<scribe> Scribe: Joshue108

Accessibility of the web of things.

JW: Summarises - Janina replied to survey

Josh engaged with WoT IG, where they referred to other use cases with a11y implications

CFC in APA - endorsed?

JS: Runs through today - I'll follow up.

JB: Which ones

https://www.w3.org/2002/09/wbs/1/wot-uc-priority-202005/results#xq9

JW: So the question is to look at the additional use cases that WoT IG are suggesting

<janina> https://lists.w3.org/Archives/Public/public-apa-admin/2020Jun/0004.html

JW: From my reading they all have a11y properties and are relevant.

JS: Do we have any clue on how they are going to work on this?

Make sure our goals are aligned.

JOC: Gives background on meeting with WoT

<jasonjgw> Josh notes it is encouraging that the Web of Things Interest Group is developing accessibility-related use cases.

JS: I responded already on most.

<jasonjgw> He suggests we have a supporting role in collaborating with them on use cases.

JS: I flagged that we will have a position on them
... <Discusses hot house agriculture, a11y and WoT>
... Presses that if you have a UI, there are a11y implications.

JB: There is a high rate of disability in agriculature.

JW: +1 to Janina about use cases with UIs being relevant.

They are all about support for adaptability on the user side.

We are familiar with some of these - the individual applications are less important.

Less interested in the application domain vs recurring theme.

JS: Good point. There is a recurring theme.

<SteveNoble> Speaking of agriculture - see http://www.agrability.org/

Personalisation matters more to APA at the mo.

We have a spec in Personalisation TF.

CSS Media Queries gives us more support for these things.

Things are ripening.

JW: Any other comments on that perspective?

SN: I posted link to national Agri project.

JB: Great.

JOC: Should we gather those themes?

JS: There is a sales, cash money aspect here.

In terms of the specific area - we don't really care, as long as they know the UX matters.

SH: I looked at these, and struggled to think how we can contribute beyond me too and +1 etc.

But it was a struggle tbh

JB: Skimming through..

There is a compliance element here - people may not make the connection as they may default to stereotypes.

We can build hooks in for various domains - where functionality x needs to be accessible.

JB: Discusses farming and contemporary counterparts and compliance.

JS: Concurs and extends..

SOme of the domains Josh brought include home automation, vehicles, public spaces with shared tech etc

So there is diversity..

But not as broad as what Janina was suggesting.

We should think about common themes, with MMI, personalisation etc

There is a commonality of interest between these things.

JS: Sounds like agreement - there is an APA specific one - that is the portability of UIs.

+1 to Janina

JS: This is one of the ways a UI becomes personal.

We can also bring in the relevant APA work we have to support this.

There may be API specific stuff for Ag, or factory stuff.

Portable personalisation is horizontal

SH: In terms of feedback how do we do that?

JOC: WoT have GH, mailing list etc - I'm happy to mail stuff for chairs or do all three.

SH: What are they looking for? What do we need to do?

JS: I think they are trying to prioritise them.
... We have made our point, over which ones to go for.

We are offering that we can support them in area x or y.

There is a commonality

JW: If we do identify areas they are considering, thats great - we can identify more. That is RQTFs role.

So what is the next move?

JOC: Gives overview of next steps

JB: +1 to Josh

SH: +1

JS: I nominate Jason, they have heard from Josh already and me on the survey.

Coming from Jason, the locus of activity would be in RQTF.

JW: I'll draft something.

JS: Sounds good.

JW: We can develop on list - something else that occurred to me - a strategic question.

The interplay between having a UI reside in the environment or in the users device.

How is that division of responsibility? I've spoken with Mark about this, a la GPII.

I'm not sure how interplay works technically

JS: I think its a case of both/and

<gives reasons>

Some are not powerful enough to build UIs.

Lots of low tech devices that will be aggregated at a collection point

Low power protocols are important..

Then there are smart assistants.

More powerful, can be dealt with directly or on a mo phone

<gives home example of listening to book in multiple rooms at home while moving around house>

JB: We need to support these universal applications

JW: In some contexts it is important for the device to have UI, may have speech UI etc

In emergency, accessible options are important.

There is interplay

Flexibility needed.

We will need protocols and applications that support multiple ways

JS: Interoperability is important so we can mix and match.

I was happy to play music from A lady upstream via G lady speakers.

JOC: We could pimp the WCAG SC on multiple ways to cover sensor enabled networks and devices that are available/able to generate UIs

JS: Sounds interesting

I'd like to have a way to direct non generating data streams where we could build our own devices.

WIRED is talking about building apps without righting code.

Aggregating UIs based on data needs and preferences.

JOC: Just to say again that this is great WoT are actively looking for a11y feedback for these use cases etc.

<good stuff all round>

JS: Next step is on me to circulate and we will take it from there.

JW: Anything else?

Web and Machine Learning workshop

JW: This is for APA, and making decisions on participation?

JS: This has generated a lot of interest in the APA meeting - we need to be there to learn and point out some critical stuff and the implications.
... Is Jutta planning on being involved?

JB: Yes, she has submitted.
... <gives background>

There may be multiple a11y presentations.

Lead staff on this is aware of need - a robust set of presentations is needed.

Current ones look at risk of AI bias etc based on existing research etc.

Datasets can be trained to overcome bias.

Looking at disability as functional outlier - business goes for central core use cases - you need to design for edges etc.

More extreme data sets are needed.

JS: Jutta has good stuff there.

Do we want to make her one of APAs voices?

JB: I've done that with WAI Director hat, but sure, would be nice if APA can as well..

JS: Good.

https://www.w3.org/WAI/APA/wiki/MLWorkshop2020

JW: We have an XR question.

JB: The discussion that Janina and Josh has started in Silver, I understand where that comes from..

This is scribe.perl Revision of Date 
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Succeeded: s/Director hat/Director hat, but sure, would be nice if APA can as well./
Succeeded: s/presos/presentations/
Default Present: jasonjgw, Joshue, scott_h, SteveNoble, becky, janina
Present: jasonjgw Joshue scott_h SteveNoble becky janina Joshue108
Found Scribe: Joshue108
Inferring ScribeNick: Joshue108
Found Date: 01 Jul 2020
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)