wilco: I have been working on getting the list completed
... we've wrapped up updating the ACT Rules in ACT-R CG to the CR format
... I think it's done now
... then I started filling out the document for exit criteria
... a lot of things are sorted, but still a couple todos left
<Wilco> https://www.w3.org/WAI/GL/task-forces/conformance-testing/wiki/CR_Implementation_Tracking
wilco: under C4 there is at least one rule that is a satifsfying test, we don't have that in ACT-R
... there was an outstanding item to update the ARIA rules
... (there are 3 of them)
<Wilco> https://act-rules.github.io/rules/
wilco: I think we can map these to ARIA 1.1 and they can be ued to complete this requirement
<Wilco> https://act-rules.github.io/rules/c4a8a4
wilco: maybe a better idea: Document title is descriptive may be a satisfying test
romain: I agree
shadi: it is for HTML, mayne not for other formats?
wilco: if it's inapplicable it isn't, but if it passes it is satisfying
... that's probably easier to do
... does that fit with our exit criteria?
... I think that works
... anyone disagree?
anne_thyme: [asks clarification about what satisfying test means]
<Wilco> Rules that can be used to determine if an accessibility requirement is satisfied are called satisfying tests.
wilco: quoting the spec "Rules that can be used to determine if an accessibility requirement is satisfied are called satisfying tests."
anne_thyme: if we're testing for 1.1.1 and it only tests specific elements (img), the rule passes but only for thoses elements
Wilco: right, but the one we propose is "Document title is descriptive"
anne_thyme: is it a satisfying one?
... the rule is only satisfying when the title exists, it is checked elsewhere
romain: I was wondering if these rules were separate because one is automatable and the other may not be
Wilco: I still think it means that if the rule is applicable, it's satisfying
... so it would pass the exit criteria
anne_thyme: yes, I think you can say that
... but then basically we're using the applicable outcome to ???
Wilco: I think we can say that "Document title is descriptive" is satisfying when it is applicable
... we need to update this rule, and it can complete C4
... the next one is C5
... we can use a rule mapping to ARIA to complete this exit criteria
romain: works for me, there will be the EPUB one as well
... this can be done by next week
anne_thyme: I think that requires at least a 2-week cfc phase
Wilco: oh right
... let's look at the timeline
... for C8 we need other technoologies
... Romain, where are we on the EPUB rules?
romain: the rules is alsmot ready on its own, but I'm not sure if/how I can reuse ACT-R rules
... because the test subject may not be explicitly defined, and the definition might be slighly differrent in EPUB and ACT-R
... does it work if I add a general document which defines what "page" means in an EPUB context?
anne_thyme: for some rules we changed page to something else, by trying to use more specific concepts like DOM
... I don't know if it's applicable to EPUB
romain: it's why the test subject being kinda implicit can be an issue
... if the test subject is an EPUB, I cannot apply a rule which test subject is a DOM tree without explaining how to get one for an EPUB
anne_thyme: Kasper from SiteImprove definitely has an opinion about this, we could setup a meeting to talk about this
romain: sure, we can talk more about that
Wilco: I'll join
... that should get our C8 answered
... the next one is about the data format
... Jey is working on putting the data format we have on the web site
... some of these things are missing (data from Alfa)
anne_thyme: please let us kwow as soon as you hear something
... I think they said a while ago that everyting was OK
Wilco: I think it's enough for the automated one (C24)
... something is missing for C25 (semi-automated)
... about the no-keyboard-trap rule
... we have an RGAA rule
... not sure if we have a second one
... maybe something fro Kathy?
kathyeng: I'm hoping to have something today
Wilco: that would be great
... I think it means we're almost there, awesome!
kathyeng: is that the first one that you need, or do you think you need more?
Wilco: if you can get the keyboard trap one, that's enough for the exit criteria
... I'm trying to get other implementors as well, but 2 is what we need
... I would like to hear from Shadi what are the steps to get our of CR but he's not here :)
... regarding the timeline, yes we would need to go through a cfc process for ACT-R
anne_thyme: if we skip the CfC phase it would be bad
Wilco: I agree, which is why I'd like to propose it ASAP
anne_thyme: OK, let's try that
Wilco: that's my update for the CR
Wilco: Mary Jo do you think we should skip this issue until Alistair is on the call?
maryjom: yes
Wilco: it's linking to section 3 instead of section 4.3, easy to fix
maryjom: we discussed these issues last time but didn't have anybody assigned
Wilco: I think I said I'd do it
[Wilco self assigns]
Wilco: I don't know why these maps to auto-wcag
shadi: I used the same example as from before
... I was thinking of using ACT-R, but then some examples would use auto-wcag and other act-r
... I felt this was a separate issue
Wilco: right, we should update all the references
... the PR looks fine
... do we have an issue open to change auto-wcag to ACT-R?
shadi: no, I'll create a self-assign
Wilco: anything else on this PR?
shadi: we might need to update the EARL JSON as well
Wilco: we probably have to change the context too?
shadi: yes, that's what I meant
Wilco: maybe embed the context is the cleanest thing to do...
... this PR closes #362 then
... can we merge the PR and let the other changes to a new issue?
[no disagreement]
Wilco: on the CR exit criteria there is a list of the rules we can use for publishing
... how many rules do we think are appropriate for a first batch?
... there are quite a number of media rules, and other quite basic automated rules
... what do people think is the right direction?
anne_thyme: I think that once we choose which ones are ready, we should make sure they don't have open issues
... right now it's difficult to track which rules have issues and which don't
Wilco: that should be fixed now
anne_thyme: but someone has to explicitly put the info and label the issues
Wilco: good point, can you create an issue on ACT-R about that?
anne_thyme: will do
Wilco: shadi do you think we should rather have a small batch? a particular topic or focus? or submit as many as we can?
shadi: to submit to the AG?
... I think we should use the same rule set as we use in the implementation report
... there will be manual and automated rules, from various vendors and implementations
<Wilco> https://www.w3.org/WAI/GL/task-forces/conformance-testing/wiki/CR_Implementation_Tracking
shadi: this is our best chance of suggest, so I would go for that set
... maybe between 10/12 rules
Wilco: it's probably a diverse set, but shows the spread of the work we can do
shadi: looks good
... do we have anything about the perceivable principle?
[shadi and wilco go over the covered principles]
shadi: this looks about the range I would expect. It would be good to have something under "Understandable" as well
Wilco: do we want to submit EPUB and ARIA rules in the same batch?
... we can't, as they don't have multiple implementations
shadi: I don't think the AG group would be in position to approve that
... I would include these rules in the report, but not submit to AG
Wilco: sounds good
... last topic: I'm not gonna be around next week, do we want to meet?
maryjom: I won't be there either, I don't think we can have a meeting
Wilco: ok, then let's try to get these exit requirements done, and catch up in 2 weeks!
maryjom: I updated the availability survey, it might help during the summer