Silver Task Force & Community Group

26 Jun 2020


jeanne, Fazio, Rachael, Francis_Storr, OmarBonilla, crispy, CharlesHall, bruce_bailey, Lauriat, Jan, kirkwood, Joe_Cronin, JakeAbma
Bruce, Chris, Peter, Angela
Shawn, jeannezakim, take up


<jeanne> chair- jeannezakim, take up

<jeanne> chair+ jeanne

<jeanne> chair: jeanne, Shawn

<CharlesHall> can I expatriate Earth?

<scribe> scribe: sajkaj

subgroup updates (only changes from Tuesday)

js: Any updates not provided Tuesday?

<CharlesHall> FunctionalNeeds has update

df: Maturity Modeling has planned first conversation 8AM Pacific next Friday 3 July
... Probably will move to Wednesdays for recurring schedule

js: Jan for clear lang?

jm: Have been comparing functional needs we gathered with COGA draft docs and working on completing the template
... So making progress

js: Anyone else?

new proposal for scoring (Rachael)

subgroup updates (only changes from Tuesday)

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

ch: Meet yesterday on Functional Needs and have doc ready for wide group review

<CharlesHall> The scope of the Functional Needs Subgroup is to draft and achieve a standard consensus based list of the core and intersectional needs of people and include at least the following:

<CharlesHall> Definition of (1) functional needs and (2) functional outcomes that can be referenced by WCAG3 as well as other W3C groups and documents.

<CharlesHall> Editorial method or style guide for describing a functional need.

<CharlesHall> Consider and account for needs that have been identified in and by: current policy and regulatory documents (globally); current W3C documents; existing research; and public comment.

<CharlesHall> Identify where functional needs are or may be unique to a specific context (such as while moving) or specific technology (such as an XR environment).

<CharlesHall> An evergreen document governance model that allows gaps identified by new research to be quickly added to the master list in a manner that facilitates informing all inheriting and derivative documents.

js: Wow! Do you think this becomes a separate note?

ch; Not yet discussed

ch: Looking for feedback here on the goals

js: Ambitious -- may need prioritization

df: Thinking Maturity Model can bring some of this together into a coherent structured map -- suggest we stay in touch on that

js: Functional Needs will be integral to all our work
... Also think it's well ordered

<CharlesHall> A statement that describes a specific gap in one’s ability, or a specific mismatch between ability and the designed environment or context.

js: Definition will be important ...

ch: Example follows in doc

<Fazio> sounds a lot like The Who definition of disability

<Fazio> WHO

ch: Nearly all are written as "use with" or "use without"

<kirkwood> well done

rm: Want to get this to a wider audience--very exciting

<Lauriat> +1

ch: So far want feedback only on goals

rm: Agree, but even that should have wider feedback

js: Also want to suggest weekly meeting

ch: We're going to communicate more frequently via our own list

df: Did I miss the meeting announcement?

ch: Don't recall how sent

js: Is call info on wiki page?

<CharlesHall> https://www.w3.org/WAI/GL/task-forces/silver/wiki/Functional_Needs_Subgroup

<CharlesHall> https://www.w3.org/2017/08/telecon-info_a11y-functional

<CharlesHall> alternate Thursdays

js: Very delighted by all this
... Anything else on functional needs? ...

new proposal for scoring (Rachael)

<Rachael> slide deck: https://docs.google.com/presentation/d/1IceTYOyGitApczya4vat4gPk9_I-7hIwpqTtXGusEZk/edit#slide=id.g8208eb709f_2_92

js: Rachael has been working on this ...

rm: trying to share screen ...
... Recalls bringing adjectival approach to silver some weeks ago ... has overlap with Jake but different format/scope
... Looking for feedback
... DISCLAIMER: Have put wording in, even if it isn't fully appropriate -- just to have something to work with -- no means final!
... Two definitions for concepts we've not yet named
... functional categories
... Scope
... whatever it is, page, sampling, task, etc etc
... a word applied to some functional category
... we need tight definitions, tight as possible
... so far it's just what a score sheet might look like
... I used 0-5
... believe 0 level is psychologically important
... will talk about what each of these "look like"
... put in excel
... off link 4
... top is functional categories
... below that is what's currently like guideline but also outcomes
... a score is based is not percentage--only if 100%
... basic is programmatically determinable
... that's level 1
... next level up is manual test with easy judgement call
... is image decorative or meaningful
... manual test with quality assessment
... e.g. does alt reflect actual meaning
... 3 about current wcag
... above things that take more to meet, may need to consider affordances
... does alt use plain lang

rm 5 is user testing with at

<jeanne> +1

rm: pausing to ask whether people are tracking?

<Chuck> +1

<bruce_bailey> +1

<kirkwood> _1

<kirkwood> +1

<Jan> +1

rm: next two slides drill down for examplaryapplication
... not attempting to be a complete list, but indicative

rm; you'd have to pass all 1 to get a 1

rm: similar for 2, etc
... [walks example with images contrast]
... next clear written content ...
... key is the scoping--however one scopes, all tests within each area must pass
... sees several challenges ... but does get to a final score
... to one decimal point
... equates 3 to bronze
... think would need a functional score in every category to avoid bias -- so req for 3 or greater in each functional category

<Zakim> bruce_bailey, you wanted to check that levels and score is the same thing

bb: looks great -- want to ask levels and score, are these interchangeable words

rm: pretty much so

<Zakim> Lauriat, you wanted to speak to the maintenance and score changing when we add new things…thing.

sl: wanted to ask about score change -- when new changes are added
... less concerned because conformance claim will apply to a particular version
... but wonders about test levels themselves
... makes sense to describe the levels
... but how to handle the guidance that doesn't have tests at specific levels?
... e.g. you have tests at 2 and 5, but not others
... if pass 5 would pass all -- but there's no req there be a test at every level

rm: yes
... Tried to make each test group about the same value -- as what JF has been asking for
... believe problem is creating the tests and decideding how to get everything balanced

sl: so if one category didn't have a level 1 test, you'd have to pass 2 of that category and at least 1 of the other categories

rm: yes

ca: seems simple and easy -- where did the complexity go?

rm: believe into creating the task guidelines

sj: and maybe into scoping?

rm: yes

js: Thanks! Fits in with some that we've been doing with Jake
... If level 5 test scores 3 good ---- are all levels the same -----

rm: a level 5 test would never score a 3; each level tests are modular, you pass or you don't

<Fazio> I thought WCAG 3 didn't want pass or fail

rm: the tests themselves are binary; probably a weakness of this approach

js: believe we could add some complexity and not undermine ...
... how to handle multiple instances that score differently?

<CharlesHall> I recall a goal of Silver was simply that pass/fail was not the only method – not that we wouldn’t use it at all.

rm: focusses fixes first on auto testable, then on up
... I tried making these percentage based and it just got crazy

<Lauriat> (since Rachael already spoke to it)

js: think a weakness is that it doesn't meet our goal of giving people a better understanding of how accessible they are
... perhaps we need to revisit that discussion

sl: it does help, though
... so test level 3 equates to wcag 2.x, and it does go above that

<Chuck> +1 to Shawn

<Fazio> I was hoping fitting all these components in a maturity model builds a roadmap for continuous improvement by highlighting a11y weak points, identifying appropriate dev teams to address it, budget etc

<Fazio> This along with Jake's would coincide

sl: provides a framework to express going above what wcag can do today

rm: would take you to a report like the excel I just posted
... one advantage is to show improvement over time; also how it works for various categories

<bruce_bailey> +1 to what Shawn said about overall approach is not pass/fail even if tests are pass/fail

omar: like requiring 3 or greater for each category
... will help in procurement

<Jan> +1 to 3 or greater = bronze

<jeanne> +1 to 3 or greater = bronze

<Lauriat> +1

<Zakim> jeanne, you wanted to say that I like how you need to do level 5 tests with users in order to reach gold

js: also really like that you have to do user testing to get above a 4

<Jan> +1 to 4 or more requiring level 5 tests, including users

rm: looking back at pros and cons in her slides ...

df: is there some kind of compatibility for wcag 2.x? if you pass 2.x you're bronze?

rm: synergy, yes, but not 1 to 1 mapping; e.g. no clear lang in 2.x
... but if you already pass 2.x you're close

sl: don't think last two listed negatives are negatives, but they are intertwined
... until one has test, these are just granular tests
... but by defining scope and expressing test results of that scope; then you know how accessibility things are vis a vis the scope defined which is better than what we can do today

ca: think the legal beagles might find this more understandable

<Lauriat> +1

<Chuck> +1 to Jeanne

js: we do need to do some testing to demonstrate validity to ourselves; a next step

<OmarBonilla> +1 to this being more understandable for the lay person

df: would we be choosing this over jake or vice versa?

js: believe they're evolving in similar directions and we need to look at the diff and see whether they can be merged
... in any case we need to test before making determinations

sl: believe jake's approach goes beyond of what would be in conformance
... so don't see them as mutually exclusive

<Chuck> +99

js: I found it easy to digest

sl: yes

js: Notes we meet June 30 but not July 3

<Jan> Thanks, Rachael - this was a great presentation.

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: 2020/06/26 19:05:11 $

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 Fazio Rachael Francis_Storr OmarBonilla crispy CharlesHall bruce_bailey Lauriat Jan kirkwood Joe_Cronin JakeAbma
Regrets: Bruce Chris Peter Angela
Found Scribe: sajkaj
Inferring ScribeNick: sajkaj

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]