W3C

- DRAFT -

Silver Task Force & Community Group

16 Jun 2020

Attendees

Present
jeanne, JF, CharlesHall, janina, Joshue, Lauriat, Fazio, Francis_Storr, Chris_P, ChrisLoiselle, Chuck, michaelcrabb, kirkwood, OmarBonilla, Joshue108, Jan
Regrets
Chair
SV_MEETING_CHAIR
Scribe
Joshue108

Contents


<scribe> scribenick: Joshue108

Holiday schedule

SL: Holidays coming.

JS: Most people in US with take Fri 3rd off,

<Chuck> +1

<Lauriat> +1

Please +1 if you are in IRC

<michaelcrabb> +1

<jeanne> +1

JS: Will we cancel Tues 30th?

<CharlesHall> Friday is also Juneteenth in the US and some orgs are making it an official holiday.

<JF> +1 to being away on the 3rd

<ChrisLoiselle> Unknown on July 3rd. On vacation 6th through 10th of July.

MC: Yes if reduced participation, but not if only a few

<jeanne> Tuesday June 30?\

JS: Lots not available on Fri 3rd.

<Chuck> +0 Tuesday June 30th

Please +1 if you are not attending call that day

<JakeAbma_> 0

Tues 7th?

<jeanne> Tuesday the 7th?

<Chuck> +0 Tuesday 7th

<JakeAbma_> 0

<JF> -1 for Tuesday 7th - vacation

<ChrisLoiselle> +1 to Not being available July 7th.

JS: We could still meet.

<jeanne> Friday July 10

<Chuck> +0 Friday 10th

Friday 10th?

<ChrisLoiselle> Congrats JF!

JS: Ok July 3rd cancelled rest of meetings will go ahead.

SL: This Friday is also a US holiday I'll be off

JF: Lets ping other groups on this.

<jeanne> June 19th unavailable?

<Chuck> +1 June 19th

<Lauriat> +1

JS: June 19th - +1 if unavailable
... We should go ahead with that.

SL: Fine

<jeanne> Conclusions have no meeting on July 3rd.

Sub-group check-in

<Lauriat> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Main_Page#Sub_Groups

SL: Makoto emailed alt text status - no update, but new members

<ChrisLoiselle> No new update from Visual Contrast

JS: You meeting this week?

CL: Yup

<michaelcrabb> XR Update

Clear Language - more members meeting this week

Headings Sub Group

JS: Not actively working on this - changes to approach in conformance architecture

<scribe> New member from Intel

MC: We are going through XR User Needs doc and putting stuff together

Cross ref with functional outcomes

Happy to answer questions on GH vid

JanS: To suggest, when you have a write up - there is an active captions CG - they may have feedback

MC: Yes, defo - was approached by ImmWeb Captions CG

When we have this done we will share - via GH

GH === Github

JS: Q?

Resuming consensus check on Scoring Proposal

ISO Modelling

DF: Am working on this, sceduling meeting have participants etc

JA: Add me

Conformance architecture testing

JS: Jake update?

JA: Not really

reviewing content, looking if we can do atomic testing.

If that fits

WIll talk more soon - will start to test processes like search and login

And see how they fit

JS: Q?

(laughs)

Functional Needs

MC: Group getting set up

Meeting bi-weekly

Keeping group small at the beginning

Mailing list avail from group homepage.

CH: I've work in progress

JS: Are you using the FAST as a basis?

MC: It's one of several that we may pull together

JS: Jake is looking at that, likes it.

MC: I hope to tie the work of the sub group with Silvers
... I hope the outcome of this group will be used here and in Silver

Maybe standalone doc

We want them to be usable etc

JA: Lets discuss offline

MC: Makes sense

JS: Thanks all

Resuming consensus check on Scoring Proposal

SL: This is where we left off on Friday

<jeanne> https://docs.google.com/document/d/1l4ETg-VJepvbcINFta3exwX-DM29O4Q14XPpKkA170Y/edit#heading=h.ocfedtnacj7u

We established that we dont have concensus on conformance scope

<Lauriat> Sampling: Sampling is a method of achieving conformance, but it is not included in conformance.This is consistent with the way WCAG2 addresses sampling. The current proposal for Representative Sampling is in the Editor’s Draft. Silver Sampling follows the WCAG2 model and recommends that the rules of WCAG-EM adapted to include non-web and tasks. The principles of WCAG-EM can be applied for now.

SL: A way of sying rep sampling is not a part of overal conformace

but here is how you can use it outside of the guidelines

SL: Q?

JF: This has changed

<struggles with assertions>

SL: Which bit

JF: <gives example>

SL: Thats not it.
... Illustrates point being discussed

JF: Is working.. am concerned about scores of post launch

SL: This is rep sampling not post launch

JF: <further outlines his issue>

CA: I hear JF, am confused

<discusses measurements on component level>

JF: <gives sporting analogy>

JS: Can we stop with analogies?

They are confusing for some users.

JF: A conformance report is a summary - they can then use it for future work etc

<further illustrates reporting usage vs atomic scoring against a larger part>

SL: Please back to rep sampling

JF: <goes back to rep samling>

SL: I know what it is.

MC: IMO The reported unit of conf is tasks

doesn't line up to pages etc

I think rep sampling is a way of going further and should be encouraged

<Fazio> can it be a part of the maaturity model?

JS: JF descs where we want to end up, but what about the steps?

We are looking at the pieces here, starting at the end result is not helpful.

JF: <tries to understand>

SL: Same as WCAG 2 today

JS: There may be more than one definition and mechanism for conformance WCAG 3.0

we dont know yet

JS: Thank you Janina

<JF> +1 to Janina - have we explored other options?

JS: Jake working on other models.

I want to talk about this - and the sampling in that proposal may work in this way

We want to propose two tracks - a11y in use and tasks

Both will apply -

JanS: Thats my understanding

SL: Not mine - up for discussion

JF: < no concensus>

JS: Ok, lets talk about flow chart

I've adapted and flattened,

We can expand this

dont want to be restrictive - technical side in known to us

Info on functional needs can be passed thru the process

Entity reporting => Info available => Split out

How to => Standardised Questions

JS: Gives overview of new rating

Need to work out how to do usability testing economically

JS: A11y in use testing from easy to difficult

Each section should be defined

Does it meet criteria x or y and these qualities make it easy or difficult

Without hiring outside help can be averaged into a total score

<discusses failure equation>

Total scope => this.(bronze, Silver, Gold)

JS: Q?
... <recaps>

<gives semantic heading ~ structured/unstructured>

% of passed / number of instances

JA: Something to add.

Two things - 1 may not always be technical

2 some implementations may not be required but may be best practices and should be included

<discussed points subtleties>

JS: s/Total scope => this.(bronze, Silver, Gold)/TotalScope(score) => this.(bronze, Silver, Gold)
... We are still working on % passes for most things

<talks about exceptions>

CA: Re: Johns example..

<riffs on Johns example>

Is that how averages are worked out?

JS: We are trying to work this out with solid example

<gives though experiment example in order to normalise variable guideline outcomes>

JS: We need to capture the impact of failures
... So a score makes sense for a site.
... We want to try this out and see if it makes sense

<overview of previous funtional needs with real numbers>

A site may still pass - we don't want that.

and need to test with test and real data

SL Regarding Janinas point atomic tests and the outcomes

<Agreement between JS and SL>

JS: <Func outcomes of guidelines normalisation>

Needs to be normalised so statistically valid.

JS: We need to work out how to weight the guidelines in a way that is fair

Will look for details of consensus in minutes

When weighted by disability they tend to not be fair and introduce bias towards Cognitive users

All barriers are equal.

Lets talk Friday

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: 2020/06/16 14:30:09 $

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)

Succeeded: s/@@/post launch/
Default Present: jeanne, JF, CharlesHall, janina, Joshue, Lauriat, Fazio, Francis_Storr, Chris_P, ChrisLoiselle, Chuck, michaelcrabb, kirkwood, OmarBonilla
Present: jeanne JF CharlesHall janina Joshue Lauriat Fazio Francis_Storr Chris_P ChrisLoiselle Chuck michaelcrabb kirkwood OmarBonilla Joshue108 Jan
Found ScribeNick: Joshue108
Inferring Scribes: Joshue108

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: 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]