See also: IRC log
Wilco: Deadline to release at end of August
... Need to get to an agreement of the draft next week.
... Will send out at end of the meeting. Take a look and if you have any
comments. Mostly editorial but if major concerns we want those too.
... To achieve a draft next week, we really need to review
Wilco: We want at least 2 to 3 examples of rules. First one is aria-describedby
Romain: There is a bit of mismatch between the specification and the format of the rule. Should update the rule to match the specification.
Wilco: Only significant thing to do is to have examples.
Romain: Need unique identifier. It is not mentioned explicitly. The Accessibility Requirements of Test Input Types also need to be clearly stated.
Wilco: Ok, you have made it clear that the rule needs to be updated to match specification. Do we agree though that this is a good example of a rule to include?
Tobias: I think this is a suitable rule. The assumptions need to be clearer. For example, what do we mean by "properly used" in this context.
Wilco: Who would like to get this rule into shape so we can publish it as an example?
Tobias: I could take a look at this one.
Wilco: Tobias can we get this done by end of week? Let me know if you have questions.
Romain: Having an example is great but would also be good to have an example of a page.
Wilco: We need to create test cases. Excellent suggestion but no time for that now.
Romain: Understood, can wait til next release.
Wilco: Anne suggested Not to Publish
Tobias: Assumptions were kind of muddy.
Needed to reread the first assumption multiple times to understand it.
... Confused by Step 1 that it should resolve in Do Nothing as outcome.
I don't think I've seen that for any tests. Is this on purpose?
Wilco: We've used this in the past but it is not covered in rules format. Good point.
Tobias: Also assumption states html lang attribute is taken into account but it's actually not here.
Wilco: Good point. Let's not discuss rule too much.
Tobias: Rule is good but needs work.
Wilco: Ok, let's pass on this one.
Wilco: Again, No to publish from Anne and
team
... What is the objection? Is it too straight forward?
Tobias: I think this one is suitable
Wilco: Correction, there is no objection to publish this rule.
Tobias: Do need to correct description. Not completely correct.
Wilco: Do we feel this is too straight forward or do this show the point efficiently?
Stein Erik: I think it is good to be straight forward
MoeKraft: This would need updating in the formatting to match specification.
Wilco: Yes. Any takers?
Kathy: What are you asking?
Wilco: Before publishing these rules, they
need to fit the draft rules format. There is some shuffling and renaming
that needs to be done.
... May need Accessibility Supported and Test cases to validate the rule
with.
Kathy: I can help out a little bit but
might take me a bit longer.
... Can you send me some details?
... Can we edit in github?
Wilco: I have a repository.
... I don't mind where we do this. I will make sure to include it in the
spec for publish.
Wilco: We got no votes on this one. Do we want to take 5 minutes to read?
https://auto-wcag.github.io/auto-wcag/rules/SC1-4-2-audio-control-audio.html
Romain: This rule is semi-automated. There
are some user input questions. That is not described in our rules
format.
... Should we describe this in the rules format for interactive steps?
Wilco: This is really just one way of
dealing with user questions.
... The rules format does not describe how you are supposed to do this.
... I'm not sure.
Romain: In beta, we state that the test mode for the rule itself but say nothing about the test steps.
Wilco: We used to do that in Auto WCAG but
it's really obvious. They are easy to identify.
... Identifying whether automated, semi-automated or manual did not
provide a lot of value. Do we need to talk further about user input?
... Let's add this to our issues list.
<rdeltour> +1
Romain: I think it is okay. Because the spec doesn't mandate this maybe one can word the step in their own way. We can wait for next release to further define.
Wilco: Good point. We are writing these
rules in plain language so that they can be executed by QA and not
necessarily automated.
... We can return to look at this again. Moe, create an issue on Github.
Romain: There is inconsistency. Step 5 is confusing. It is checking for video.
Wilco: Agreed, need another branch. But I do like this rule. Shows creative approach of the rules format.
MoeKraft: The description contradicts itself. How about, "This test checks that any sound that plays automatically when a web page is loaded stops after 3 seconds or the web page provides a user control to turn off the sound."
Kathy: Why are we restricting step 5 to just video?
Wilco: Not sure why step 5 says video. This
was written awhile ago. If I can recall, I think we wanted to look at
HTML5 technologies.
... Instead of using old technology like Flash or Silverlight. Are there
any other ways of doing it?
Kathy: Could just link to mp3 file
Wilco: But that would need a different rule
Kathy: so restricting only to audio and video elements. Why do we need a manual test for video?
Wilco: Ok, I need to think about this more. I think step 5 should be "audio or video" element
Kathy: Description of step 5 should be just "if audio is playing"
Wilco: What do folks think about publishing this one?
Tobias: I like it. It's quite clean from the longer rules.
Wilco: Ok, I think it needs a bit of
updating. Anyone who can update this in the coming week?
... Ok, I will do it.
... Ok, so we have 3 rules. Let's try to get them done by end of week.
MoeKraft: I can generate a new draft today.
... What about "test execution steps" issue?
Wilco: Do you think the changes are small enough to move in?
MoeKraft: It is just updating terminology
Wilco: We will send out a survey with the
latest draft. Please put your comments in the survey.
... Thank you very much!