– DRAFT –
AGWG-2025-01-14

Introduce updated WCAG2ICT work statement

GreggVan: I retired as of 2 days ago

<kirkwood> congrats!

Chuck: we have a list of topics for future agenda, in case anyone has

Introduce updated WCAG2ICT work statement

Chuck: we'll be introducing WCAG2ICT work statement today, maryjom can you introduce it?

maryjom: I have opened an issue for tracking the AG WG review of our proposed updates to the work statement

<maryjom> w3c/wcag2ict#611

maryjom: we updated the work statement to provide more details re phase 2 of our WCAG2ICT note update, we want to do additional work to the note

maryjom: we want to keep the note maintained over time, eg keep it up to date with WCAG and our current thoughts on applying WCAG to non web

maryjom: we want to make sure we have the latest and greatest advice for how things like non web ICT and documents are interpreted

Chuck: thanks maryjom

CSUN Planning update

Chuck: chairs have been discussing what we should do with the one day we'll have to meet, the Monday

Chuck: what we think will happen: we'll meet the entire Monday, probably 9ish to 4ish

Chuck: morning session likely continues discussion from TPAC

Chuck: afternoon session we want to do something that best uses the folks that can attend, we're thinking conformance now… if there's a volunteer who can set up and plan an evening dinner that would be great

Chuck: any questions re CSUN planning?

Onboarding discussion

<Jennie_Delisi> Possible remote option?

stevefaulkner: do we have a room?

Chuck: not yet

kevin: it's on my list

Chuck: we have ~3-5 new members per month… we're thinking to start 30 mins earlier one meeting a month, we're looking for a volunteer to lead that session

Chuck: to help with onboarding

Chuck: first couple of sessions the chairs will join

MJ: I'm interested in leading some of the onboarding sessions… but not the first one so I can see how it goes

Chuck: think it would be fine for a chair to do the first one

Review Handbook updates with exercise

alastairc: for context… updating the subgroup handbook, guidance for all of the subgroups, in terms of taking the early draft of a particular guideline we've been working on and getting it ready for our next publication

alastairc: one specific bit of it … that is tricky… getting together for the basic requirements and decision tree

alastairc [shares screen]

alastairc: what we're trying to do with the decision tree is make it clear which requirements apply in a certain scenario… also for user agents. Which is something that varies per platform and over time

alastairc: potentially different technologies have different requirements and capabilities

alastairc: we're looking for this to guide people in what requirements to fulfill

alastairc: [switches to 'clear meaning' tab in document]

alastairc: we start with a scope, 'for each … in …' , then we describe when to continue or stop

alastairc: programmatic determinability is dependent on the accessibility supports set

alastairc: early feedback we got is that there isn't enough content for foundational requirements to have clear pass or fail

alastairc: what we're trying to do with the foundational testing section… to have enough in there so that you can normatively test what's in the foundational requirements

alastairc: 'focus appearance' is fairly different, has easier decision tree, just two steps

<kirkwood> can you share link to the decision tree refinement?

alastairc: in the 'foundational testing' section for this one, we provide a summary

<alastairc> https://docs.google.com/document/d/1y5r2ubojyx4kHOu01HqyC1WLkIPhJibwJigAHwLNwrI/edit

alastairc: the foundational testing section is in this case longer than the decision tree

alastairc: I've drafted doing this with the keyboard subgroup work too

alastairc: each of the requirements in that new session, it's a 'continue', each would need to be met

https://docs.google.com/document/d/1y5r2ubojyx4kHOu01HqyC1WLkIPhJibwJigAHwLNwrI/edit

<bruce_bailey> Looks great, addresses the recent conundrum keyboard subgroup was trying to work through!

<alastairc> https://docs.google.com/document/d/1obzGKEp9802Z8_IoDRIYEwZ5mvwRtYyiGYW_eXGGSFQ/edit?tab=t.0

alastairc: please create your own copy and spend about 10 mins filling that in

stevef: re the keyboard one you joined… it didn't seem to mention moving between views

alastairc: for the new definition of view that would pages, would probably the first one

<alastairc> Word document version: http://alastairc.uk/tmp/Human%20language%20decusion%20tree%20example.docx

giacomo-petri: we use the same structure for labels …the link to the methodology that is presented, it should be a bool, yes or no. A sequence is not correct, I think

giacomo-petri: potentially the link to the methodology should be before the yes/no

<stevef> have to step away for 10 mins to finish cooking daughter's food

giacomo-petri: if we're going in 'yes', we assume we are passing? but shouldn't the test be before the condition / yes/no?

alastairc: the intent was to say 'yes, and', eg 'as part of that yes, it needs to do X, Y or Z'

alastairc: the goal was to get people in the right branch, to know if the req applies or not

alastairc the question in the decision tree is to get you in the right place, the answer within that would potentially tell you which requirement is needed. So the 'yes' doesn't mean 'pass', it's just 'yes this is the scenario you're in'

<ChrisLoiselle> Hi, had to drop this meeting for another customer facing call. Apologies. Great work as always.

GreggVan: this seems to try and make everything flat/linear… think it doesn't work.

GreggVan: either you have to do A, or you have to do B. Sometimes one of those branches itself has requirements

GreggVan: sometimes we have to do one or the other, sometimes both.

<Jon_avila> I agree sometimes, if you stop then you don't find other failures. We almost need a yes and continue

GreggVan: if this is a decision tree, it should be organised so that it makes sense… we shoul figure out all the different ways one can fail

<Zakim> alastairc, you wanted to answer Gregg, where it isn't always A or B.

GN: @@@

alastairc: there's more options that one of the following

GN: but then exactly one has to be true?

alastairc: whether something is true, in some cases, could depend on the accessibility support set, or if it 's not good enough you may have to go to the next step

GreggVan: as a hypothetical example… if three things all need to be true. The way to do it would be to say 'the following three things need to be true'

GreggVan: then each step, you'd say 'if this is true…' and to say underneath the steps 'if any fail, you don't pass', or something like that.

GreggVan: we should avoid the suggestion people could stop testing as soon if one specific item is not met

GreggVan: then my second point is, we probably need hierarchy and one question at the top in the case that there are tests depending on whether one variable is true or false

GreggVan: to avoid having the same question upside down

bentillyer: I don't agree with the notion that folks wouldn't read the whole decision tree once they failed one it em, I certainly would

alastairc: I do agree we could probably make it clearer

alastairc: the tricky bit for the text alternatives… we kind of have three scenarios

alastairc: some on the author, some on the user agent… GreggVan are you saying step 3 would be indented after step 2?

Chuck: I think Ben and Gregg can be simultaneously right

GreggVan: 'fail' could be read as 'don't continue'

GreggVan: the other two choices are subsets of the first questions, not separate questions

alastairc: I'll slightly modify the exercise

<alastairc> https://docs.google.com/document/d/1obzGKEp9802Z8_IoDRIYEwZ5mvwRtYyiGYW_eXGGSFQ/edit?tab=t.0

alastairc: for anyone who is not sure, probably stick with the original. As an alternative, you can take one of these and put it in your copy and refit it

ljoakley: I got a bit confused with everybody's comments… everyone who works with computers knows how to read a flow chart… I thought that's what we were trying to get to with the decision trees

GreggVan: me too, but not visual

alastairc: one of the difficulties is with providing UA support and allowing that to be an optional branch

<Zakim> GreggVan, you wanted to ask quick question

GreggVan: I think part of the issue is we're treating it as a static page, we might be able to only show the parts of the tree that are relevant

<Zakim> Chuck, you wanted to check our remaining time

<alastairc> Template: https://docs.google.com/document/d/1obzGKEp9802Z8_IoDRIYEwZ5mvwRtYyiGYW_eXGGSFQ/edit?tab=t.0

<alastairc> Examples: https://docs.google.com/document/d/1y5r2ubojyx4kHOu01HqyC1WLkIPhJibwJigAHwLNwrI/edit?tab=t.3ia49ubv8hjo

<alastairc> Also, has anyone spotted the deliberate difference between Focus Appearance and Clear meaning, and which did you find easier/better?

<DJ> I find the Clear Meaning decision tree clearer/easier

<alastairc> So having the extra step is easier than combining it? Interesting

<alastairc> Template: https://docs.google.com/document/d/1obzGKEp9802Z8_IoDRIYEwZ5mvwRtYyiGYW_eXGGSFQ/edit?tab=t.0

<alastairc> Examples: https://docs.google.com/document/d/1y5r2ubojyx4kHOu01HqyC1WLkIPhJibwJigAHwLNwrI/edit?tab=t.3ia49ubv8hjo

<DJ> I think of it less as an extra step and more of a separated question with two branches

<DJ> Like the structure of a choose your own adventure story

agreed with DJ

alastairc: did anyone get far enough to share?

GreggVan: [shares screen]

<Jennie_Delisi> Gregg is on mute but may be trying to speak

Subgroup working sessions

<Chuck> bruce: Keyboard subgroup would like to work from where Alastair got.

<Chuck> alastair: I suggest bringing over to your doc.