– DRAFT –
WAI-Adapt Weekly Teleconference

agenda/

adenda?

<Lionel_Wolberger> present_+

Introduction, logistics. Any Updates?

Janina expects to be gone Aug 20-22

matatk: I may be gone on the 20th too

<Zakim> matatk, you wanted to ask about time off

TPAC Preps

Lionel_Wolberger: Let's revisit next week

matatk: Reminds people to register especially if in person

matatk: Notes schedule is decided, but we're still working details

Lionel_Wolberger: Recalls to print more AAC handouts

Symbols explainer

Russell: Just checked and nothing new on 240

matatk: Think we should decide whether we want response ahead of TPAC, or whether we're OK to get closure at TPAC

matatk: If ahead we should be clear on our agenda

matatk: Believe no news from Unicode? Would we know it's not going in this year?

Russell: Believe I would have heard by now

matatk: Are their minutes public?

matatk: Might we start using?

Russell: Advice is to not use until formally adopted

Russell: Believe minutes may be public, will look

matatk: Would be good to have a sense whether there are serious blockers that would put us back more than months

matatk: With that understanding we can figure out next steps and agenda

janina: The "and therefore we should" means changes to the draft
… we care about user agent rendering, and how authors will 'decorate' their content with the correct symbols
… we need tooling for that
… this means there are two documents to update
… once we get confident that unicode will be there, we can take these steps to rely on it

matatk: If time permits, I suggest we do a mapping excercise

matatk: It would be good to do a mapping exercise from our key to different symbol sets
… a mapping from our key to a relevant symbol in another set, e.g., ARASAAC

matatk: For now with BCI, but would give us a sense of how viable

matatk: Notes his tool also included rendering; and helps in viability review

matatk: I propose we select a large froup of pages, say 1M
… decide some portion of the target pages to decorate (perhaps just headings)

matatk: We can test rendering to test how it looks rendered apart from semantic accuracy
… then have a script screenshot the page
… we would then see if symbol rendering would break pages
… the mappings would be part of the registry

matatk: mappings could become registry updates
… the testing, on rendering would be associated with the specification

matatk: we can work on apart from unicode

janina: ARASAAC is not going into Unicode.

Russell: Agree

janina: The value proposition is that we (a) enable the decoration of symbols and (b) they will render in the symbols ideal for the end user

Lionel_Wolberger: Should we be writing an Explainer before TPAC?

janina: Need to figure out priorities and allocate time as best we can.

janina: We have some next steps.

matatk: If we find unicode not in, it's a reason to go back and update 240

matatk: If the outcome is we'll wait and implement, that's a good outcome

matatk: But also is let's take the registry path in that case; also a good outcome

Russell: I think we're understanding it's not in this year

matatk: Only to know if there's a substantive reason

matatk: We need acceptance from WHAT to push further on adoption with Unicode

Lionel_Wolberger: Our main Adapt TF page from W3C links to https://www.w3.org/WAI/about/groups/task-forces/personalization/
… suggest Roy bounce that link to https://www.w3.org/WAI/about/groups/task-forces/adapt/

Lionel_Wolberger: Suggest our Explainer move symbols up top

<Zakim> matatk, you wanted to discuss mapping exercise and to discuss testing

WKD Github actions and PRs

matatk: Have read most ... very exciting

matatk: I read the materials
… thanks to Abhinav for the overview
… there is room for Adapt to define semantics and types of links
… Linksets supports a rich language for stating, here are links from a particular page
… e.g., we could state 'here is the accessibility statement for this page'
… e.g., we could state 'here is the relevant help and support for this page'
… the significant challenge I see is a "one to many" issue
… there is no wild card for assigning these mappings
… Given that I could not find this, it seems a site would have to assign a linkset for every single page
… There is a standard for URL wildcards
… so logically, these should be able to be combined

Abhinav: To restate, if a given destination e.g., home, is common to many pages
… how do we assign this common 'home' page to every single page

https://www.rfc-editor.org/rfc/rfc9264.html#name-link-target-object

matatk: We could ask Phil these questions

<Lionel_Wolberger> +1

matatk: If Phil says no, perhaps we will consider modifying linksets to meet our needs

Abhinav: So we could look into extending the RFC if we need to add that capability?

Lionel_Wolberger: Who's the standards body here?

matatk: IETF (same as well-known URIs)
… We are talking about IETF and RFC 9264

<Abhinav> { "linkset":

<Abhinav> [

<Abhinav> { "anchor": "https://example.net/bar",

<Abhinav> "next": [

<Abhinav> {"href": "https://example.com/foo1"}

<Abhinav> ]

<Abhinav> },

<Abhinav> { "anchor": "https://example.net/boo",

<Abhinav> "https://example.com/relations/baz" : [

<Abhinav> {"href": "https://example.com/foo2"}

<Abhinav> ]

<Abhinav> }

<Abhinav> ]

<Abhinav> }

{

"linkset": [

{

"anchor": "https://example.com/*",

"href": "https://example.com/home",

"rel": "home"

}

]

}

Abhinav: I suspect this will not support wildcards