Meeting minutes
<Lauriat> https://
Measurable guidance editing (minutes from Tuesday)
<Lauriat> https://
<Lauriat> https://
<Lauriat> https://
Shawn: <pasting in reference links>
Shawn: The topic for this is one of the issues that was filed for WCAG 3 requirements. We drafted a response to the issue, brought to WG. First link is to survey results.
Shawn: 2nd link is to issue, and 3rd link is to the discussion in AGWG.
Shawn: The WG said this should go back to Silver TF for modification as per our discussion. However, the discussion was around the requirement rather than what the issue seemed to talk about, which is the design principal.
<Lauriat> https://
Shawn: <adding additional resource links>
Shawn: That goes to design principals. We have Wilco which filed the issue. Feel free to jump in.
<Lauriat> https://
Shawn: Current wording under conformance model <pasted>
Shawn: This is partially on requirements and partially on conformance model.
Shawn: In requirements we have noted certain accessibility guidelines are clear, some less so. Concern about statement that some needs cannot be measured with pass/fail.
Shawn: The requirements agree with the statement. We are not saying that we need more than pass/fail, we say we need more than true/false.
Shawn: Wilco, did I interpret correctly?
Wilco: yes.
Shawn: Response is "yes, we will make pr to change to pass/fail"
Shawn: FROM pass/fail to true/false
Wilco: Not sure that's what's going on here. I see it, eventually we do have to boil this down to... does something conform to the requirement? Ultimately this is true/false or pass/fail.
Wilco: This does eventually get to a binary option. Even within that if you have a scale, it has to come down to "did you meet the requirement or not"?
<Lauriat> https://
Shawn: The requirement itself very much agrees with that. We need to have tests or procedures that can be used to verify.
Wilco: How is this different from WCAG 2? That's not clear to me?
Jeanne: The gradients. How WELL do you pass? We want to go more than the minimum. Here's the bottom (pass/fail), but we want to say "you did it better".
Jeanne: That's how we think we can include COGA requests.
Suzanne: Wilco is basically saying "I want to review later, and determine if it falls into a category, and determine if they pass at a level".
Suzanne: This is another option - ... for a sub-group, we could have 5 different banks that target a different population. WCAG requirement is trying to allow a company to make their website, go through a process, and get credit for that.
Suzanne: Would still be pass/fail. But it wouldn't be pass/fail in that you could look at the site and know for sure that it did or did not do that. Does wilco agree we should leave the window open? I think we should.
<Zakim> Lauriat, you wanted to distinguish testing from conformance
Shawn: I want to review 2 aspects. Confusion came from us having this wording.
Shawn: Distinguishing testing, measuring and conforming. Testing and measuring can happen in multiple ways. Conformance has lines of pass/fail at different levels.
Shawn: I think this section needs some clarity around the wording to distinguish between aspect of testing and conformance. If we reword such that it speaks to the multiple measurements.... that build up to overall conformance, would that make it clearer?
Wilco: I'd like to see that wording.
Sarah: Some thoughts. I think we are talking objective measures so when we say pass/fail statements and clarity around them, it's still "you did or did not". Maybe part of what we are hung up on is the pass/fail concept.
Sarah: That we are providing requirements and either you've met the requirement or you haven't. That word "requirement" came up with AGWG. And "guidance and guidelines". I could see how we could get tripped up on.
Sarah: This is less about moving away from binary statements to ... we want to move that into ... address that sometimes the requirement is more subjective than other times, and makes more difficult to determine if you've met the requierment or have not met the requirement. Maybe clean up the confusing language.
Sarah: Essense is there but not clear. We are using terms that mean something else in this doc.
Shawn: Agreed. That's one key aspect. Another that Wilco pointed out is it seems to make assertion that some different needs are not measurable, when we are really trying to convey that they require different ways to measure.
Sarah: One more thing... I feel like we are getting tied up in conformance and specifications. If you think of WCAG as a spec that tells people how to achieve accessibility, that feels like one kind of "thing". That's guidance.
Sarah: Conformance feels much more aligned with requirements. "I need to meet these requirements to conform to this standard".
Sarah: I'd love to see us talking about that specification aspect. Or keep them separate. Smushing them together gets us stuck.
Sarah: Difference between "how to do it" and "did you do it".
<shadi> +1 to Sarah on separating technical specification from guidance
Shawn: The aspect of this opportunity is essentially to say "we have an opportunity to have more guidance with more flexibility on measuring that can build up to conformance."
Sarah: I think that trying to get a 2-fer out of it makes the work hard. Make it so that "how to do it" also allows for measuring if you did do it.
Wilco: For me it boils down to true/false statements. Color contrast with a number for example.
<jeanne> sarah has a very thought-provoking point about separating the how to do it from the did you do it.
Wilco: Or text and a decision tree, with true false statements. Or Rubrik, with scales. You say yes or no at each level. What I don't quite see...
Wilco: How that's any different from WCAG 2? If we are just testing the content are we proposing to test different things than the content? Will we try to measure with user input? Statements?
<Zakim> jeanne, you wanted to answer wilco
Jeanne: We are trying to put the pass/fail aspect at a different level. We found during silver research that they love the guidance, hate the sc. Too ambiguous or difficult to parse.
Jeanne: We want the guidance to be separate. We could talk about things more clearly.
Jeanne: And put he pass fail at a different place than in the guidance. So we could say "here's how you do, here's how you measure" and here's how the pass/fail works.
Jeanne: That will enable us to include more difficult guidance from COGA and low vision, because they couldn't fit it into a true/false.
Shawn: an illustrative example...
<Lauriat> "All non-text content that is presented to the user has a text alternative that serves the equivalent purpose…"
Shawn: For non text content today, ignoring the "exceptions", we have fairly straight forward guidance that is easy to understand. <reads from sc>
Shawn: With this example, it notes serves the equivalent purpose, but doesn't explain what that means, which is benefiicial. If it tried, it would be overly convoluted. This example separates that out. We can have ACT walk through decision trees.
Shawn: Come up with something equivalent. Jeanne is that along the lines of what you were describing?
Jeanne: Yes.
Shawn: Wilco did that make sense?
Wilco: Thinking on it.
<Zakim> Lauriat, you wanted to ask about alt text as an illustrative example
<shadi> https://
Shadi: I thought I was following along, but now I feel lost. First, Wilco asks if you are planning to introduce other types of test. I expected the answer to be "yes".
Shadi: Multiple ways to measure does include this, not just looking at content but other things. Maybe I heard the answer was "no". This is where I first started to get lost.
Shadi: I got more lost when we discussed putting guidance first, and pass/fail in a different place. At the end of the day they are still there. It will be measured or determined by a pass fail statement.
Shadi: Maybe it's only me, but I got a bit lost and I don't have a clear example.
<Lauriat> "The visual presentation of text and images of text has a contrast ratio of at least 4.5:1…"
Shawn: I think the overall answer is we do want to include other ways of testing. To give another example... we have color contrast guidance. For minimum contrast we have ... <reads>
Shawn: has 4.5:1, we could in WCAG 3 have guidance specify a minimum contrast but have it centered on user need as opposed to the specific test to indirectly say this is how you me the user need.
Shawn: This could give us some flexibility on determining if we met, w/o having to update the guidance.
<Zakim> Lauriat, you wanted to give another example that maybe helps, color contrast
<Zakim> SuzanneTaylor, you wanted to share draft based on what Jeanne shared
Suzanne: I tried to write down what Jeanne expressed.
<SuzanneTaylor> "Measurable Guidance: Certain accessibility guidance can be stated easily in a single pass/fail statement. Others, far less so. There are needs of people with disabilities, especially cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement. By expressing how-to implement and how to measure separately, as well as by allowing multiple
<SuzanneTaylor> means of measurement, we can allow inclusion of more accessibility guidance."
Suzanne: I can paste into IRC.
Suzanne: <reads>
Shadi: I have some editorial suggestions that to me are quite distracting on this requirements item.
Shadi: Feel free to take offline. One of the things, change the word "especially" to "such as", it creates a polarization in the community. I don't know if there's research to prove.
<kirkwood> agree ‘especially’ to ‘such as’
Shadi: The other thing that trips me over, first 2 sentence <reads> I put in conjunction, I don't know why these statements are there, they are statements of fact. But putting in conjuction, are we promoting more or less guidance?
<SuzanneTaylor> "Measurable Guidance: There are needs of people with disabilities, such as cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement. By expressing how-to implement and how to measure separately, as well as by allowing multiple means of measurement, we can allow inclusion of more accessibility guidance."
Shawn: The introduction was to state that the problem we have opportunity to solve that lead us to review multiple ways to measure. If you have a suggestion to flow better I'm open to it.
Shadi: Thanks Suzanne.
Shawn: <reads>
<jeanne> +1 to ST's edit
Sarah: I like how this is taking shape. Are we still talking about measurable guidance?
Sarah: In a way I feel like we are talking about trying not to do rigid guidance. That might just set the stage wrong. We may want to look at the label again, and make sure it labels what we are talking about.
Sarah: Maybe somebody else.
Wilco: I think this change is helpful to me, I find this clearer.
Wilco: What does come up, a question about normative and making sure that normative text is still sufficiently testable. It is a thing that will be discussed.
Chuck: It will be discussed, let's keep it separate.
Shawn: Back to Sarah's point. I feel this is talking about measurable guidance. It's kind of oblique. Takling about a couple of things. The overall objective of including more guidance than we included before.
Shawn: The other is we will still have the guidance provided by WCAG 3, and will still be measurable, so we can determine if it's met.
Sarah: It makes me bummed that this is what the good news is. When I read the revision... there's an opportunity to be aspirational and inclusive within the context of this section.
Sarah: Might be something that we could say that doesn't focus on the measuring part. That we are trying to broaden coverage.
Shawn: This is one opportunity around the conformance model. I'm inclined to agree with you. We have a chance to include more guidance, the way we fill that opportunity is through different ways of measuring.
Chuck: good topic, but goes outside of answering Wilco's issue.
Shawn: I think we've arrived at a proposed re-wording.
proposed RESOLUTION: Used updated response.
<sarahhorton> +1
<Lauriat> Measurable Guidance: There are needs of people with disabilities, such as cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement. By expressing how-to implement and how to measure separately, as well as by allowing multiple means of measurement, we can allow inclusion of more accessibility guidance.
<Lauriat> +1
<Wilco> +1
+1
<KimD> +1
<SuzanneTaylor> Measurable Guidance: One area of exploration is to innovate in the area of ways to know whether or not you have followed guidance. There are needs of people with disabilities, such as cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement. By expressing how-to implement and how to measure separately, as well as by allowing multiple
<SuzanneTaylor> means of measurement, we can allow inclusion of more accessibility guidance.
Suzanne: We may still be able to address Sarah's points. ...reads...
Suzanne: makes sense to acknowledge that it's not solved but different ways to measure. Does that meet Sarah's requirement?
Sarah: Thanks Suzanne. I'm glad to see it there!
<kirkwood> I don’t think the disability type “cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement” change to “understand” to “measurable”
Sarah: You've opened a beautiful door, and its very excitement. I almost want to say "...for example..." you are reframing it by saying its hard to know if you've followed guidance and we are going to come up with ways to solve that.
Sarah: One solution might be how to implement and measure separately. You've opened the door to this exploration of innovation. It's very exciting!
<SuzanneTaylor> Measurable Guidance: There are needs of people with disabilities, such as cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-understand pass/fail statement. In addition, one area of exploration is to innovate in the area of ways to know whether or not you have followed guidance. By expressing how-to implement and how to measure separately, as well as by
<SuzanneTaylor> allowing multiple means of measurement, we can allow inclusion of more accessibility guidance.
Suzanne: This might solve it.
kirkwood: I had a very similar feeling with what Sarah said. I think that rather than understandable we should say measurable. I don't like the idea that these aren't understandable guidance. We can still do binary, it just needs better language.
<Lauriat> +1 to kirkwood
Kirkwood: We can do a better job, it can be done. We haven't figured out how to measure at this point. I would change that one word there.
<Lauriat> Updated word:
<Lauriat> Measurable Guidance: There are needs of people with disabilities, such as cognitive and low vision disabilities that cannot be stated in a single easy-enough-to-measure pass/fail statement. In addition, one area of exploration is to innovate in the area of ways to know whether or not you have followed guidance. By expressing how-to implement and how to measure separately, as well as by allowing multiple means of measurement, we can allow inclu[CUT]
<Lauriat> accessibility guidance.
Shawn: Easy enough to measure...
Shawn: When we bring to AGWG we'll probably have more wordsmithing and clarity, but I think we have the concepts clear at this point.
Shawn: Are there any other aspects before we resolve?
proposed RESOLUTION: Used updated response.
+1111111
<Lauriat> +1
<SuzanneTaylor> +1
<jeanne> +1
<sarahhorton> +1
proposed RESOLUTION: use updated response as proposed in latest draft of measurable guidance
<janina> +1
<Wilco> +1
<jeanne> +1
<kirkwood> +1
<KimD> +1
<SuzanneTaylor> +1
<sarahhorton> +1
RESOLUTION: use updated response as proposed in latest draft of measurable guidance
Shawn: 5 minutes left. Any updates from sub-groups?
Subgroup updates
Jeanne: Update to be pasted...
<jeanne> The Alt Text subgroup used to meet every week, but we're having biweekly meeting since last month. We are waiting for the new templates for the How-to and Method documents. Please let me know if there is anything we should work on for now.
Jeanne: From Makoto.