<dj> scribe: dj

Introductions / meeting items

alastairc: introductions?

<ben_tillyer> I’m having issues, currently connecting from my phone instead as I have no client on my work machine

alastairc: any burning issues people would like the group to discuss later?

Introduce Outcomes Pull Request https://github.com/w3c/wcag3/pull/62

<Rachael> PR link: https://github.com/w3c/wcag3/pull/62

<Rachael> Link to Summary for this meeting: https://github.com/w3c/wcag3/discussions/60#discussioncomment-9200175

alastairc: introduced last week, lots of changes since then
... some items we think need more discussion:

<Rachael> Preview link for pull request: https://raw.githack.com/w3c/wcag3/guidelines_update_april_24/guidelines/index.html

alastairc: https://github.com/w3c/wcag3/discussions/60#discussioncomment-9131345
... are we sure that VR and other future technologies will not allow author to control audio (instead of user)
... does anyone know?

mbgower: current state of technology: OS controls that matrial
... fits in with 508 requirement
... stereo could potentially be a problem
... similar to Reduced Motion

Rachael: (chair hat off) i thinking leaving it in at this point makes more sense, because we can remove it later

<Chuck> +1 keeping it in for now

Rachael: on the second question, i think adding spatial audio requirements are useful

JenStrickland: question for Mike: what's the connection between here and Section 508?
... also agree with Racheal on leaving it in because many websites now try to control sound

jon_avila: to the 2nd point: i think we need both
... example: audio games for the blind assume that you can hear spatially
... 1st part: it seems like the only solution is to use mono audio only, which seems heavy handed
... i'd like a more nuanced approach

alastairc: it seems like there are two requirements: preventing harm and loss of information

<Zakim> mbgower, you wanted to say that it needs to be phrased from an author viewpoint and to

jon_avila: yes, and the specific problem with harm for some people is more about how specifically stereo is used, not just it's ppresence

<JenStrickland> Ah, I think what Mike Gower is suggesting is looking to Section 508 as an example for how to document it in the guidelines.

mbgower: 502.2.1 - platform software defines a11y features and allows people to use them
... 502.2 - no disruption of a11y features

502.4 - platform a11y features

scribe: we can't define that level of stuff in the web,
... but where there are standards that define that, we can put wording in to say "what are the author responsibilities to support these things"

<Zakim> alastairc, you wanted to say it would be useful to leave in, even if there is no author method yet

<mbgower> https://www.ibm.com/able/requirements/requirements?version=v7_2#502_2_1

<mbgower> https://www.ibm.com/able/requirements/requirements/?version=v7_2#502_2_2

alastairc: (chair hat off) i think it'd be useful to leave it in right now even if there's nothing web-specific

<kirkwood> are we covering that it can cause induced dizziness with balance/proprioception issues? (sorry if covered)

<mbgower> https://www.ibm.com/able/requirements/requirements/?version=v7_2#502_4

ben_tillyer: this really helps me as someone with an audio processing disorder

<jon_avila> Recorded audio in video can use it - so it is available today on the web.

ben_tillyer: minecraft's sounds subtitles show you what direction a sound comes from
... and contrast increases/decreases with distance to the sound
... so i definitely think we should leave this in
... especially if we can influence browsers to do that

<alastairc> Poll: Keep the audio-shifting outcome, and add "Important information is not presented in spatial audio alone." and "Important information does not require depth perception." Please + / - 1

ben_tillyer: or at least add something to the audio/video components to handle this

+1

<JenStrickland> +1

<Chuck> +1

<kirkwood> +1

<mike_beganyi> +1

<julierawe> +1

<giacomo-petri8> +1

<Francis_Storr> +1

<jon_avila> +1 to both

<mbgower> 0

<Glenda> +1

<laura> +1

<JakeAbma> 0

<Azlan> +1

<ben_tillyer> +1

<Jennie_Delisi> +1

<Rachael> +1

<jtoles> +1

<Jen_G> +1

<dan_bjorge> +1

<tburtin> +1

<ToddL> +1

<ljoakley> +1

<dan_bjorge> (at least +1 to including "be able to turn it off", not sure "not conveyed through ... alone" is necessary if "not conveyed through audio alone" and "not conveyed visually alone" are already separate reqs)

RESOLUTION: Keep the audio-shifting outcome, and add "Important information is not presented in spatial audio alone." and "Important information does not require depth perception."

kirkwood: is this the right space for the dangers of diziness and spatial audio?

alastairc: yes
... (this is probably platform right now though)

Multi-step process

https://github.com/w3c/wcag3/discussions/60/#discussioncomment-9131447

Rachael: delijennie-a11y's comment
... i think we should postpone this until we get into more depth, but wanted to bring this up to the group first

Jennie_Delisi: in response to julierawe's comment
... i'm fine if the group decides to remove it, but i'd like to discuss her comment first

julierawe: wanting to keep multi-step process in the conversation for now so we don't loose it
... i'm not sure if we do need to split it, but i don't want us to loose the multi-step process stuff

<kirkwood> +1 to keeping the multistep process. making a distinction to context. timing and keeping context and current location

dan_bjorge: i think it makes sense to split them

<alastairc> Poll: Keep multistep-process for now, and discuss further when we get to those outcomes. Please + / - 1

<Chuck> +1

<ToddL> +1

<mike_beganyi> +1

<julierawe> +1

<dan_bjorge> +1

<jtoles> +1

<kirkwood> +1

Jennie_Delisi: aria current isn't sufficient for this,

<Zakim> mbgower, you wanted to say do we need to emphatically differentiate levels of multi-step processes? difference between key combinations and a multi-'page' process?

<kirkwood> 2nd one. ‘mulit page’ process

<mbgower> scribe: mbgower

Ben: To build on what Mike was saying, in two-factor authentication, we need to be careful of how we define the scope of what constitutes a multi-step process

Alastair: We may be getting in the weeds. Coming from 2.x, it covered multi-page processes, and that's my assumption of our starting point.

<Chuck> +1 keeping it

Alastair: We are keeping for now.

<kirkwood> +1

RESOLUTION: Keep multistep-process for now, and discuss further when we get to those outcomes.

<kirkwood> +1

Alastair: We've been talking about this for 30 min. Last call for this PR
... I encourage you to look through and make any other comments

Rachael: The framing language is going to be important to this. One is in the PR. One is in the Editor's Note above the guidelines.

<alastairc> https://raw.githack.com/w3c/wcag3/guidelines_update_april_24/guidelines/index.html#issue-container-generatedID

Alastair: Yes, that framing language is important for scope and context.
... The other main focus was text alternatives
... We were working in subdocuments. That information has been pulled into the main document.
... I also tried to tidy up the 'top bits'
... We're looking at the Image alternative outcome

[Reads from Draft outcomes]

<kirkwood> covered -> described

<alastairc> Image alternative: Images have text alternatives, unless the image is part of an operable control and the alternative is covered by the name of the control.

Alastair: Does the new text improve the division between images and controls?

<jtoles> alternative is covered by the name of the control -> alternative provides a name for the control

Ben: I feel like if the alternative to the image is provided by the name of the control, that's just another way of saying the image is decorative

<jon_avila> What does this mean for decorative images?

Gundula: I do not agree that an image is the correlative.

<laura> +1 to Gundula

Gundula: I don't feel it applies to icon buttons.

Rachael: Is the intention for this to be scoped for controls?

Alastair: The intention was to scope out images whose purpose is to provide the name of a control. A search icon, for example, is going to have a name of "search".

<jon_avila> Or adds to the name of the control

Dan: +1 to what Gundula was saying. The second part seems redundant.

Rachael: We're all circling on the same question.

<GN015> Suggestion: Image alternative: Images have text alternatives, unless the image identifies an operable control and the alternative is provided by the name of the control.

Ben: Dan said it eloquently, but to try to reword... All images have to have text alterantives, unless it's in a button. But that doesn't make any sense to me. We'd still want that to have an alternative.
... If this was to pasted somewhere without the knowledge of the separate scope, that context would be lost.

<GN015> What about this suggestion: Image alternative: Images have text alternatives. If the image identifies an operable control, the alternative is provided by the name of the control.

Alastair: The decision tree used to be 'is this in an operable control?'
... Dan commented on this. I was trying to tackle this, and the only way I could see to do this was if an image was providing a name for a control.
... We have images that are in controls, images that are more than just the control, and images that are the control.

<jon_avila> +1

<kevin> +1

Rachael: Do we start with all images?

DJ: COuldn't we just say something as simple as 'unless the image overlaps with the name of the control?'

Gundula: Could you explain why we need to separate the 3 cases?
... I understand that the failing is that if the image is an operable control, it describes the function not the image

Alastair: The main reason is we have a separate outcome about naming controls. So you can be passing and failing two different things.

<ben_tillyer> I read this as “images have text alternatives unless in this single use case where a text alternative name exists”

Alastair: There's the between case, such as infographics, where you might be interacting with the images
... where they provide both information and controls
... I wonder if the decision tree doesn't start with 'is it an operable control'?

<Rachael> +1 to reorganizing hte decision tree

Rachael: I agree we need that branch of the tree. I just think we can move it down.

Alastair: Maybe around "important", since if it is functional, it is important

<ben_tillyer> +1 to this new branch

<Zakim> Jennie_Delisi, you wanted to talk about complexity and relationships

Jennie: I wanted to make sure we don't lose one of the pieces that Laura Carlsson brought up. In those complex images that have both operational and static images, we have to ensure the decision tree and descriptors can carry that key piece of relationship.

Alastair: Do you know what subgroup that was in?

Jennie: It was in a group that Chuck was leading covering complex images like infographics and graphs.

<alastairc> Poll: Keep the former (simpler) outcome text of "Images have text alternatives", and adjust the decision tree to put controls lower down.

0 need to thnk about it more

<ben_tillyer> +1

<Rachael> +1

<dj> +1

<jtoles> +1

<Azlan> +1

<julierawe> +1

<dan_bjorge> +1

<mike_beganyi> +1

<giacomo-petri8> +1

<GN015> +1

<kevin> +1

<jon_avila> +1 assuming we define images and text alternatives

<JenStrickland> 0 same as Mike, as images may not have text alternatives in some situations

Alastair: This document is open for comments if anyone has more to add. But it seems to have agreement.

<Jennie_Delisi> 0 same as Mike

Alastair: Jon is asking about definitions. A nice segue

<ToddL> 0

<tburtin> 0

<laura> 0

[Goes to Definitions heading in document]

I believe we use the terms in the Definitions in the decision tree.

scribe: I haven't updated the Test Procedure

The "Draft (very draft) methods" I thought might be useful for discussion.

Alastair: Moving to Discussion Questions, we've discussed the outcome [strikes out first bullet]
... The draft is open for comments
... Gregg added a "Future Note" 'already programmatically determinable'
... I had assumed we would handle AI-provided alternatives inside the decision tree.
... Points to second broze method "relying on user-agents..."

<Poornima> scribe: Poornima

WCAG 2 Issues

<mbgower> https://github.com/orgs/w3c/projects/56/views/1

mbgower: we put out another round of changes last week
... there are 12 things added in, and the wcag approval which is 6th on it
... we tried to address those in the next meeting. one item we are discussing # 3734

<mbgower> https://github.com/w3c/wcag/commit/4c06779f241ee04db5465700f8eb8284de44ea8b

<alastairc> Apologies, I should have said more clearly: This is the WCAG 2 content now, we have finished on WCAG 3

mbgower: this direct commit was modified after discussing
... if anybody already reviewed this, that commit is on the record.
... other than this, 11 others brought to attention and got thumbs up
... we are looking for people to go through PR's and do thumbs up
... for # 3642 there is email going through, i put a direct link to go through this draft proposal response

<mbgower> https://github.com/w3c/wcag/wiki/WCAG-2-Task-Force-process#errata-changes

mbgower: another thing is we have been putting some things on pending or normative in the project board
... basically we are pulling guideline from editorial wcag erata document. Two types of editorial changes in wcag 1. no text is changed (say link is broken) 2. fix the typo
... other thing is 'substantial change' which clarifies or changes the context or some add more clarification
... 3 kinds of scenarios 1. understanding document/sufficient techniques 2. normative changes/documents (actual PR itself) 3. Introductory (say in published documents)
... i just put in a matrix in the task force page
... informative parts of normative docs editorial will be listed in erata and more changes goes for call for consensus in the workgroup
... normative parts will always go for call for consensus in the working group
... long story short, errata - except for informative docs, any change is going to be inside that annually published area except the broken links for an example

This is scribe.perl Revision VERSION of 2020-12-31
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/danges/dangers/
Succeeded: s/loking/looking/
Succeeded: s/images that in controls/images that are in controls/
Succeeded: s/docuemnt/document/
Succeeded: s/thank you!//
Default Present: ToddL, Francis_Storr, rscano, kevin, Jennie_Delisi, dj, JakeAbma, mike_beganyi, jtoles, ben_tillyer, kirkwood, mbgower, JenStrickland, giacomo-petri, shadi, Jen_G, Azlan, sarahhorton, Chuck, tburtin, ljoakley, Laura_Carlson, Glenda, Poornima, Frankie, Rachael, alastairc, GN, Nayan
Present: ToddL, Francis_Storr, rscano, kevin, Jennie_Delisi, dj, JakeAbma, mike_beganyi, jtoles, ben_tillyer, kirkwood, mbgower, JenStrickland, giacomo-petri, shadi, Jen_G, Azlan, sarahhorton, Chuck, tburtin, ljoakley, Laura_Carlson, Glenda, Poornima, Frankie, Rachael, alastairc, GN, Nayan, giacomo-petri8, GN015
Found Scribe: dj
Inferring ScribeNick: dj
Found Scribe: mbgower
Inferring ScribeNick: mbgower
Found Scribe: Poornima
Inferring ScribeNick: Poornima
Scribes: dj, mbgower, Poornima
ScribeNicks: dj, mbgower, Poornima

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


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

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.)