Minutes of Silver meeting of 29 January 2019

== Formatted minutes
https://www.w3.org/2019/01/29-silver-minutes.html


== Text of Minutes

    [1]W3C

       [1] http://www.w3.org/

                                - DRAFT -

                  Silver Community Group Teleconference

29 Jan 2019

Attendees

    Present
           Lauriat, KimD, AngelaAccessForAll, kirkwood, jeanne, JF

    Regrets

    Chair
           Shawn, Jeanne

    Scribe
           jeanne

Contents

      * [2]Topics
          1. [3]Requirements
          2. [4]Accessibility Supported
      * [5]Summary of Action Items
      * [6]Summary of Resolutions
      __________________________________________________________

Requirements

    <Lauriat>
    [7]https://w3c.github.io/silver/requirements/index.html

       [7] https://w3c.github.io/silver/requirements/index.html

    <scribe> scribenick: jeanne

    <Lauriat>
    [8]https://www.w3.org/TR/WCAG21/#identify-input-purpose

       [8] https://www.w3.org/TR/WCAG21/#identify-input-purpose

    There is a thread on the AGWG about what is a failure condition
    for 1.3.5 Identify Input Purpose

    Shawn: Many solutions are "postpone to Silver" and there is
    good discussion about Silver and the conformance model
    ... I hope it makes it easier to include some of these issues
    as requirements for Silver.
    ... How far are we from going back to the working group to
    address the Requirements

    Charles: two questions for Silver: Does Failure apply to
    Silver? Yes - it is a score of zero
    ... the other is does Accessibility Supported have a zero? I
    don't know that.

    Jeanne: The difference between Programmatically Determined and
    Accessibility Supported is that Programmitcally Determined
    means it is coded to standard, and don't have to have an
    assistive technology, Accessibility Supported means that the
    content developer is responsible for finding a solution, even
    if the assistive technology doesn't support it.

    JF: I agree with JEanne. 1.3.5 is an important example.

    Shawn: IN terms of saying which tests to run for a different
    task, you look at the Methods available to meet certain
    guidance.
    ... for example, if there is no method to set a default
    language in a VR town, then this doesn't apply.
    ... It highlights that the VR platform needs to support default
    language, not that the content developer has to find a way to
    do it.

    JF: Currently with Flash and WCAG, the author is responsible
    for finding a way to do it.

    Shawn: The thread in 1.3.5 in AGWG provided some suggestions of
    how AGWG and Silver could address the problems in 1.3.5 and it
    gives an opportunity for AGWG to relook at the Silver
    requirements.

    jF: Not sure there is consensus in AGWG

    Shawn: We want to make sure that the overall working group
    agrees that these are suggestions we should investigate.

    <Lauriat>
    [9]https://w3c.github.io/silver/requirements/index.html#oppotun
    ities_conformance

       [9] https://w3c.github.io/silver/requirements/index.html#oppotunities_conformance

    Shawn: These are not included in the formal Requirements het
    ... I think some of these are pretty solid, but I want to do a
    sanity check.

    Jeanne: We should also review Detlev Fischer's comments,
    because he had comments that talked about how Germany handles

    <Lauriat> Github issue started by Wilco:
    [10]https://github.com/w3c/silver/issues/41

      [10] https://github.com/w3c/silver/issues/41

    Jeanne: some of these issues

    JF: I will discuss this at Deque

    Jeanne: I will reach out to TPG and LEvel Access to see if they
    would accept current testing for for Bronze level and usability
    accessibility accessment for SIlver and Gold level.

    Shawn: We can test mechanically for alt text, but we also need
    the cognitive assessment of the quality of the alt text.
    ... In the Opportunities section, is there a better way we can
    express the direction we want to go to and the problems we want
    to solve.
    ... I want to say that these are the problems we want to solve
    instead of writing Requirements of how to do it.
    ... for example: In Silver we want to include more measurable
    outcomes, like alt text quality. The end of the day, the user
    has a lousy experience, so how we do we measure that?

    JF: But how do we do legal conformance?
    ... How do we measure quality in a legal conformance.

    Shawn: We already have this problem. WCAG says "that serves the
    equivalent purpose" which is undefined.

    JF: That is the big problem.

    Shawn: We want to improve the problem. We want to give a way to
    express that in COnformance. Today it is pass fail, We could
    have "it passes, but its a lousy experience." We want to
    express that.

    Jeanne: We could do Bronze is WCAG AA, and the measurement of
    quality is in SIlver and Gold level.

    JF: What's the difference between Silver and Gold?

    JEanne: We have roughly talked about using internal evaluation
    for Silver and PwD with Gold.

    Charles: IT gives a qualitative response, so that helps improve
    where we are today.

    JF; I'm interested in a testing terminology that Charles talked
    about is a Cognitive Walkthrough.

    Charles: It's not the number of users testing, it is the
    variety of human functional needs represented in the testing.

    JF: I don't see that scaling -- it keeps out the smaller
    companies -- only the largest companies.

    Cyborg: It creates a demand for a new kind of business for
    people with disabilties for testing.

    Shawn: It's still valuable to get more usability testing to get
    to higher levels in Silver. I want to know if AGWG will agree
    that this is a valid need that Silver needs to address.

    JF: Which brings us back to 1.3.5 - if they used microdata they
    could get to bronze, but couldn't get higher because it doesn't
    exist. But if they used @autocomplete, they could get at least
    to Silver.

    Shawn: Yes, it gives a way for people to express conformance
    when they are using innovative techniques, or you can express
    that they tick the boxes, but have a terrible user experience.

    JF: WCAG 2 tried to improve that. But it has limits. For
    example, alt="alt text" is not accessible.

    Shawn: The example of a meme, does the alt text describe it.
    ... I'm not concerned with the details now, because I want to
    get input from more people who have more detailed experience.
    ... up until now, we have gotten so bogged down with AGWG in
    the details of how to do.
    ... let's clean up the Requirements, make sure that we
    highlight the functional needs that we want to include in
    Silver.
    ... do you think we could bring it to AGWG?

Accessibility Supported

    Kim: HOw are we wholistically addressing conformance

    Shwan: The author makes a site, it works beautifully and then
    JAWS does an update and it's broken.

    Kim: The example of a site that can't work right because of a
    browser issue. Today, WCAG requires that it is the author's
    problem.
    ... an example of making it working for a mouse but not
    keyboard because it causes a security breach.

    Shawn: For Japanese users, the screen reader doesn't support
    ARIA, so the end users don't have an option.

    Kim: What do we do about it?

    Jeanne: We had part of the Methods prototype. The language of
    page example had VR platform that said it was the platform
    problem.

    <Charles> have to drop for another meeting

    Shawn: That doesn't address the problem that different Methods
    may work better than others.

    JF: The person using a WYSIWYG editor is an AUTHOR, the
    developer is the CREATOR.

Summary of Action Items

Summary of Resolutions

    [End of minutes]