<ChrisLoiselle_> scribe:ChrisLoiselle

<ChrisLoiselle_> Chuck: Survey commenting will follow the presentations

<Chuck_> https://www.w3.org/WAI/GL/wiki/Meetings/Silver_Deep_Dive_2020-08

<ChrisLoiselle_> Chuck: Time and Details for Deep Dive link will documented soon.

<Rachael> https://docs.google.com/presentation/d/1zUqVZnSKEmQuRpd7aTLvVeI_A-IV0ly3qt3glqqpNBc/edit#slide=id.g8c4beda184_0_171

<ChrisLoiselle_> Rachael: Provides link to slide deck, first 10 slides will be presented, rest is background information.

<stevelee> F5

<ChrisLoiselle_> Rachael: Slide 2: Goals and Disclaimer.

<ChrisLoiselle_> The presentation tries to consolidate into a concrete example, based on discussion to date. Provides a starting point to decisions to be made.

<Fazio> community group can't get in

<Fazio> do we want to pause?

<ChrisLoiselle_> For disclaimers, Rachael has put stake in ground on cut offs for scoring on conceptual model. Pulled from 2.1 a/aa mapping and act tests as well as silver sub group work.

<SBH_> can you past the meeting id/password in the IRC?

<ChrisLoiselle_> Try logging into zoom then starting the meeting

<ChrisLoiselle_> Rachael: Documentation Hierarchy for conformance (slide 3) talks to Functional Need Categories, guidelines, functional outcomes, tests, tests have atomic, contextual, holistic categories.

<Rachael> Essential, Visual, Auditory, Speech, Motor/Mobility, Attention, Language & Literacy, Learning, Memory, and Executive

<ChrisLoiselle_> Functional needs were reviewed, Rachael pastes in what is within slide 39 of presentation.

<ChrisLoiselle_> Examples will be reviewed on slide 38.

<ChrisLoiselle_> Example on slide 3, Functional need : Visual, Guideline Visual Alternatives. Functional Outcomes: Group tests outside of conformance. Example: Provides text alternative for non text content. Then Rachael talks to atomic, contextual and holistic.

<ChrisLoiselle_> Holistic are tests that involve entire path. Contextual more difficult to meet (AAA) and atomic, Tests whose requirements can be applied across systems.

<ChrisLoiselle_> Chuck: Are the atomic automated? Or could they be?

<ChrisLoiselle_> Rachael: Most are detectable / automated to review warnings or failures but human judgement is needed.

<ChrisLoiselle_> JF: AAA and contextual comment around AA vs. AAA and the easier to meet. Evaluation of requirement may be harder to determine.

<ChrisLoiselle_> Rachael: Talks to contrast and testing.

<ChrisLoiselle_> JF: I would like to possibly remove AAA from contextual to larger audience review.

<ChrisLoiselle_> Team (Silver TF) to review that wording and overall wording of presentation

<ChrisLoiselle_> Steve: Does this cover ICT?

<ChrisLoiselle_> Note: Silver will review as a group, not just Chairs in regard to wording.

<ChrisLoiselle_> AWK: Does this apply to non W3C technologies, i.e. PDFs?

<ChrisLoiselle_> Rachael: It is meant to be a broad range of technologies that are address. Web content...

<Ryladog> +1 to AWK

<alastairc> This is the scope: https://www.w3.org/2019/12/ag-charter#scope

<ChrisLoiselle_> Definition of path is presented in slide 4 of presentation. Scope related definitions of Path and View.

<ChrisLoiselle_> Path: Single view of the complete series of views and the specific components and content needed to complete a task from end to end.

<ChrisLoiselle_> View all content visually and programmatically available without an interaction equivalent to loading a new page or state. Views include pages, states.

<ChrisLoiselle_> MikeP: I was involved in wcag to itc group. WCAG to ITC task force looked at non web software and documents.

<ChrisLoiselle_> We talked to view in various conversations and we couldn't come up with agreement.

<ChrisLoiselle_> Many others went over definitions of view, so it is hard topic to define in a robust way

<Mike_Pluke> https://lists.w3.org/Archives/Public/public-wcag2ict-tf/

<ChrisLoiselle_> Rachael: Could you provide a link to your work Mike P? Thank you.

<Zakim> JF, you wanted to ask if "view" and "screen" would be synonomous?

<ChrisLoiselle_> Conversations were around UI Context within June , July and August during that time frame

<ChrisLoiselle_> JF: Talks to use of screen as the terminology vs. view

<Fazio> is that XAUR transferrable?

<Fazio> screen may not translate to XAUR

<ChrisLoiselle_> Rachael: If we used screen, below the fold, are those counted? Items off screen for screen readers? Conceptual , back to view / screen, does that make sense from a high level?

<ChrisLoiselle_> Steve : Path is user focused. View is what the system is presenting one point in time.

<ChrisLoiselle_> +1 to Steve's point.

<JF> +1 to Steve's interpretation

<Ryladog> +1 to Peter

<ChrisLoiselle_> PeterK: Talks to the view discussion...the charter stated we couldn't change language. We had to give meeting to terms used. The restriction on 3.0 is less so. We should keep in mind the horizontal and vertical.

<ChrisLoiselle_> Steve: Agrees with Peter's comments.

<ChrisLoiselle_> Rachael: Talks to slide 5, Scope, four ways conformance. Path, View, Content, Component.

<ChrisLoiselle_> Conformance is defined for paths. Minimal conformance for views.

<ChrisLoiselle_> Details of conformance are talked to in slide 5

<ChrisLoiselle_> JF: does a view contain a complete path?

<ChrisLoiselle_> Rachael: A view might. I.e. a CNN article read through

<ChrisLoiselle_> JF: I.e. a single screen or path can be a view?

<ChrisLoiselle_> Rachael: Yes.

<Fazio> is path synonymous to processionals?

<ChrisLoiselle_> JF: Thanks.

<ChrisLoiselle_> AWK: What is the difference between component and content?

<Fazio> process not professionals

<ChrisLoiselle_> Are components not content?

<JF> perhaps "Editorial Content"?

<ChrisLoiselle_> Rachael: To discuss content vs. component in detail off line and re-word to separate further what is meant.

<ChrisLoiselle_> Rachael: Takes note to clarify process per David F.'s request

<ChrisLoiselle_> Gundula: Editing a row and interacting with a table, is it content or component?

<ChrisLoiselle_> Rachael: Goal is to provide the ability to a developer of the content vs. the interactive component to be tested.

<ChrisLoiselle_> A view is all of the tests.

<Zakim> AWK, you wanted to ask if all content on a view is part of a path? Can a path be defined that results in some content on a view not needing to be tested?

<ChrisLoiselle_> AWK: If all the content on a view is part a path, horizontal nav example to get to a page. Or 4 screens on a path, each screen would need to be reviewed entirely?

<alastairc> Sounds like, in current WCAG 2.x terms, you would define a set of pages and would review each page in scope.

<ChrisLoiselle_> Rachael: All 4 screens would need to be reviewed.

<ChrisLoiselle_> Stefan: talks to component being part of content or if they should be split out.

<ChrisLoiselle_> Rachael: Agrees, will be part of future meeting.

<ChrisLoiselle_> Rachael: Talks to slide 6, Scoring Process.

<AWK> You're going to walk through a specific example in the coming slides, yes?

<Zakim> JF, you wanted to ask about automated scans and dashboards in this model

<bruce_bailey> i feel like this approach lends itself to a dashboard well enouigh

<bruce_bailey> each functional category could be its own "dial"

<Chuck_> q/

<alastairc> When would you score things off the path? Also, for the "path", I thought it was defined as a set of views, but then you mentioned things in the footer not being on the path?

<Zakim> alastairc, you wanted to ask about "90% on the path", would you have tested things not on the path?

<bruce_bailey> slide 14, one failure in path, would be less confusing if at least one rating was 99% instead of 100%

<stevelee> scribe, steve

<stevelee> scribe: steve

<Chuck_> scribe: stevelee

PeteKorn: does this take into the account the 'spoons concept' (cognitive friction)

<alastairc> Would that be built into the higher level tests for context & user testing?

rachael: yes in that it takes a complete path in

but it averages

we need to discuss the agregation and averaging in depth

rachael: you will see better when I send the spread sheet out with which yo uca nexpereiment

PK: this model is a looking at subset paths not the majority

if any drops below metric than all we can say is there is a failure that is only for a path

think wee need to address for top down too

<Zakim> AWK, you wanted to ask how only 45% of instances in view 1 can pass and there are no failures?

<bruce_bailey> good point about testable/repeatable leading to bottom-up analysis

GN: unsure how it scales to software (data driven apps like stores). When underlying content changes to single data item might break the conformance

Rachael: tha tis another conversationwe need ot have

AK: need explanation of example with 45% of a view passed but no failure in path -

<Zakim> bruce_bailey, you wanted to confirm: Bronze is better than "Substantially Conforming" ? (slide 7)

rachael: this a good one

bruce: is 45% better than substantial?

rachael yes bronze is 100% - but these are arbitrary and could change

? should everything be automatable?

rachael - I do not thinks so but JK mention dashboards

scribe: imeant once you fill i nthe test results everyhting in the spreadsheet is autpmated

<kirkwood> JF/JK

scribe: i meant once you fill in the test results, everyhting in the spreadsheet is automated

SBH: thinking about SaaS with rapid updates - should we could we keep rescoring for conformance?

<Chuck_> +1 on Rachael's understanding

rachael: I think conformance statements are a snapshot but a11y testing can be ongoing

JF: agrees, assumes a level of automated CI/CD a11y testing and snapshot against a version

SBH: yes we are doing as a best practice, can't say how often we need to refresh conformance
... world is moving towards SaaS

rachael: another point to discuss

<JF> I also want to note that "testing" and "scoring" ARE different things

dmd: mental health seems like a functional requirment
... please explain "essential"

rachael: I put mental health in essential needs as appears in the new groups list which is slightly different as both groups ar small and hight risk

<alastairc> That risk will depend on the lower-level tests I think.

dmd: concerned about risk for possibitly of litigation around mental health

AK: sounds like paths must be defined very specifically - eg alternative routes to same next step on path where only one is on the path

ie someone might take a different path to that expected

<david-macdonald_> Essential, is # 1 in the list, with mental health as part of that... concerned about the ambiguity around it. Large opening for litigation. Hard to defend against because of the subjective nature.

df: included stress -> mental fatigue -> negative effects in Coga research. It is important to address but hard ot measure levels

rach: I copied the various source functional needs (slide 36)

ra: for example essential probably need to be own category so it has higher impact in agregation

is this the right approach - what is good, what is bad

chuck: survey will be to facilitate commentary and then deep dive 'til laug 11

jf: what present next week will not be radically different

Issue Survey https://www.w3.org/2002/09/wbs/35422/wcag21-issues-2020/

<sajkaj> Ciao!

question 1 - SC 1.3.5 Technique "Use type attribute on fields in simple login forms"? #504

stefan: please clarify what if required field?

<Chuck_> In cases where a field accepts two types of entry / two values for user purpose (as in combined user name/user email fields), it is valid to give either one or the other value or leave out the designation of input purpose altogether" only if not mandatory. We have updated the text of 1.3.4 "Understanding Input Purpose" to cover this situation.

ac: don't think is related. if a field could be name or email can be ambiguous so can leave? That's not if mandatory but if it matches the mapping

if it doesn't have a clear mapping it should not fail

alan in othr words can give one, other or neither

RESOLUTION: Accept response to question #1.

question 2 - labelled navigation landmarks should be added to sufficient techniques for 2.4.4 #619

<alastairc> Detlev's updated version: 2.4.4 is scoped to links rather than groups of links, and we have ARIA13 which is a technique for 1.3.1 info & relationships.

<alastairc> Unless AT support demonstrably improves to allow access to link context contained in a landmark label from a link that already has focus, we consider a landmark label insufficient to convey link context.

stefan: Unsure about Detlev's landmarks are exposed in screen readers and not sure how that works with links

AK: stikes me as similar as discussion using headings as part of context of a link whic hwere removed. I would prefer headings too if we use landmarks

AK should treat both consistently

stefan: landmarks usage is not clear to me - what can a landmark have ot do wit ha link?

RESOLUTION: Accept Detlev's response.

question 3 - 1.4.11 hover state clarification #849

<bruce_bailey> https://www.w3.org/2002/09/wbs/35422/wcag21-issues-2020/results#xq3

<bruce_bailey> https://github.com/w3c/wcag/issues/849

ac: think laura suggesting expansion to my brief response - no issues

RESOLUTION: Accept Laura's response.

question 4 - Clarity on 1.4.11 as it applies to "Cards" #856

<bruce_bailey> https://www.w3.org/2002/09/wbs/35422/wcag21-issues-2020/results#xq4

<bruce_bailey> https://github.com/w3c/wcag/issues/856

GN: think the srepsonse should refer to cards as the question does
... also think the understanding section could be improved

stefan: "card" should be clarified and relationship to link - thinks the original thing got a bit lost. is this about cards or about broader scope?

chuck: we could consider GH's broader solution so ...

GN: want to object as is on WCAG 2.1 as answer should be understanding to clearly state it for 2.1

AC: agree is 2.1 issue. But...1)need to respond 2) how update doc for 2.1 and 2.2. so need to look through text an dgood to have an example

suggest. Agree response and then someone creat PR wit hGN's text plus an example

ac suggest I add cards to response

<Chuck_> For links such as "cards" which have clear text and/or image content, non-text contrast does not require them to have a contrasting border or background, as the understanding doc states under boundaries:

stefan: so card is just a decorated link? Not a more complex container? Need to define.

ac: lets leave definition question used in repsonse

<alastairc> Also adding "We will look at adding an example to the understanding document to improve that paragraph."

RESOLUTION: accept response and ammended. Defer evaluating changing understanding document.

This is scribe.perl Revision of Date 
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/Chairs to review/Team (Silver TF) to review/
Present: Francis_Storr Rachael GN015 jeanne ChrisLoiselle_ shanew Raf Nicaise JF Fazio StefanS Mike_Pluke KimD PeterKorn bruce_bailey stevelee JakeAbma alastairc MichaelCrabb Laura kirkwood Katie_Haritos-Shea SBH_ Sukriti sajkaj ok
Regrets: Brooks_Newton John_Kirkwood Jennie_Delisi Charles_Hall Detlev
Found Scribe: ChrisLoiselle
Found Scribe: steve
Found Scribe: stevelee
Inferring ScribeNick: stevelee
Scribes: ChrisLoiselle, steve, stevelee

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.)