Minutes of Silver meeting of 27 November 2018

Formatted minutes:
https://www.w3.org/2018/11/27-silver-minutes.html


Text of minutes:

   [1]W3C

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

                                - DRAFT -

                  Silver Community Group Teleconference

27 Nov 2018

Attendees

    Present
           jeanne, Jennison, Charles, Makoto, KimD,
           AngelaAccessForAll, kirkwood, Shawn

    Regrets
           Luis

    Chair
           Shawn, jeanne

    Scribe
           jeanne

Contents

      * [2]Topics
      * [3]Summary of Action Items
      * [4]Summary of Resolutions
      __________________________________________________________

    Last weeks work was in this document:
    [5]https://docs.google.com/document/d/1aBoQ1HDindVnFk_7Ljp-whpK
    3zAiqAdgJxsgpqsNpgU/edit#heading=h.imldn9ebvi3e

       [5] https://docs.google.com/document/d/1aBoQ1HDindVnFk_7Ljp-whpK3zAiqAdgJxsgpqsNpgU/edit#heading=h.imldn9ebvi3e

    Jeanne: This is a brainstorming document
    ... I created a new document for the alt text example
    [6]https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0Le
    Bqhu9DudyWp_IwovWc0/edit

       [6] https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0LeBqhu9DudyWp_IwovWc0/edit

    Auto-WCAG rule: Image has an accessible name
    [7]https://auto-wcag.github.io/auto-wcag/rules/SC1-1-1-image-ha
    s-name.html

       [7] https://auto-wcag.github.io/auto-wcag/rules/SC1-1-1-image-has-name.html

    Jeanne: talking about Auto-WCAG test.

    Charles: When governance overlaps with scale, it becomes
    unmanagable to put them into the guidance.
    ... this example is very specific to HTML
    ... if we followed a format like this we wouldn't be able to
    keep up with the scale.

    Jeanne: This would be a part of a method

    Charles: But this would proliferate the Methods

    Shawn: This is what we are trying to work out with this
    prototype. We would want to have Methods be higher level than
    the tests. So we wouldn't want to have a comprehensive list of
    all the ways we can meet this particular guideline.

    Jeanne: David had an interesting idea that we could write
    Methods for existing technology, and then write one Method for
    for new technology that is more like a success criterion.

    Shawn: I disagree. That is just moving success criteria to the
    Methods level. The user need is in the Guideline, and the
    person who is creating a new method has to have their own
    burden of proof that they satisfy the user need.

    [some back and forth about this concept]

    Shawn: for the image alt text example
    ... the Google docs document uses images, it doesn't use an
    <img> tag. Everyone has access to the alternative text, but it
    isn't following any method normally used.

    Charles: So if this example isn't in a Method, how do they
    claim it without a catch-all Method.

    Shawn: A catch-all Method would perpetuate that the problem we
    have today. The onus is on the organization that uses this new
    Method.

    Charles: So how would they prove it?

    Shawn: If it is in a VPAT, they would say that they did
    something different and document it. They can write it in their
    documentation. If it is a legal case, they would have to put it
    in the legal document. What they do, depends on the context.

    Charles: Do they ever get incorporated back into the guidance?

    Shawn: Sometimes. In this example it would not, because the
    Google docs document is a proprietary technology that others
    don't use. If someone came up with an example that is a new way
    to add an image in HTML, then it would be included in the
    Silver Methods.

    Jennison: I like the idea of keeping it open, but what would
    happen if someone claimed conformance and someone else thought
    it would not be, what do we do?

    Shawn: We would have to sort it out. The Method might be valid,
    but we wouldn't want to include it. It could also be duking it
    out in court. Or it gets worked out in the accessibility
    community

    Jeanne: I would like (down the road) to empower the
    accessibility community to vote new Methods up and down and
    have them work out what is best.

    Shawn: That would be a valuable way to manage the scaling

    Charles: What about using a picture without an image element?
    Like a Kiosk

    Shawn: If they create a new kiosk OS that has a different
    lamguage, then they still have to meet the user need.
    ... we can't expect people in the working group to be expert in
    so many technologies

    Charles: Doesn't this raise the problem that Wilco raised about
    the cost of testing Silver?

    Shawn: Only for new technologies that aren't using existing
    Methods.
    ... that wouldn't mandate more work and cost, but it would
    allow them to create new methods if they had a reason to do it.

    [8]https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0Le
    Bqhu9DudyWp_IwovWc0/edit#

       [8] https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0LeBqhu9DudyWp_IwovWc0/edit

    Jeanne: What I wanted to accomplish by bringing up the example
    of the ACT rule, is that the more granualar the tests are ---
    not the Methods -- the better we can explore the edge cases.
    That helps us write better Methods and Guidelines, which helps
    keep the cost down.

    charles: I am concerned about the number of tests, and needing
    to number them.

    Shawn: I share your concern. Keeping it so it isn't 100 Methods
    for a given piece of guidance.
    ... I have less of a clear image in my mind about the tests
    ... the tests are hierarchical, and there are a lot of them.
    ... I don't have a good answer for that yet.
    ... I want to get more information as we work through the
    examples.

    Jeanne: I would assume that we are not going to give automated
    test information to any user who isn't working on automated
    tests?

    Shawn: We need to figure out the level testing that we have for
    each Method. We want it to be concrete enough to be useful.

    Charles: granular is more consumable, it is the scale that
    concerns me.

    Jennsion: this discussion is about how to display it -- the
    presentation of it all. We don't want to overwhelm people with
    information, but we want to give people what they want in
    chunks or drill down.

    Shawn: we need to think about how we want to give them the
    information and how they will find it.

    Jeanne: Please take a look at the Conformance Example of Images
    need Alternative text
    [9]https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0Le
    Bqhu9DudyWp_IwovWc0/edit#
    ... and start entering your ideas and thoughts for tests.

       [9] https://docs.google.com/document/d/1IK83Gfxz01zFWgNtaCWCL0LeBqhu9DudyWp_IwovWc0/edit


Summary of Action Items

Summary of Resolutions

    [End of minutes]
      __________________________________________________________