W3C

- DRAFT -

Accessibility Guidelines Working Group Teleconference

09 Oct 2018

Attendees

Present
MichaelC, Brooks, alastairc, Detlev, JakeAbma, david-macdonald, kirkwood, jon_avila, Katie_Haritos-Shea, gowerm, Chuck, AWK, Laura, marcjohlic, johnkirkwood, JF, Mike_Elledge
Regrets
Chair
SV_MEETING_CHAIR
Scribe
Chuck, Mike_Elledge

Contents


<alastairc> agenda order 5, 6, 7

<AWK> +AWK

<Chuck> scribe: Chuck

Notices - TPAC agenda, upcoming requests from TFs, no meeting week after TPAC, last chance for process review pre-TPAC, PDF techniques

<AWK> TPAC Agenda: https://www.w3.org/WAI/GL/wiki/Meetings/TPAC_2018

TPAC agenda

alastairc: Things coming up we want to make people aware of, not expecting lots of feedback.
... Not changed much from last week.

AWK: Related to silver and task force updates (ACT) we expect next week we will have ACT Task force facilitators to give working group a preview, and we hope to get a preview from Silver as well.
... We hope to get to the meet of topics, instead of spending time getting anybody up to date. Hopefully will help us be productive at TPAC.

alastairc: Any q on TPAC?

AWK: Has everyone registered? Registration is changing sometime soon.

<alastairc> scribe: Mike_Elledge

No meeting the week after TPAC.

<AWK> No meeting on the 30th

Since it's the week after tpac, we'll skip that meeting after, people needing to catch up.

Process review -last change

Thanks to IBM and Mike Gower that filled in survey.

<AWK> https://www.w3.org/2002/09/wbs/35422/process-results-pt1/results

Thanks to others as well.

Will review this week, come up with proposals for tpac. Decision process, workflow, tools and editing/updating.

You have a couple more days to make comments!

PDF techniques. Have a lead to take on techniques.

<Ryladog> Where will the meeting be?

<alastairc> I think it's Edinbrough (Scotland)

awk: Duff Johnson very involved in PDF association. Want to provide guidance for conforming to 2.1 with pdf. They are working on techniques.
... Discussed concerns about specific products, Acrobat, Open Office, and that content tends to become stale. Most techniques don't talk about specific tools, so will look into getting specific vendors out of techniques.
... But vendors could provide links to documents with techniques. So WCAG could provide links.

jk: Who is heading that up? PDF association...Duff Johnson?

awk: yes

Techniques for approval - https://www.w3.org/2002/09/wbs/35422/TechniquesforApproval/

<alastairc> https://www.w3.org/2002/09/wbs/35422/TechniquesforApproval/results

Understanding text-spacing, stylish to stylus update

#1 will be quite straightforward. Doc for text based spacing. Suggested extension for testing. Has been banned by Chrome and Firefox. Was capturing user info.

ac: There is another version that is okay, forturnately: Stylus. Added to understanding document. Everyone seems happy with it.

dm: Should we mention that there could be confusion between that and similar sounding product?

ac: Not sure how we would do that.

dm: have a note wherever we mention it.

ac: Almost dont' want to mention the other one.

dm: There is a similar one with similar name.

ac: Has been banned, so not available with Firefox or Chrome.

dm: Had a class when found it couldn't be used!

ac: Got around the deletion...

dm: Think it's still in Chrome with a note.

me: q?

ac: Hoping what we have will be enough

a

q

<Detlev> Don‘t mention stylish

awk: Worry that we would have to make lots of revisions!

ac: Resolve to publish 487

RESOLUTION: Publish PR487
... Publish PR487

Technique: wrapping long URLs for reflow

ac: PR 486

<alastairc> https://rawgit.com/w3c/wcag/tech-reflow-url/techniques/css/reflow-url.html

ac: Rawgit is closing down

dm: Oh NO!

ac: Currently links work. Can still use it for a few more months for existing urls. DK about new ones.

<alastairc> https://rawgit.com/w3c/wcag/tech-reflow-url/working-examples/css-reflow-url/

ac: If you have long urls, it can wordwrap. example that is useful to see how it works. page puts url in foreground. First example pokes out of container. Use CSS when set toggle button.

mg: Reverse two examples. Second will stand out.

ac: detlev comments.

detlev: Eitehr focus on long strings, or be about urls with test on url.

<AWK> "Allowing for Reflow with Long URLs and Strings of Text" ?

<AWK> may come up with a lot of lengthy german words?

ac: comes out of experience by low vision task force. Does apply to long test, but have seen only with long urls. New title update comments: might come up more in German.

<laura> +1

ac: Don't think anyone in lvtf would object to different title.

detlev: Talks about long space, not exclusively about urls.

ac: Andrew's suggestion for title will make it clearer.

dm: We have two urls, first wraps, second doesn't so use toggle button.

ac: depending on browser, first may wrap as well. Maybe have just one example that doesn't wrap.

dm: First one wraps in Chrome.

ac: Doesn't wrap in IE.

brooks: Wraps in firefox.

ac: Take out first example.

dm: second wraps quite a bit, has ampersands.

mg: Second one actaully overlaps.

ac: Will poke Jim to make sure not missing something. Will remove first example.

<Brooks> The first example wraps in Firefox, but the second example doesn't wrap by default.

ac: Does it require 1280 x 740?
... Using same procedure as other LV items for consistency.

awk: if you have content that has columns, room to zoom in, columns don't all go away, it may be that scrolling is not required. That the content overruns the container.
... Is it always going to be scrolling, or specific window size. Seems straightforward wrt items wrapping correctly within their containers.

ac: Would that be reflow or increase text?
... Would seem to be easier to have consist procedure with reflow. Image widths, long urls, flexbox and grid layout techniques.

<laura> +1 to keep it consistent.

<david-macdonald> overflow-break: break-word: Allows words to be broken and wrapped within words.

<david-macdonald> overflow-wrap:

mg: Occuring with german words not just urls. Not just anchor text. Do we want to say waht elements it applies to.

ac: Most browsers cannot apply to specific elements, in IE have to apply to all content. Didn't find any harm in having it on. TL Styles group: can't wrap because...

mg: Enough to say does not apply only to urls.

ac: Andrew's change should suffice.

dm: typo over-break vs. over-wrap?
... first bullet?

awk: Disconnect between first and second bullet.

ac: Should check that. Browser specific things to check and correct.

awk: Word-wrap and over-wrap.
... Need to change bullets only. Text is fine.

dm: overflow-wrap.

ac: Changes to title, examples should be singular,

awk: Don't think we can do anything about examples...

ac: Keep procedure consistent with other examples.

awk: Example 2, which would be come #1, text says "no style applied". But follow with CSS, so what does that mean?

ac: All the script does is talk to the class, probably jsut remove the text.

awk: Agree.

<alastairc> Title: Allowing for Reflow with Long URLs and Strings of Text

awk: propse if changing the titel, not sure live example points back to technique, since the only palce you get is live example. Disconnect between text and technique now.

ac: Helpful to have link back to technique from example. Should lien them up. Change first example to very long text.

dm: Mary poppins reference...supercalifragiliciousexpiallidocious...

ac: If we make changes...is everyone happy with technique.

<laura> +1

ac: Any objections?

+1

<Detlev> +1

dm: guess that's what the meaning is, long url...prior to seeing the technique would not fail long url. Now will.

ac: fair enough. if have it as sufficient technique for that type of content, otherwise would get horizontal scrolling.

mg: Not sure it would follow. not an automatic failure technique.

dm: since have technique that show how to wrap it; corollary is that there is a failure.

ac: does it need a re-review.

awk: nope.

<Ryladog> 1

<Ryladog> +1

<Brooks> +1

<Ryladog> :-)

<david-macdonald> can live with it

ac: RESOLUTION: Publish PR486 as amended

RESOLUTION: Publish PR486 as amended

RESOLUTION: Publish PR486 as amended

Technique: media queries for sticky elements

jf: Want to go back to dm's comment. Are we writing a failure for that?

ac; don't have in list.

jf: Don't need one.

<alastairc> https://rawgit.com/w3c/wcag/tech-media-queries-sticky/techniques/css/media-queries-sticky.html

jf: Just wanted to verify...glad working on Sufficient Techniques (positives) first

ac: Focusing on positive techniques to start.
... Advisory technique to use media queries..normative do not rule out sticky headers
... If you hae a sticky header, use a minimum height, so won't stick.
... mg least happy. Not clear that this was an example.
... Maybe depends on window sizing...
... have to go to quite small size before it is unstuck..
... Not dependent on width. Vertical media query is waht applies.
... Okay unless we can clarify.

jake: make height of browser two-thrids of screen, in landscape will be dropped. For tablets, too. More than 150% will also be dropped. Tried to find some middle ground. An example sticky headers will be dropped. Might be clearer if make header and footer a little bigger.

ac: simple in principle, but not always easy when zooming around in browser. Can scroll down slightly, will see it doesn't disappear. could make those a little bigger.
... detlev comment to clear usp some things in text

detlev: in user testing sticky header when user tabs backward focus can be bheind sticky header. Should we point out those potential problems? Or is this a specific technique?

mg: yes drives me nuts, detlev!

Katie: when i tab forward and back, and clsoed out the sticky header, it comes back into view. so I can see where I'm tabbing.

ac: if wndow is small enough, still get header at top of page, just not stuck anymore. as I tab it opens up the sticky header, so haven't lost focus. Seems to be working just fine.
... Not so much this exaple, if in quite a long page, tabbing back through content, link gets hidden from sticky header. Example doesn't show that.

katie: Would have to see that.

ac: In terms of htis technique, shouldn't say it doesn't cause problems. Should say from a zoom pov not a problem on the desktop. Need to qualify it.

katie: agree

<Chuck> scribe: Chuck

<Ryladog> Bye Mike

alastairc: We can deal with detlev's comments by simplifying slightly. No need to talk about complexity, just sticky headers. Covered keyboard tabbing.

AWK: Did you say this was advisory?

alastairc: Yes, in that ... I'm assuming advisory, in that it is a useful thing for people to do, but it's not something we would fail according to normative text.

AWK: Understood. In the technique there's nothing that says advisory. We should make sure that's clear and categorized appropriately.

alastiarc: I assumed that was in the understanding doc. Should it be in the technique?

AWK: Just need to be sure it's crystal clear that we aren't saying "what do you think this is...?"

alastairc: I haven't done an advisory before, good question.

AWK: A lot of times we have language that calls it out. I will take a look and see. I was also confused by medium complex. If that's out, that's great.
... In the description, it says that the basic principal of the .... I thought that the approach to unfixing sticky headers seems strange to say the basic principal is these two bullets.
... Just struck me as strange to read.

dm: Just looking at... Doesn't have anything that says it's advisory itself, it's the link that says it's advisory. Not sure if that's true without.

<david-macdonald> https://www.w3.org/TR/WCAG20-TECHS/C18.html doesn't say its advisory

alastairc: Feels like it's worth going through with another understandability check on the text around where it mentions complexity. There's a couple of other points.
... Any comments about the validity or what it's trying to do?
... Assume not.

katie: I think it's really good.

ac: I think it is almost good to go. It would be good to have a quick refinement on the text and check in with Jake to make sure we have not changed anything significant.
... Final approval next week. too many things at the moment to put it up for publishing, hopefully refined and ready for next week.

RESOLUTION: Leave open, finalize next time.

Technique: failure for hover / focus content not dismissable

<alastairc> https://rawgit.com/w3c/wcag/tech-failure-hover-focus-content-not-dismissable/techniques/failures/tech-failure-hover-focus-content-not-dismissible.html

ac: I think has a slightly clearer but longer title in the fully rendered version.
... One thought it's ready, one would like some discussion.
... detlev commented on his own technique, wants to do some changes.

detlev: Not sure at what point the document was taking out. I thought you had a concern with removing the documentation. It's about dismissing hover content.
... Any kind of strange keyboard shortcut. That would not be a way of meeting this sc. I think it should be any other documented keyboard shortcut.
... That's my only concern.

ac: Remembering conversation from late august. My point was that the sc doesn't mention anything about documented. Was difficult to rely on that.
... detlev: would like to refer to it as documented other shortcuts. Andrew is pointing out that ...

detlev: It should be changed then.

ac: Could be my fault....
... awk's main comment is that as a failure...

awk: this might be as good of an example of as good an example as we could have as to why it's hard to write failure techniques. To be a failure technique, it has to fail all the time.
... This has three bullets. The first one has an out, there's a variety of exceptions. If we have a failure technique, everything must fail.
... I think that something like this, having a success technique or a few would be easier than one failure, just because... the procedure would have to be 6 - 8 things long.

<JF> +1 to AWK

detlev: Three bullet points... would you agree that failing any of those three would fail the success criteria. Any of the three would be a failure?

awk: yep.

detlev: It's wise in the failures to... that's why I wrote the second one. I think it has to be a separate failure.

awk: I do agree. If you just focus on the dismissability, you have to pull in checks... as part of not addressing dismisable. Probably also need to include something about the user agent.

detlev: That makes the test procedure more awkward and longer. I can try to add that in next week.

ac: I think that those examples do try to talk about it and cover it. Would it be easire to translate it into a sufficient technique to providing escape?

detlev: Sure. It's an option.

ac: I think especially in a case where we have multiple bullets and things, a long procedure, it would be nicer all around to say "in this scenario, use the escape key..."

awk: I think we can have failure techniques, they both have challenges. If you have a failure you have to catch all the edge cases, neither is a trivial effort.

katie: Not a trivial effort. We need both, success and failures. I think that him digging in further would be important to identify the particular criteria has these ...
... Also have the success technique. None of this is trivial.

<laura> +1 to Ryladog

ac: Yes, a matter of priortization. Do we prioritize sufficient techniques over failures?

dm: Widen the conversation... I'm a little uncomfortable widening the conversation in that way. I am in the camp that we need failure techniques. If you have sc and no way to fail...
... Then we wouldn't be providing a service.

ac: Didn't mean to go into wide conversation, detlev seems happy to tackle this next week.

<Ryladog> Team FAILURE, or Team SUCCESS = We need to be on both teams...:-)

detlev: Having said that, I see that I have trouble working with github in ipad. Just trying to select the right branch to make a trivial change. Not easy to operate with touch.
... Will have to wait until week after next. If you want me to focus on sufficient technique, I will priortize that and put failure on the back burner.

ac: Andrew was very even handed. I was trying to work out which was quicker and clearer. Change to success technique, or expand the failure technique?

detlev: Should have a good working example. I would have to look around, or someone who knows of a positive example.

ac: If anyone has any ideas, please email. I'll try and sort out the branches. And I've used a particular app on the iPAD. I'll send you later.
... Not ready to publish that one.

<david-macdonald> passing example? https://jqueryui.com/tooltip/

RESOLUTION: leave open and detlev will think about improving procedure or convert to success technique.

<alastairc> https://inclusive-components.design/

ac: in terms of those examples, I would mention "inclusive components" site.

detlev: Yes, I will look there first.

ac: his javascript is usually quite clear.

<alastairc> agenda

ac: That's the end of our programmed material today.

Another other business

ac: we've had a failure light week this week. Next week won't be so light. We have silver, act, possibly some other things for people to read or review.
... week after that is TPAC.
... Pleanty of things to do. Whole list of github techniques that are ready for initial review.

<alastairc> https://github.com/w3c/wcag/pulls?q=is%3Apr+is%3Aopen+label%3A%22Ready+for+initial+review%22

<JF> Regrets for next week - will be at a11yTO Conf, and then flying to Europe on the Wednesday

ac: Anybody have anything else to add or discuss?

jf: I'll be at TPAC Sunday.

ac: looking forward to it.
... Anyone else?

<Ryladog> Thanks!!

ac: wrap up.

<alastairc> trackbot end meeting

Summary of Action Items

Summary of Resolutions

  1. Publish PR487
  2. Publish PR486 as amended
  3. Publish PR486 as amended
  4. Leave open, finalize next time.
  5. leave open and detlev will think about improving procedure or convert to success technique.
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2017/02/06 11:04:15 $

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/glad working on SCs/glad working on Sufficient Techniques (positives) first/
Succeeded: s/priortize/prioritize/
Default Present: MichaelC, Brooks, alastairc, Detlev, JakeAbma, david-macdonald, kirkwood, jon_avila, Katie_Haritos-Shea, gowerm, Chuck, AWK, Laura, marcjohlic, johnkirkwood, JF, Mike_Elledge
Present: MichaelC Brooks alastairc Detlev JakeAbma david-macdonald kirkwood jon_avila Katie_Haritos-Shea gowerm Chuck AWK Laura marcjohlic johnkirkwood JF Mike_Elledge
Found Scribe: Chuck
Inferring ScribeNick: Chuck
Found Scribe: Mike_Elledge
Inferring ScribeNick: Mike_Elledge
Found Scribe: Chuck
Inferring ScribeNick: Chuck
Scribes: Chuck, Mike_Elledge
ScribeNicks: Chuck, Mike_Elledge

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 09 Oct 2018
People with action items: 

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


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]