<Chuck> meeting: AGWG-2024-02-20
<laura> scribe: Laura
Chuck: Welcome.
... Chuck: Any new members or roles?
... (none)
... March 6 will be having an on boarding class.
... will have announcement on list.
<Chuck> https://github.com/w3c/wcag3/discussions/51
Chuck: will be having requirements updates and review card sort next week.
<Rachael> https://docs.google.com/document/d/1tpgJaoIHC4IywvFkfXrfa8mEcSlEtAP-P0_46vNqynY/edit
rm: 2 versions.
<Rachael> https://docs.google.com/presentation/d/1p2HBgjatnLQBxeSMSFPPiafy91K3jXj5sCGYxaJystY/edit#slide=id.p
rm: google doc and slides.
... take a moment to go through structure.
... Guideline: High-level, plain-language content used to
organize outcomes.
... Outcomes: Verifiable statements that allow testers to
reliably determine if the content being evaluated satisfies the
user needs identified in the Guideline.
... What/Where/Why Statement: Proposed way of writing an
outcome or supplementing an outcome written as a statement.
Alternative way of writing based on EN 301 549 “Conditional
Clauses/Requirement/Notes.
... Method: Detailed information, either technology-specific or
technology-agnostic, on ways to meet the outcome as well as
tests and scoring information.
... Example information: Introduction, description, examples,
tests, resources
... Tests: Mechanism to evaluate implementation of a
method.
... Techniques: Technology-specific approach to follow a
method
... Assertion: A formal claim of fact, attributed to a person
or organization. An attributable and documented statement of
fact regarding procedures practiced in the development and
maintenance of the content or product to improve
accessibility.
... User Needs: The end goal a user has when starting a process
through digital means.
... How Tos: Advice written in plain language, including
information on how to get started with accessibility.
Originally at Guideline level.
<Zakim> shadi, you wanted to ask about difference between "EN 301 549 conditional statements" and "tests"
<Jennie_Delisi> * Very helpful to share this before the conversation. Thank you @Rachael!
Shadi: differences between EN 301 549 and tests?
re: proposed way to write an outcome.
rm: we have examples from a
couple of weeks ago.
... still some ability
... there was a was a previous decision to have like content
together.
... just talking about structure.
... 5 questions
... 1. Do we have different types of Outcomes (Quantitative,
Qualitative or Assertion) or do we have Outcome statements that
have different types of Methods
... 2. Are Outcomes and Assertions peers?
... 3. Do user needs belong with the guideline or with the
outcomes?
... 2 more Detailed Structure Questions
... 4. Do we want to include the What/Where/Why information and
if so does it become a normative part of the Outcome?
... 5. Do we want the How To information associated with the
Outcome or with the Guideline?
<Rachael> slides at https://docs.google.com/presentation/d/1p2HBgjatnLQBxeSMSFPPiafy91K3jXj5sCGYxaJystY/edit#slide=id.p
RM: Option 1: Qualitative &
Quantitative Outcomes / Peer Outcomes & Assertions / User
Needs under Guideline
... (see slide deck)
... what other questions should we be asking?
Julie: Is there an example?
RM: could try to put one together.
<alastairc> +1 on creating a quick example, I'll have a go
RM: Guideline level text
alternatives. Then under that user needs.
... Qualitative Outcomes - text alternative y/n
... Assertions , then user testing.
... we could write it out.
<sarahhorton> Examples are really helpful!
chuck: clear time with user need closely related to outcome. (chair hat off)
<kirkwood> very good question
dan: Qualitative could have an outcome that is not an assertion?
rm: assertions can't have comparative outcomes at the end.
ac: chair hat off. Wondering about assertions. maybe a method under an outcome?
dj: why can't we have multiple structures?
<kirkwood> +1 to potential of assertion being method per alistair
rm: to ac look at slide 7.
<alastairc> Yes, slide 7 (number 5) is my mental modal.
rm: to dj we will have tagging.
But need a structure for the way we work.
... next model Qualitative & Quantitative Methods with Peer
Outcomes & Assertions. Where the user needs are is key to
this one.
chuck: option for user needs seems more organic (chair hat off)
rm: we are seeking initial reactions. This is a kickoff meeting.
mg: thinking of another model for
quantitative vs qualitative. Could have an assertion too.
... some may not be obvious.
<Zakim> alastairc, you wanted to comment on flat structure for methods/assertions
mg: need to have an assertion to get to more qualitative measures.
ac: maybe have a flat structure for methods/assertions.
<Zakim> jeanne, you wanted to say that "1-5 words" is not qualitative. Equivalent purpose is qualitative. Assertion is not the results. It's the process.
Jeanne: 1- 5 words is
quantitative. Assertions can not be measured.
... Equivalent purpose is qualitative.
... Assertion is not the results. It's the process.
... need to be careful with qualitative/quantitavive.
<alastairc> So that's a +1 to flat list of methods?
<dj> example of overlay spoofing: https://adrianroselli.com/2020/06/accessibe-will-get-you-sued.html#Spoofs
Jeanne: as quantitative can be spoofed.
<Zakim> alastairc, you wanted to comment on the spoofing.
ac: not harmful at this point to separate qualitative/quantitavive.
<kirkwood> +1 to Alastair not helpful to split
<sarahhorton> +1 what's the benefit of separating qualitative and quantitative outcomes from the perspective of defining standards that need to be applied regardless
ac: But we need to be mindful of it.
<Bri> +1 Jeanne's concerns
dan: have we made design on multiple methods or one method to satisfy outcome.
ac: it is a building block approach. To start assume an "all" must be met by default, allow some to fall away if harder methods/tests are accomplished.
dan: more comfortable with a decision tree method.
rm: methods are sufficient. But there is flexibility.
<alastairc> I think to start with we'd assume an "all" must be met by default, allowing some to fall away if harder methods/tests are accomplished
Detlev: structure seems too complex.
<Zakim> Chuck, you wanted to close this part of the conversation
chuck: Conversation will continue on Github.
<Rachael> https://github.com/w3c/wcag3/discussions/52
rm: WCAG 3 structure with
definitions.
... Then we have instructions.
... Then we have comments . Reply to comments. Do a "Thumbs up"
or "Thumbs Down "or "Watching Eyes".
<sarahhorton> For all of this work it would be beneficial to include examples
chuck: down voting could mean informative. There is not an in between.
ac: wondering if methods could be done in What/Where/Why?
Graham: what is the definition of normative?
<alastairc> Defined by WCAG: https://www.w3.org/TR/WCAG22/#interpreting-normative-requirements
rm: Normative content is required for conformance. It is difficult to change. Informative content is provided for information purposes and not required for conformance. It is much easier to change.
<alastairc> Currently the "guideline" level is normative, but isn't really testable (shrug emoji)
scott: not sure what the alternative would be if I say something is not normative.
dj: understand docs normative?
<Zakim> Rachael, you wanted to say thumbs up indicates that the level is normative and thumbs down indicates that the level is informative
rm: In GitHub do a thumbs up if
you think it is normative. down if you think it is
non-normative.
... understanding docs in 2.x are akin to howtos in WCAG3.
chuck: this is a place to start from.
scribe change?
<Francis_Storr> scribe: Francis_Storr
dj I feel like the failure techniques should be normative but they're currently not.
alastairc: can you clarify?
dj concept of adversarial conformance. someone could be using an insufficient technique and claim they're supporting a criterion.
<Zakim> alastairc, you wanted to comment on how the technology specificness relates to normative
<jon_avila> They could still meet another way though even if the fail a failure technique.
<dj> adversarial conformance: https://adrianroselli.com/2023/10/2-4-11-adversarial-conformance.html
alastairc chair hat off. one thing to be aware of is updating normative things takes longer. It's a ~six-month process to update some content
scribe: putting things as
normative does push up the time scales.
... things have been moved to the WAI publication space where
things can be updated more quickly
<Zakim> mbgower, you wanted to say failure techniques
scribe: I think there needs to be some normative technology specific content so that if something didn't work in a technology, it got filtered out
<laura> s/can not /cannot /
<laura> s/quantitative. /quantiative. /
mbgower I don't think I've ever thought of failure techniques as being normative, but maybe it's something we could consider for 3.0
<Chuck> https://github.com/w3c/wcag/pull/3380
Chuck There are two WCAG 2 PRs to review
mbgower There are 2 PRs that cover normative info. The first was approved a while ago.
<alastairc> (It will go for CFC soon)
mbgower the first is pretty straight forward. It's just fixing up some HTML where we have strong tags with code tags. There's nothing being changed, it's just updated for presentation purposes.
Chuck the remainder of the call is on WCAG 2.
<Zakim> Chuck, you wanted to clarify that we transitioned to WCAG 2
JenStrickland Looking at the change, what is it intended to apply?
<Chuck> Francis: When we had the rea decision of WCAG 2 space, to Mike's point, this was purely presentational.
<Chuck> Francis: Page looks better. DFN, that is something we can add on.
<scotto> i'm not sure there'd be any value in adding dfn elements
<Chuck> Jen: You can get rid of code class, and style the dfn for the presentation and convey the semantics.
<Chuck> Francis: Need to review the W3C style sheet.
<Chuck> Francis: If we strip it out, we probably have to write lots of extra stuff.
<Graham> why wouldn't this become a <dl>?
<Chuck> Francis: And updating W3C stuff in WAI space.
<Chuck> Jen: I think I'm understanding.
<Chuck> Francis: It's in the spec. That's the hook for the syntax.
<Chuck> Jen: Does it help with assistive technology?
<Chuck> Francis: No.
<Chuck> Francis: It's not a definition list, there is an HTML dfn element, inline, that says that "this instance is the defining instance on the page". Just popping in a nested dfn tag within the code tag.
<mbgower> and FYI, that is in the spec too
<Chuck> Jen: DFN is an obscure thing that isn't very used.
<laura> s/quantitavive. /quantitavive. /
<alastairc> +1, let's leave it as is
scotto I'm not really sure if there's any value in doing this as there's no benefit to assistive technology
<alastairc> I think Jen was saying to use <code ...><def>term</def></code>
scotto these aren't really terms
kevin isn't this a description list?
<alastairc> It could be a <dl> with <dt> & <dd>
<dj> <dfn> details: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/dfn
scribe: this is a list of input controls and what they're defined as, which is a description list.
dj If we are doing `dfn` elements, we'd have to change the surrounding code per the HTML spec for the element
<laura> s/marrch 6 /March 6 /
alastairc we don't want to make big changes as the process gets longer and there are more styling issues to deal with. I think we'll discuss it in our Friday meeting and I'll catch up with Kevin about this. We could potentially consider a definition list.
alastairc moving onto the next one
<Graham> <dl> is being used within current guidelines, so i am not sure why we can't use that?
mbgower this is altering the normative definition of "single pointer"
scribe: the original definition
listed the method as well as what a single pointer was
... (reads the original definition)
... this change says what a single pointer is
<alastairc> This definition is used by 2.5.1 Pointer Gestures, 2.5.2 Pointer Cancellation, and 2.5.7 Dragging Movements
mbgower this removes a bit of the noise of the existing content
alastairc are people happy with this? The next stage is to put a CFC together, but before that we wanted to put it before the group to see what people thing
s/people thing / people think/
<Chuck> https://lists.w3.org/Archives/Public/w3c-wai-gl/2024JanMar/0025.html
Chuck there's no voiced objections, so I'm moving onto WCAG 2 updates
mbgower we released another of the 2-week proposed updates to 2.x.
scribe: there's 2 items that are
in there, one is #1803
... removes an unnecessary reference to WCAG 2.0. Removes
outdated content, corrects link to error suggestion, adds ARIA
link
... (shows removal of WCAG 2.0 reference)
... (shows clarification of requirements)
... (shows addition of ARIA 2 technique)
mbgower we wanted to draw this to the group's attention in case people had missed the discussion
mbgower this is the one where the ACT discussion went back and forth
Chuck anything else to add?
<mbgower> https://github.com/orgs/w3c/projects/56/views/1
alastairc if this has tweaked your interest, then please get involved. Check the email you hopefully got from Mike last week.
scribe: if you can't see anything on the project board, you probably need to log in.
<Rachael> WCAG Card Sort by Interface Type: https://uxd-library.optimalworkshop.com/optimalsort/t9ac1u32
Rachael I want to make sure people have a link to the card sorts
<Rachael> WCAG Card Sort by Interaction Type: https://uxd-library.optimalworkshop.com/optimalsort/bd4b3ava
scribe: take a quick glance and
see if there are any problems. We'd ideally like feedback on
these by next week.
... if you need accessible versions of the card sorts let us
know and we'll send you Excel documents
This is scribe.perl Revision VERSION of 2020-12-31 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/why can' we/why can't we/ Succeeded: s/differnces/differences/ Succeeded: s/abiguity/ability/ Succeeded: s/taliking /talking / Succeeded: s/stucture /structure / Succeeded: s/morre /more / Succeeded: s/want inital /we are seeking initial / Succeeded: s/quantiative /quantitative / FAILED: s/can not /cannot / FAILED: s/quantitavive. /quantiative. / Succeeded: s/quantitavive /quantitative / Succeeded: s/quantitavive. /quantitative. / Succeeded: s/descion on multuple /design on multiple / Succeeded: s/Converstation /Conversation / Succeeded: s/nortmative?/normative?/ Succeeded: s/ingent is /In GitHub / Succeeded: s/thumbs up you /do a thumbs up if you / Succeeded: s/information on/example of/ Succeeded: s/marrch/March/ Succeeded: s/aand /and / FAILED: s/quantitavive. /quantitavive. / Succeeded: s/sttatify outcome/satisfy outcome/ Succeeded: s/is non normative/if you think it is non-normative/ FAILED: s/marrch 6 /March 6 / Succeeded: s/put on together/put one together/ Succeeded: s/user testing/, then user testing/ FAILED: s/people thing / people think/ Succeeded: s/be we /But we / Succeeded: s/design/a decision/ Succeeded: s/on 3/in WCAG3/ Succeeded: s/block approach./block approach. To start assume an "all" must be met by default, allow some to fall away if harder methods/tests are accomplished./ Default Present: bruce_bailey, Graham, Laura_Carlson, Francis_Storr, gpellegrino, shadi, Jennie_Delisi, wendyreid, kevin, JakeAbma, Azlan, mgarrish, dj, mbgower, dan_bjorge, sarahhorton, Bri, mike_beganyi, JenStrickland, tburtin, Detlev, julierawe, Rachael, Gez_Lemon, kmcgeephd, alastairc, kirkwood, DanielHE, scotto, giacomo-petri, maryjom, Frankie, Wolf, Ben_Tillyer, jtoles, jeanne, Jen_G Present: bruce_bailey, Graham, Laura_Carlson, Francis_Storr, gpellegrino, shadi, Jennie_Delisi, wendyreid, kevin, JakeAbma, Azlan, mgarrish, dj, mbgower, dan_bjorge, sarahhorton, Bri, mike_beganyi, JenStrickland, tburtin, Detlev, julierawe, Rachael, Gez_Lemon, kmcgeephd, alastairc, kirkwood, DanielHE, scotto, giacomo-petri, maryjom, Frankie, Wolf, Ben_Tillyer, jtoles, jeanne, Jen_G, Frankie Wolf Regrets: Roberto Scano, Todd Libby, Lori Oakley, Shawn Thompson Found Scribe: Laura Inferring ScribeNick: laura Found Scribe: Francis_Storr Inferring ScribeNick: Francis_Storr Scribes: Laura, Francis_Storr ScribeNicks: laura, Francis_Storr 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: 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]