– DRAFT –
AGWG-2022-05-03

<Jennie> Scribing commands for those that want to try: https://www.w3.org/WAI/GL/wiki/Scribing_Commands_and_Related_Info

<JF> agenda

future topics

dl: @@

Announcements

WCAG2ICT

ca: exploring reopening WCAG2ICT TF

<AWK> +AWK

jb: background - after WCAG 2 we were asked to map to non-Web

non-normatively, what WCAG provisions would apply and how?

<Judy> existing WCAG2ICT Overview: https://www.w3.org/WAI/standards-guidelines/wcag/non-web-ict/

<Judy> existing WCAG2ICT Note: https://www.w3.org/TR/wcag2ict/

^ are background on what was produced

we´ve been asked to update for post 2.0

propose to reconstitute TF with constrained scope

draft work statement

Mary Jo Mueller expected to be a co-facilitator, exploring second balancing one

mjm: for those who don´t know me, I work at IBM accessibility on many technologies and standards

goal of WCAG2ICT refresh is update guidance for 2.1 and 2.2 SC

also address open issues on the previous version

<jon_avila> FYI Annex C of EN 301 549 has some guidance on how to apply WCAG 2.1 to software - that may be helpful to look at as part of this exercise.

address topic of ¨closed products¨, to which generalized AT cannot interface

<janina> Noting some WCAG2ICT issues listed in Appendix B of the Challenges Note draft. Don't believe we moved them to github issues, but that could be done.

bring together global standards reps

ca: will have survey open for next call, further discussion on that

ISO

WCAG 2 went through a process to be co-endorsed by ISO as international standard

W3C is de facto standards body for Web

but some orgs need ISO references

we decided not to do this for 2.1, partly because of its rush and associated issues

have been asked to do so for 2.2

important that this come from W3C to ensure harmonization

saz: will there be a need to address comments from the previous ballot

jb: there are some details to be looked at

such as ISO ensuring that they will post an accessible format

there may have been issues slated to address in subsequent versions

think that´s done or OBE, but will check

Chartering Conversation https://www.w3.org/2002/09/wbs/35422/one_or_two_groups/

https://www.w3.org/2002/09/wbs/35422/one_or_two_groups/results

ca: there have been questions about how things would look in a 2-group scenario

we find it´s best to discuss that by reference to concrete draft charters

these are drafts for discussion, not final proposals

1-charter group would have the kitchen sink

2-charter version has a WCAG 2 and related materials group, and a WCAG 3 and related materials group

rbm: want to clarify goals related to the survey

<Rachael> https://github.com/w3c/wcag/pull/2348/files

have made the straightforward changes

^ is for the single-group

<Rachael> propose approaches to known challenges, including those listed above.

remove ¨website¨ from conformance

drop ¨all¨ from ¨address all issues¨

editorial on functional needs list

also request to remove it

wf: why ¨website¨ removed in conformance?

<jon_avila> Website is too limited - what about web apps?

<alastairc> Ah, website as in tackling the full website rather than the conformance of websites. I see.

awk: same question

<Zakim> GreggVan, you wanted to say it is not just websites but pages processes etc.

jf: typo

gv: there´s all sorts of words for the content we target

<Zakim> alastairc, you wanted to say we mean (I think) whole-site conformance.

how about just ¨web¨

<jon_avila> What about web app conformance?

<GreggVan> +1 to alistair

ac: challenge with conformance of complex sites, maybe say ¨whole website conformance¨

<alastairc> Our general scope is "web technologies", so yes, but the tricky bit we are trying to tackle is "whole website"

<janina> Believe "such as" should cover unlisted web-based techs, like watches; etc

<Chuck> poll: option 1) whole website conformance, option 2) "web conformance", option 3) no reference to web

gv: withdraw, there are cans of worms

<Rachael> 1 or 2

<Wilco> 1

<jweismantel> 1

<joeyang> 1

<janina> +1 to #1

<Chuck> 1 or 2

<2>

<Jaunita_George> 2

<alastairc> 1, 2 ok

<JakeAbma> 1

<Azlan> 1 or 2

<jeanne> 1

<GreggVan> 1 or 3 (to match other issues that don't say web

<ShawnT> 1 or 2

<Francis_Storr> 1 or 2

<MelanieP> 2 or 1

<laura> 1 or 2

<maryjom> 1

<JF> 3

<Jaunita_George> I think we don't want to limit ourselves to websites...

<alastairc> E.g. "Complex conformance challenges, such as third-party content, whole website conformance, scoring, accessibility statements, etc."

<Zakim> JF, you wanted to ask if XR = 'web'?

ca: leans towards 1, anyone not tolerate?

<alastairc> 1+

<Wilco> WCAG 3 is only scoped for web

jf: though we wanted to go beyond web

ac: we address web technologies, regardless of how they´re delivered

<Rachael> Context: Examples of known challenges include ...Complex conformance challenges, such as third-party content, website conformance, scoring, accessibility statements, etc.

in this bullet, we´re responding to specific conformance challenges, of which whole sites are one

jb: W3C uses web-based technologies

there are W3C standards related to XR, RTC, etc.

<Zakim> Judy, you wanted to note "web-based technologies" is common usage in w3c, and w3c and wai are both developing standards and accessibility user requirements for these areas

gv: <wordsmithing>

<AWK> Suggesting "Complex conformance questions, such as third-party content, scoring, accessibility statements, etc."

<Zakim> Jennie, you wanted to ask about whole web

<Zakim> Rachael, you wanted to restate context

<Rachael> Context: Examples of known challenges include ...Complex conformance challenges, such as third-party content, website conformance, scoring, accessibility statements, etc.

jd: some reviewers might not know what we mean by ¨whole website conformance¨

rbm: ^ is context for why this wording is here

<alastairc> Suggest: Either except 'whole website conformance', or remove.

it´s a sample list of a sample list

awk: suggest s/challenges/questions/

more clear without reducing scope

<Jennie> +1 to Shadi's "entire"

<Rachael> Context: Examples of known challenges include ...Complex conformance questions, such as third-party content, scoring, accessibility statements, etc.

<AWK> Andrew's Examples of known challenges include... Complex conformance questions, such as third-party content, scoring, accessibility statements, etc."

<alastairc> +1, let's move on.

<Chuck> +1

<GreggVan> ah good point Rachael -- I move to 1 or AWK suggestion of leavign that one out

<Wilco> +1 to moving on

<Jem> +1

<kirkwood> +1

<JakeAbma> +1

<Jennie> +1

<GreggVan> +1

<ShawnT> +1

<jon_avila> +1

<Azlan> +1

<jeanne> +1

<SuzanneTaylor> +1

<Francis_Storr> +1

<MelanieP> +1

ca: accepting the AWK proposal

<Jaunita_George> +1

<Rachael> +1

rbm: what about keeping / removing list of functional needs?

jf: had proposed another

rbm: request was to add conformance to bullets, worked that into context-setting

<alastairc> JF - see line 180

jf: requested ¨defined model¨

<alastairc> Is that on the same topic?

rbm: noting as open issues

<scribe now lost>

<Rachael> Themes: Do we keep the functional needs list, and do we commit to a "defined model" instead of "answering questions including ...complex conformance challenges"

ca: refocusing, there were competing suggestions for the functional needs list, add to or remove

ja: functional needs sub-group now part of APA, maybe the list not needed here?

<Rachael> https://github.com/w3c/wcag/pull/2348/files

rbm: not about committing to specific work, it´s explanatory for scope of work

<Rachael> lines 248-261

<Chuck> +1 to removing, maybe including a link

<Wilco> +1 to remove it

ac: might be better to link than include for charter

<Rachael> +1 to removing and adding link

<michael> +1 to remove

<jeanne> +1 to remove

js: we should be indicative rather than exhaustive for WCAG 3

<Jem> no objection

rbm: ok to substitute with link?

<GN015> +1 to janina

<maryjom> +1 I agree with Janina, Shawn.

ca: no disagreements

<jon_avila> +1 to remove list and add link

ca: back to complex conformance challenges

rbm: request that scope commit to a conformance model for wcag 3

we previously said we´d address conformance questions

ca: think if we list, need to list ´em all again

<jeanne> -1 to calling out just one challenge

jf, would that address your comment?

jf: <reads from survey response>

I request by end of charter we have defined what conformance model looks like

rbm: hesitant to do so explicitly

<Wilco> -1, it's already in the document

plan wide reviews, expect lots of comments

<michael> +1 to not making explicit

<alastairc> +1 to Rachael, and it's already in the 2.1 scope as something that goes into the wide review(s).

don´t think we can have finished by by end of charter period

jf: think we´ve waited long enough

<AWK> +1 to JF's concern.

think we need a forcing mechanism to finally define conformance

<AWK> Doesn't mean that the conformance model can't change if the work of the group makes a change necessary

we keep kicking the can down the road, want to say by 2024 we know where the can will land

<Zakim> alastairc, you wanted to say it's already in the scope.

ac: the scope section addresses the point

<Chuck> poll: add the edits proposed by JF to section 1.2 "Focus for the 2022 - 2024 Charter"

<Wilco> -1, it's already in the doc

<jeanne> -1

<Chuck> poll: add the edit proposed by JF to section 1.2 "Focus for the 2022 - 2024 Charter"

<alastairc> -1, already in the doc

<Lauriat> -1

<Rachael> -1 already in document

<JF> +1

<Chuck> -1

<janina> -1

<Jem> -1

<sarahhorton> -1

<ShawnT> -1

awk: I agree with JF on addressing conformance model

I heard RBM saying we´re not committing to having that done by end of charter

but AC saying already there

<shadi> +1

rbm: I understand the discussion to be about making conformance model final and unchangeable by end of charter

<alastairc> See under "https://raw.githack.com/w3c/wcag/charter-2022/charter.html#scope-wcag3"

don´t think we can go that far

<Rachael> yes, I do

<alastairc> This wide review of WCAG 3 will include the following:... A conformance model, and

awk: when we go to wide review, will there be a conformance model?

<alastairc> "This wide review of WCAG 3 will include the following:... A conformance model"

or a request for input on one

<laura> +1 to awk

I say there should be one, which we´re open to changing based on feedback

saz: +1 to AWK

<jon_avila> I agree with Andrew. We want a full model - but not one that can't be changed at all in the wide review.

if the milestone is wide review draft, operative word is ¨draft¨

things aren´t unchangeable

would like language about having taken a good step ahead

while avoiding misinterpretation about done deal

jf: I want to know how the conformance model will work

it gates other activities

so think this should be more than just in scope

<Zakim> alastairc, you wanted to say it includes "This wide review of WCAG 3 will include the following:... A conformance model"

<AWK> Seems like there is inconsistency between the 2.1 WCAG 3 section's "A conformance model, and" and the text in the WCAG 3 Conformance model.

ac: <quotes from charter where it says these things>

<AWK> We should make these sections consistent

jf: want a working example of WCAG 3 conformance model

at least differences between levels

wf: I see it as already there

<quotes>

jf: s/define scope of conformance model/define working conformance model/

<Chuck> poll: add the edit proposed by JF to section 1.2 "Focus for the 2022 - 2024 Charter"

<alastairc> Suggest: "In this charter period, AG WG will define *a* conformance model for WCAG 3."

<Chuck> ack

<SuzanneTaylor> +1 to "working conformance model"

<Chuck> +1

<laura> +1 to "working conformance model"

<ShawnT> +1

<JF> +1

<jon_avila> +1

rbm: suggest JF additions with a caveat

<shadi> +1 to alastairc

<janina> +1 to initial

<JakeAbma> +1

<Wilco> +1

<Jaunita_George> +1

<jweismantel> +1

<Lauriat> 0, seems redundant

<alastairc> "In this charter period, AG WG will define a working conformance model for WCAG 3."

<sarahhorton> +1 with "working"

<Rachael> straw poll Change "scope of conformance model" to "working conformance model

<michael> +1

<Rachael> 0

<alastairc> +1

<JF> +1

<AWK> +.5 Can live with it but think that the "working" is already implied by the draft status

<Jem> +1

<SuzanneTaylor> +1

ca: poll passes

<sarahhorton> I can scribe

<Zakim> JF, you wanted to remind the W3C Process document

Visible controls (No Survey)

alastairc: Acknowledge Gregg's effort
… wanting to make progress, need to wrap things up, haven't gotten past core issue of applying conventions
… haven't made enough process, need to pull SC out

Chuck: Recommended approach? No survey, last week resolved to finalize or pull, addition resolution needed?

alastairc: at stage to pull it, no further resolution needed, didn't make enough progress

WCAG 2.2 Focus appearance https://www.w3.org/2002/09/wbs/35422/wcag22-focus-appearance-enhanced2/

michael: Resolution, yes, add something that it was brought up and addressed?

<michael> Okay

alastairc: Have in minutes, brought up as topic, has scribe

<Chuck> https://www.w3.org/2002/09/wbs/35422/wcag22-focus-appearance-enhanced2/results

Chuck: [reads survey]

Chuck: 4 agree, 1 revert, 2 something else

<Jem> https://github.com/w3c/wcag/pull/2341/

<AWK> Can we see this in context of the full SC?

<Jem> https://github.com/w3c/wcag/pull/2341/

alastairc: Had 2 topics, came up with 3rd which is link

Question 1 - Adjacent contrast

<alastairc> "has a contrast ratio of at least 3:1 against adjacent non-focus-indicator colors, or is no thinner than 2 CSS pixels."

alastairc: non-focus indicator, new term

<Chuck> proposed RESOLUTION: Accept PR 2341 to address issue 2333

<Zakim> AWK, you wanted to ask if overlapping is included in adjacent?

<alastairc> Plain preview: https://raw.githack.com/w3c/wcag/30ec82a645f256c6a4a1a4fd9b603ae1746f868e/guidelines/sc/22/focus-appearance-minimum.html

AWK: Brings up other question re scope of adjacency, e.g., focus indicator dot in control? Is that adjacent? Could address in understanding, just make sure clear

Detlev: PR, not sure understand, 3 bullets, meets all, 2nd stipulates contrast and next says "or", e.g., 2-px border, does it not have to 3:1?

<Zakim> alastairc, you wanted to say adjacent / change of

alastairc: Meets all, must have changed compared to unfocused, falls down with dark indicator, dark button, trying to catch that with last bullet
… without punishing gradient indicators, because gradient won't have contrast with self

Detlev: 2px indicator, slightest contrast possible, would it still pass?

alastairc: Would have to meet change of contrast
… anything inside controls working on change of contrast, adjacent comes with up against other things

<Chuck> proposed RESOLUTION: Accept PR 2341 to address issue 2333

alastairc: Other things in PR?

alastairc: Resolve normative change first, not accept PR

<Chuck> proposed RESOLUTION: Accept normative change of PR 2341 to address issue 2333.

<alastairc> +1

<Chuck> +1

<Detlev> +1

<Jem> +1

<jon_avila> +1

<ShawnT> +1

<bruce_bailey> +1

<jweismantel> +1

<Francis_Storr> +1

<laura> +1

<GN015> 0

<joeyang> +1

<JakeAbma> +1

RESOLUTION: Accept normative change of PR 2341 to address issue 2333.

Question 2 - Adjusting sub-component requirement

<alastairc> https://github.com/w3c/wcag/pull/2341/files

Wilco: Does this include "a" to "the" for bounding box

<Wilco> https://github.com/w3c/wcag/pull/2341/files#diff-444c5829eebc004b2bed36e50ccd93bbc460c08d46c2bf15d4675daa34bfd9bfR21

alastairc: Just the non-focus indicators

<alastairc> Sorry - that was focused on "non-focus-indicator colors"

Chuck: [reads survey]
… 7 agree, 1 adjustment

<Jem> #2 question in https://www.w3.org/2002/09/wbs/35422/wcag22-focus-appearance-enhanced2/

bruce_bailey: All good with edits

<Jem> better link is https://www.w3.org/2002/09/wbs/35422/wcag22-focus-appearance-enhanced2/results#xq50

<alastairc> https://github.com/w3c/wcag/pull/2345/files

alastairc: In survey, agree with "instead"? Mike suggested advancement

<alastairc> "Where a user interface component has active sub-components, if a sub-component receives a focus indicator, these requirements are applied to the sub-component instead."

alastairc: worth discussion
… update to make in addition to what's in survey

bruce_bailey: Can be choice that it can be either one, support for that in survey, clarity that it's up to author/page owner

<Chuck> proposed RESOLUTION: Accept amended understanding document update for sub-component requirement

<alastairc> "Where a user interface component has active sub-components (for example, an opened drop-down menu shows a list of menu items), if the focus indicator is applied to the sub-component then these requirements can be applied to the sub-components instead."

<Chuck> proposed RESOLUTION: accept "Where a user interface component has active sub-components (for example, an opened drop-down menu shows a list of menu items), if the focus indicator is applied to the sub-component then these requirements can be applied to the sub-components instead."

michael: Bring back points from Bruce, focus on the "instead", tackle other topic in another PR

<Jem> +1

<alastairc> +1

<Chuck> +1

<Wilco> +1

<joeyang> +1

<jweismantel> +1

<bruce_bailey> +1

<ShawnT> +1

<JakeAbma> +1

<Francis_Storr> +1

Question 3 - Bounding box for separated links #2323

<Jem> https://www.w3.org/2002/09/wbs/35422/wcag22-focus-appearance-enhanced2/results#xq51

<alastairc> Chuck: Reading question 3 from survey

<Jem> +1 to wilko that it is too hard to understand.

<alastairc> See https://github.com/w3c/wcag/issues/2323 for some visuals

GN015: Intention to catch, original wording not clear, hope for better wording

<bruce_bailey> oops, missed "concatenate rectangles" -- i agree that is not clear

alastairc: 2 bits, normative useful, looking at size measure, "the" minimum bounding box tightens up
… disconnected parts, each component has own bounding box, if use border property and link wraps, miss out right-hand of first bit, second bit of second
… if considered separately and 1 pixel, going to fail
… if use outline won't fail, complete box around each link
… options, switch from border to outline, or use 2 pixels
… not clear how updating definition will work
… would need different change

<Jem> By the way, "concatenation" is not a plain language

GN015: Border vs outline, should not force all 4 lines for all parts, other ways to indicate

<Zakim> Chuck, you wanted to say still unaddressed concatenate rectangles

<Jem> wondering what would be downside if we don't cover/address this.

Chuck: Verbal explanation makes sense, didn't address concern about concatenate rectanges

alastairc: Suggestion to change size measure based on if it hadn't wrapped

GN015: Yes

Wilco: Every case like that need multiword link with larger indicator than necessary even if side is missing, already passing

<Jem> +1 to wilco

alastairc: Visual demonstration

alastairc: Demo using border and outline, outline fits requirement, border doesn't
… border version could fail, border rare because affects layout

<Jem> love to see the visual, but is that one word or two words?

<Jem> one word with multiple links.

alastairc: wraps, if 2 pixels passes even if 1 character

Chuck: Confused, demo supports?

Wilco: More work needed

alastairc: Don't want to create hole, [demo], opening up holes

Chuck: resolution? Not sure supporting or against extended definition

alastairc: Could lead to odd things if you use border, but border rare, things without border, adding border makes it bigger, causes things to jump around
… don't think we need to solve it, could lead to more problem

GN015: Focus indicator common, surrounding box isn't a rectangle

<mbgower> Okay, I'm back online.

<mbgower> It passes.

GN015: indicates the object continues, current definition requires two outline

<bruce_bailey> +1 to "I feel this should also be allowed" sentiment -- but also to alastair comment that it is a pass

GN015: Would like outline and border to pass

<Jem> +1 to edge case.

<Chuck> +1 to edge case

<joeyang> +1 at edge case

michael: Edge case, not sure how often problem, as is provide easy solution (outline)

michael: [walks through examples]

<bruce_bailey> link to comment under discussion in GitHub thread: https://github.com/w3c/wcag/issues/2323#issuecomment-1116069154

michael: proposed wording allows half a focus indicator and pass
… fragmented focus unusual except for text links

<Jem> agree with that it would be rare to have a fragmented focus indicator.

<Chuck> proposed RESOLUTION: Update the definition as proposed, do not accept the normative change

michael: use outline, easy technical thing, or thicker outline

<Wilco> maps can have this, countries with multiple islands

alastairc: [demo of use of border, things move around]

<Jem> half focused 4 rectangle in one reatangle would be rare.

GN015: Focus indicator with outline instead of border, or 2 pixel, past discussion to not prescribe indicator
… moon shaped indicators
… now prescribing

<Jem> AG have star rating focus indicator example

alastairc: Very rare technique to use border
… harder to meet, not saying you can't use other methods
… harder to do than default

<Zakim> mbgower, you wanted to say the requirements have flexibility but are still prescriptive

michael: requirements have flexibility but are prescriptive, need to meet one of the approaches
… exception covers scenarios except this one edge case

<Chuck> proposed RESOLUTION: Update the definition as proposed, do not accept the normative change

alastairc: "a" to "the" change is normative

<Chuck> proposed RESOLUTION: Update the normative change, do not accept the changes to understanding

alastairc: Not update definition

<alastairc> proposed RESOLUTION: Update the normative change to the SC, do not accept the changes to definition

<mbgower> +1

<Jem> +1

<Chuck> +1

<bruce_bailey> +1

<ShawnT> +1

<jweismantel> +1

<alastairc> +1

<joeyang> +1

<laura> 1

<Detlev> +1

<Wilco> 0, I agree with GN, but don't see a solve

RESOLUTION: Update the normative change to the SC, do not accept the changes to definition

Chartering Conversation https://www.w3.org/2002/09/wbs/35422/one_or_two_groups/

WCAG 2.2 Page break locators https://www.w3.org/2002/09/wbs/35422/wcag22-page-break-nav/

Question 1 - SC2.4.13 Page break locators technique #1226

<Jem> https://www.w3.org/2002/09/wbs/35422/wcag22-page-break-nav/results

Chuck: [reads survey]
… 10 agreed, 1 something else

<Jem> https://raw.githack.com/w3c/wcag/afba356fd624c0d0ba3e1c3d8739a36e91c0e01a/techniques/html/H99.html

alastairc: Change added footnotes, make them into links instead

<ShawnT> +1 to alastairc

<laura> +1

<ShawnT> +1

<Chuck> +1

<alastairc> alastairc: Just want to change the footnotes into links (TODO)

<jweismantel> +1

<bruce_bailey> +1

<alastairc> +1

<mbgower> +1

RESOLUTION: Accept amended PR 2130 to address issue 1226.

<Detlev> +1

Question 2 - Are CSS break-before/after indicators page break locators? #2259

Chuck: [reads survey]

<alastairc> Link to issue raised: https://github.com/w3c/wcag/issues/2259

Chuck: [reads responses]

michael: Conversation show why still problematic SC

<Zakim> bruce_bailey, you wanted to mention alternate formats

<Jem> I also have the same question regarding "alternative version" as Wilco, David, and Michale mentioned

bruce_bailey: Dancing around print, 504, 508, alternative formats, e.g.,braille, language is encroaching, not careful, if ePub are okay then okay

alastairc: Ran by ePub
… hinted at it and point to technique

<Jem> I think "programmatically determinable destination markers that represent page breaks may NOT be the same as in an alternative version of the document

<Zakim> mbgower, you wanted to say that this exceeds WCAG IMO

Chuck: Outside of scope of question

michael: Concern, exceeds WCAG scope, deal with web content, non-web alternative not within purview

<Jem> +1 to mbgower

<bruce_bailey> i agree that if Matt Garrish is happy, i am happy

<Zakim> alastairc, you wanted to say what I was trying to say...

michael: scoped down to page break locators, okay with existing langauge

<Jem> Epub may have something to cover alternative format.

alastairc: Have something in webpage/thing that represents page break if version different from what evaluating
… CSS page breaks aren't related

<Chuck> proposed: programmatically determinable destination markers that represent page breaks in an alternative web version of the document.

<Jem> definition form epub "Page break locators are also commonly used to provide static markers in purely digital publications (i.e., where no statically paginated equivalent exists). These markers provide consistent navigation regardless of differences in font and screen size that can otherwise affect the dynamic pagination of the content."

alastairc: things within page trying to evaluate that are pointer to something else, can do in understanding

Wilco: Can tolerate if clear understanding document with current

<Chuck> proposed RESOLUTION: reject the proposed definition change

Chuck: Updated understanding later?

<Chuck> +1

alastairc: Yes

<Jem> +1 am still confused about alternative format context

<Wilco> +1

<bruce_bailey> +1

<mbgower> +1

<jweismantel> +1

<joeyang> +1

<ShawnT> +1

<Francis_Storr> +1

<alastairc> +0.5

<laura> +1

RESOLUTION: reject the proposed definition change

<Rachael> +1

<Detlev> 0 can't tell

<laura> bye