text?"

<alastairc> New note: "A full page includes each variation of the page that is automatically presented by the page for various screen sizes (e.g. variations in a responsive Web page). Each of these variations needs to conform (or needs to have a conforming alternate version) in order for the entire page to conform."

David: We need to settle how these are combined so we have clear guidance.

Okay, well, there you go.


<david-macdonald_> meeting ended

<alastairc> oh dear, it looks like Bruce was the host!

<Kathy> yes

<laura> yes

<david-macdonald_> yes

<jon_avila> 1.4.4 says with AT. Platform magnification is considered AT.

<Kathy> have a good day everyone

<alastairc> oh well, um, let's pick it up on Thursday...

<jon_avila> 1.4.4 says with out AT

<jon_avila> sorry

<marcjohlic> i have a webex or zoom if want to continue the conversation - but w3c is tied up ?

<david-macdonald_> ok I think its an important discussion

I suggest we need a discussion about this, because I think this could be a REALLy big problem.

<david-macdonald_> agreed...

<alastairc> I have to go in 2 minutes, but it would be good to get thoughts on the #850 thread, and pick up again

<alastairc> Thanks everyone, I'll close up the minutes.

<alastairc> trackbot, end meeting

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2018/05/01 17:00:50 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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/With just 1.3.3/With just 1.4.4/
Succeeded: s/the fact/Operation: The fact/
Default Present: Laura, gowerm, alastairc, kirkwood, Greg_Lowney, Glenda, marcjohlic, AWK, Brooks, JakeAbma, Joshue108, Makoto, Chuck, Bruce, Kim_D, Kathy

WARNING: Replacing previous Present list. (Old list: Laura, gowerm, alastairc, Detlev, kirkwood, Greg_Lowney, Glenda, marcjohlic, Katie_Haritos-Shea, AWK, Brooks, JakeAbma, Joshue108, Makoto, bruce_bailey, Chuck)
Use 'Present+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Present+ Laura, gowerm, alastairc, kirkwood, Greg_Lowney, Glenda, marcjohlic, AWK, Brooks, JakeAbma, Joshue108, Makoto, Chuck, Bruce

Present: Laura gowerm alastairc kirkwood Greg_Lowney Glenda marcjohlic AWK Brooks JakeAbma Joshue108 Makoto Chuck Bruce Kim_D Kathy
Found Scribe: Brooks
Inferring ScribeNick: Brooks
Found Scribe: gowerm
Inferring ScribeNick: gowerm
Scribes: Brooks, gowerm
ScribeNicks: Brooks, gowerm
Found Date: 01 May 2018
People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)

[End of scribe.perl diagnostic output]