Ad Hoc What Next in A11y

08 Jul 2021


becky, CarlosD, janina, jasonjgw, Judy, Leticia, shadi

Meeting minutes

shadi: Project called WAI-Coop funded from E.U.

<shadi> https://www.w3.org/WAI/about/projects/wai-coop/

shadi: Support R&D in a11y

shadi: implementation of a11y directive in E.U.

shadi: Noting that a11y is not static

shadi: making sure we can support tech that emerges

shadi: providing resources to communities of interest

shadi: Being carried out by U. of Lisbon

shadi: That's our background -- we'll go around and introduce ourselves

shadi: described deliverables in such a way as could be completed in coordination with W3C/WAI APA and RQTF if useful and meet priorities/requirements

shadi: Symposia might be included if useful

shadi: So, today is to get to know each other, the context, and what might be possible and useful


shadi: Looking now to deliverables ... two main ...

shadi: 1.)Advancements in R&D: To hold symposia

<shadi> https://www.w3.org/WAI/rdsymposia

shadi: You may recall in the WAI past ...

shadi: e.g. metrics, mobile, text customization for readability, easo to read on web ....

shadi: user modeling, way finding, ...

shadi: Now committed for 3 such symposia from 2021-2023

shadi: learned to plan for 3 in parallel

shadi: Phps more lightweight?

shadi: Could include U. Lisbon doing pre workshop research deep dive; or post symposia r&d ...

shadi: 2.) twice a year roundtable discussions open to the community

shadi: goal to get wider participation

shadi: idea here is to stay high level -- more approachable; e.g. scaling

shadi: to be more practical and imm usable

shadi: e.g. how reliable is AI captioning and how reliable might we expect it to be in 5 years?

shadi: Could start as round table and become a more focussed r&d

shadi: what would it mean if we thought not 5 but 10?

Judy: q: how meeting open meetings per year?

shadi: 2 per year

shadi: only 4 left to go

shadi: Asks Carlos to report on collected topics ... and how to make it more community driven

CarlosD: noted gap between r&d and what is actually implemented

CarlosD: Looked at what's not connecting

CarlosD: did litt review

CarlosD: incl hci as well as a11y; collected keywords

CarlosD: lots of data

CarlosD: XR seems to bubble to the top last 3 years

CarlosD: haptics, nav, tracking ...

CarlosD: ai and ml also ranked high

CarlosD: also trying to come up with platform to collect

<CarlosD> https://drive.google.com/drive/folders/1uLkgAvLGFr3UApAaOtrG4guPZRo64JqK

shadi: very impressed by your analysis -- highlighted things in focus these days; but also what's not currently sexy

shadi: clearly what's not there is important

shadi: how to organize continues to be a concern; how to collect widely in a few focussed events

shadi: phps meta analysis might be first year focus

shadi: e.g. what are specific criteria for selecting

shadi: short term, long term; but also impact

shadi: to setup for other researchers as well

<Zakim> Judy, you wanted to speak to this point of framing discussions, rather than granular mini-topics, to start

Judy: q+

given the high number of potential topics and the limited symposia opportunities; concerned to position discussion for most impact

Judy: would prefer to set this work at a framing level

Judy: concerned we're not doing enough to use available tooling like AI for a11y

Judy: ai for accessible content creation

judy ai for personalization

Judy: don't want to wait 3 years to talk about that

Judy: would like framing discussion for that

Judy: accessible of remote meetings -- remote everything else -- we've learned is major

Judy: will bew relevant for a long time

<jasonjgw> Janina: notes the distinction between pure and applied research.

<jasonjgw> Janina: has concerns about very specific, known problems that arise in APA work (e.g., in recent Pronunciation Task force efforts).

<jasonjgw> Janina notes that APA is strongly grounded in more immediate needs (e.g., the recent shift toward remote meetings).

<jasonjgw> Jaina: Example - how to enable screen readers to consume SSML data reliably.

<jasonjgw> Janina notes concerns that we may be reaching the limits of accessibility APIs - the volume of accessibility-related data is growing. Example - XR - the large number of objects in an XR environment; data visualization accessibility. It may be time to envision accessibility APIs differently.

shadi: did I hear judy asking for broader freaming and janina for narrow framing?

jasonjgw: believe there's a nexus between jb and js

jasonjgw: xr realtime and nl can converge in the same application

jasonjgw: maybe ways to set this up to meet multiple needs

jasonjgw: so wondering whether we could cross topic a symposium?

jasonjgw: another potential approach to look at the pipeline from content creation through user consumption

jasonjgw: have we looked from that 30K foot view?

jasonjgw: has the strategic analysis been revisited recently

Judy: don't think we should be pure research at all

Judy: there are different levels of applied, and different framings that could help us

<Zakim> Judy, you wanted to clarify where my comment lies between "pure" and "applied" research

Judy: not so much how to talk about; but what's the space and what are the approaches

<jasonjgw> Janina: raises the concern that without sufficiently visionary approaches, we end up rearranging existing problems.

<jasonjgw> Janina: notes the need to create new solutions to reimagined problems.

<jasonjgw> Janina - Example: notes the concern that accessibility APIs are reaching their limits - a problems with far-reaching consequences.

MichaelC: priority for me is r&d that drives a11y guidance; so what we tell people to do

MichaelC: we need research on how to solve emerging problems

<Judy> [JB mentions to Janina that I think that the relevance of AAPIs is a highly "applied" topic, far from "pure research," so I think we're reasonably in the same arena.]

MichaelC: market developes approaches and we're slow to respond with duigance

MichaelC: always beena problem for me on how to pose a research question and get practical results back thatc can apply

MichaelC: clearly knowing better the theoretical knowledge will help us long term--but not sure how to factor it; knowing we needit

CarlosD: great opportunity to let research community know the practical problems you're facing

CarlosD: what are the issues that people face in the real world

<Judy> [JB: I strongly agree with Carlos that most researchers are unaware of most of the research needs that relate to digital accessibility for people with disabilities!]

CarlosD: suggests posting specific problem for symposium and asking for approaches

CarlosD: that might be a way to frame a symposium

CarlosD: we need to org the first symposium this year; phps that's our approach?

<Judy> +1 to everything Carlos just said

shadi: summarizing ...

shadi: need for specific problem solutions

shadi: don't forget outcomes need to end up in guidelines

shadi: also providing framings

shadi: we'll need several followups; propose we meet again; suggesting this time again on the 22nd?


<CarlosD> +1

shadi: any objection to this time on the 22nd?

Judy: no better suggestion for now

shadi: thanks all

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).


Succeeded: s/3/for 3/

Maybe present: MichaelC