<kathy> scribe: kathy
Chris: new member, introduces
himself
... test lead for Oracle, want to contribute and learn
Group introductions
<Wilco> https://docs.google.com/spreadsheets/d/1OSkPFocXk4K3zYLnwS78WLsWO4PvE5yRcsauyefuIUI/edit#gid=0
wilco: we use this sheet to track our work
kathy: audio element PR ready to merge, send out CFC
will: 1831 change requested
trevor: comment for jym to respond
wilco: I will look and talk to
jean yves
... Tom's 1835 has a change request
helen: new PR for iframe
#1855
... add reviewers daniel, kathy
<thbrunet> +present
<thbrunet> Apologies for running late
trevor: 1854 PR for role attribute
wilco: assign myself and helen and chris as reviewers.
<ChrisLoiselle> https://github.com/ChrisLoiselle
wilco: 1852 has reviewers
assigned
... 1850 waiting on review
daniel: will do today or tomorrow
wilco: 1833 needs reviews
... 1832 needs reviews
... call for reviews for tom's PRs
tom: 1835 has merge conflicts
helen: license issues
wilco: I will help if those get in the way
trevor: example rule by carlos,
review
... this is for links that appear inside a paragraph of
text
... looking at if the link is distinguishable from text around
it
... depending on state, the link style will change
<trevor> https://github.com/act-rules/act-rules.github.io/issues/1836
trevor: first draft attempt at
including state in rules, chose to use pseudo classes
... how do we like use of pseudo class?
helen: repeated words "any state" makes it harder to read
trevor: this particular rule might not need "any state" but other rules may
wilco: not sure it's helpful to
list any state. more interested in what states the rule might
not apply than when it does
... this direction makes the rule very big
trevor: how would you write applicability for states when rule doesn't apply
wilco: maybe an animation
... this makes rules less atomic to have to test all of the
states
trevor: better to have an atomic
rule for each state?
... trying to make clear when testing needs to be done
... not expecting examples to go through all states, just a
rough estimate
... agree over specifying is bad, but under specifying leads to
inconsistent testing
wilco: do implementations have to have multi-state testing?
trevor: that's a fair
concern
... these rules might be feasible for manual testing, more
challenging for automated
... need human to get into the right state
tom: can automate some of
it
... any state with and without - isn't that all states?
trevor: this is just for the link
tom: hovered is without
focus
... thinking default state and then the other states
... not the withouts
trevor: agree
wilco: could be many variations and combinations of states
trevor: how to reduce the number of states to manageable
tom: default, add focus, remove focus
trevor: yes, click on it to check visited
helen: in practice, look for common states and investigate when there are issues
trevor: some states are more
valuable like focus and hover
... other less common states, do we forget about them?
wilco: not convinced this is a
problem that needs to be solved
... want to list exceptions, where rule is not applied like
aria busy
... and where multi-state testing is required, aria-expanded
false changes to true when clicked
trevor: thought link might be an easy place to start
<Wilco> scribe: Wilco
Kathy: I'm looking at how states
are listed...
... Could it be an atomic rule for the default state, separate
rule for focused, separate for hovered?
... That makes this rule quite a lot to cover. Breaking it
apart seems to make sense.
... That way you don't have to cover the withouts.
... If we had one for each state, and use composite to combine
makes sense to me.
<scribe> scribe: kathy
trevor: agree but may have duplicate rules with just a difference of state
wilco: feels very redundant
... is the problem that all states are not being tested?
trevor: some are tested, but need
to be covered in rules
... cautious about combining state with subjective
helen: readability is more
difficult
... would want it to be simpler
... with so many states to test, what if one is missed
... provide enough guidance without spelling it out too
much
will: agree
wilco: could make rule
implementations difficult
... if listing states, maybe in an informal place and not
explicitly listing all states
wilco: AGWG survey this
week
... implementation matrix was approved with a few editorial
comments
... updated banner to bring back ACT Rules also approved
pending editorial changes
... update to common input aspect approved also
... other updates on survey for next week: 3 rules and minor
updates to existing published rules
helen: happy to review
daniel: hope to find middle ground on issues
wilco: will try to end meetings early