Silver Task Force & Community Group

30 Jun 2020


jeanne, JF, Chuck, CharlesHall, OmarBonilla, SBH, Lauriat, Joshue108, Fazio, Rachael, Jan, MichaelC, Francis_Storr, sajkaj, michaelcrabb, Crispy, JakeAbma
Shawn, jeanne


<jeanne> not having it

<scribe> scribe:ChrisLoiselle

Sub-group check-in

Jeanne: Subgroup check in. If you have updates, please let us know.

DavidF: We are meeting Friday at 9 Pacific. We'd need a zoom call meeting id.

SBH: We have extended members that want to contribute from many different companies.

<michaelcrabb> https://w3c.github.io/silver/subgroups/xr/captioning/what-is-captioning.html

Michael Crabb: We had a meeting yesterday. Here's the link we are working with for captioning and mixed reality.

Captions and audio descriptions are looked at two separate items in wcag, however that is difficult to do in mixed reality.

We are working toward user needs , hopefully publishing to github soon.

JF: Has there been any communication on research task force ? I.e. with Josh O'Connor?

MichaelCrabb: Yes, Josh is in the sub group. Great question. We've been working together and gathering his insights.

Jan: On plain language, we are making progress and meeting weekly.

CharlesH: Goals were written out for comment, I haven't seen any. Jeanne: Could you repost the link?

<CharlesHall> https://docs.google.com/document/d/1eJkgXqbh7dx3uD6XAy8XAANmwfbbVZ5GKb_gbsUdkVs/edit?usp=sharing

<Jan> I will review it

Scope: discuss two alternatives in Conformance Scope & Sampling from the Scoring Proposal Doc

<Lauriat> https://docs.google.com/document/d/1l4ETg-VJepvbcINFta3exwX-DM29O4Q14XPpKkA170Y/edit#heading=h.ombdu93ow71z

<Lauriat> https://w3c.github.io/silver/guidelines/#scope-of-conformance-claim

Lauriat: We've been talking on conformance on declared scope.

<Lauriat> “Conformance is defined only for Tasks. However, a conformance claim may be made to cover one task, a series of tasks, or multiple related tasks.”

An alternative proposed was to follow wcag 2.x conformance claims. Shawn pasting into IRC

Shawn: Replacing full pages, with a different unit, such as a task.

Chuck: Observation and question, with this change in conformance this scales to verticals such as PDFs ? Presumption that is one of the benefits.

Shawn: Yes. Task may not be best wording, but yes. I.e reading a PDF and understand a chart, for example.

<Lauriat> https://lists.w3.org/Archives/Public/public-silver/2020Apr/0035.html

Shawn: Pasting footnote / email on thinking of what task could mean. Follows ACT's format. Talks to atomic tasks and larger tasks.

<Zakim> jeanne, you wanted to ask Janina if this addresses the concern about scope with Challenges doc?

Jeanne: I'd like to have this commented on by Janina. Does this align with challenges document?

Janina: It may be a bit more limited than we want, but putting it on the table is a good thing.

<JF> https://www.w3.org/WAI/test-evaluate/conformance/wcag-em/

JF: Talks to a design system and scoping and concern of scaling. Do we mention wcag em? How does this impact regulators? WCAG EM talks to scoping as well.
... Using mega menu example that leads to a score, does that make the page or website conforming?

Shawn: We talk to task, not component level.
... Not a full site , a web page is the unit of measurement currently. A task is a better atomic structure. You could develop a task that covers everything in a website.

<jeanne> Scope from WCAG-EM: This methodology is designed for evaluating full, self-enclosed websites. That is, for every web page it is unambiguous whether it is part of the website or not. This includes websites of organizations, entities, persons, events, products, and services.

JF: If you have an accessible 3 step form, but you can't get to the form, you can't accomplish it. But if only scoped to the form, you can use the form, how would that be covered?

<jeanne> WCAG-EM: Principle of Website Enclosure. When a target website is defined for evaluation, it is essential that all web pages, web page states, and functionality within the scope of this definition are considered for evaluation.

Scoping and conformance claims need to match more closely.

Shawn: Real world tasks, end to end are what needs conformance.

<CharlesHall> so, we still need a definition of task that includes the path to and from the task.

Rachael: I react to the word of task, i.e. the action of being completed.

<SBH> At VMware, we call them workflows. Definitely close to the Path definition. And we disclose them in VPATs in the evaluation methodolgy section

Maybe we scope at the path level vs. task level. Beginning to end path. Tasks can be within that path.

<SBH> +1 to @ChrisLoiselle

<kirkwood> +1

<sajkaj> +1 to considering path

Shawn: I agree

<kirkwood> +1 to Racheal about path

Chuck: Flexibility and positives of this , it could encompass the entire user experience. Opposite, is that insufficiently scoped work could impact the end user negatively, per JF's examples.

<JF> An organization or author who want to make an conformance claim can select a logical sub-section of the site, application, or product.

<CharlesHall> in my experience, “action” is the lowest atomic unit, where a “task” would include {n} actions, and “path” includes the method to locate and begin the task, complete the task, and confirm the task is complete.

JF: Talks to his posting of text on claim conformance. That aspect is subjective. If the button that launches modal doesn't conform, how is that scored? Sub sections can be included as part of the site, but can't alone serve as a representative sample.

<Rachael> For consideration a draft definition of path: 1 view (everything on the screen) or the complete series of of views needed to complete a task from end-to-end. A single component is not considered a path.

Shawn: We are not stating that the conformance is at component level.

We are talking to my proposal that is not in the editor's draft.

<Lauriat> https://docs.google.com/document/d/1l4ETg-VJepvbcINFta3exwX-DM29O4Q14XPpKkA170Y/edit#heading=h.ombdu93ow71z

JF: Could you share that url again so I may review?

<Lauriat> Updated sentence: “Conformance is defined only for Paths. However, a conformance claim may be made to cover one path, a series of paths, or multiple related paths.”

SBH: Paths are very close to how we are defining things at VMWare. I do like paths, we call them workflows internally.

<Zakim> sajkaj, you wanted to talk about defining path

Janina: I do like paths

<Chuck> +1 to "path"

<CharlesHall> there is a “happy path” that the author scopes and intends, but then there is the “desired path” that the user actually takes, and they are rarely the same.

Janina: We'd have to capture path and what task means within path. I.e. more than one step here on achieving what you want.

<SBH> We then tag the paths as common and rare. Rare is error condition, Day 1 install / setup Day Omega shut down

Shawn: The full page / full context can be defined at guideline or test level.

<kirkwood> +1 to path

Janina: mapping out multiple paths would be helpful.

<CharlesHall> +1 to multiple paths

<Lauriat> +1 to multiple paths

JohnK: I was using paths terminology in order to put in standards in dept of education. In the legal accessibility realm, this helped those individuals.

<Zakim> jeanne, you wanted to say that regulators are not inherently opposed to tasks. The US airlines accessibility includes accessibility of tasks.

<kirkwood> very strongly agree with path terminology

<CharlesHall> the description of the cognitive walkthrough method we have referenced in the past includes “A task list that includes all the tasks that you will use in the walkthrough, as well as an action sequence that details the specific task flow from beginning to end.” it also includes “Determine what tasks and task variants are most appropriate for the walkthrough.”

<KimD> +1 - you'd fail that test

JF: Scoping and avoiding testing for conformance. JF talks to findable help. I.e. if findable help is not present , do we avoid this altogether? Do we fail it if they don't have it?

<CharlesHall> +1 to JF for atomic + collected

Rachael: We don't have to have the same scope for each SC.

<sajkaj> +1 strongly to flexibility in scoping

<Zakim> MichaelC, you wanted to say we can´t structurally avoid possibility of gerrymandering

Chuck: to JF, is the scoping and conformance to evade conformance the worry?

<KimD> +1 to Michael - can't control everything

<Lauriat> +1 to Michael

<jeanne> +1 to Michael

<Rachael> +1 Michael

<sajkaj> +1 to Michael. It's not our job to play police or judiciary

MichaelC: we can't control that and shouldn't constrain ourselves to do so..

<Francis_Storr> +1 to Michael

<Zakim> JF, you wanted to say it's not about tests, it's about reporting

JF: Its not about tests, it is about reporting. I.e. health of product. I think that is what regulators would look at too. I think scoping conformance reporting is a different topic.

How do we test for conformance that is testable and repeatable?

<Lauriat> +1, I don't see that in the Requirements for Regulatory Environment: https://w3c.github.io/silver/requirements/index.html#regulatory-environment

<Zakim> jeanne, you wanted to say that SIlver does not have a goal around reporting.

<Zakim> MichaelC, you wanted to say conformance reports can be challenged, but not our role to constrain

Jeanne: We don't have a goal around reporting.

MichaelC: We do want repeatable , but we can't capture all edge cases. We can define coformance and allow for testable and repeatable items. The challenges are then put toward the implementer to meet it.

<Lauriat> +1

<JF> clear intent and transparency as to purpose and goals, to assist when there are questions or controversy.

Shawn: Talks to how does a company do that now with web pages?

JF: To scaled reporting vs. pass / fail , we don't have a concept of silver / bronze or gold rating or scaled report i.e. 70 percent. What does the score mean? 3 paths on a website or the website itself? What is the scope? What does scope mean?

Shawn: We are saying the score that you claim is what is stated in your conformance claim. I.e. this is our scope and this is our claim based on scope

JohnK: On conformance claim, scoring and regulators. Regulators are looking for relative progress. Real human beings are better than scoring and how they can benefit from regulation. So if scoring is standard, that is what people are looking at.

<Lauriat> +1 to kirkwood, we provide the mechanism of measurement & transparency

I.e. it is not binary. It is progress over time and true access for people.

<Zakim> MichaelC, you wanted to say regulators can call foul on a scope

MichaelC: We need to define what a conformance claim is and what a scope is. We can't predefine all scopes. The web is too large to do that.

<Zakim> Chuck, you wanted to ask if we can do a straw poll?

<kirkwood> agree on the score

Chuck: Scoring is beneficial to executives. Simple is good.

<Lauriat> Updated sentence from before: “Conformance is defined only for Paths. However, a conformance claim may be made to cover one path, a series of paths, or multiple related paths.”

<Fazio> +0

<JF> -10

<SBH> +1

<jeanne> +1

<JakeAbma> 0

Chuck: To using paths for scoping, let us take a straw poll, do people agree with that statement?

<Chuck> +1

<Crispy> +1

<Lauriat> +1

<kirkwood> +1

<Rachael> +1 if we define path to include a single view

<CharlesHall> +1 if path is defined and variable

<JakeAbma> -0.1

<Lauriat> +1 to Rachael, I think that works fine.

Rachael: I think a valid path is you read a single page. A single instance of what you see on a page. That should be included in the word path's definition.

<KimD> +.5 - conformance *could* be path plus something or multiple paths

CharlesH: There is always more than we anticipated. I.e. scoped for happy path, but what the user took is "X" path
... Testing with people would need to be done, qualitative or quantitative , as to what people have done.

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version (CVS log)
$Date: 2020/06/30 14:30:29 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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)

Present: jeanne JF Chuck CharlesHall OmarBonilla SBH Lauriat Joshue108 Fazio Rachael Jan MichaelC Francis_Storr sajkaj michaelcrabb Crispy JakeAbma
Found Scribe: ChrisLoiselle
Inferring ScribeNick: ChrisLoiselle

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: 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]