Accessible Platform Architectures Working Group Teleconference

14 Oct 2020



Kaz_Ashimura, Janina_Sajka, Josh_O_Connor, Michael_McCool, Becky_Gibson, Cristiano_Aguzzi, Sebastian_Kaebisch, Jason_White, Kunihiko_Toumura, Martin_Alvarez_Espinar, Tomoaki_Mizushima, Ken_Ogiso


<scribe> scribe: Joshue108

<kaz> agenda: https://github.com/w3c/wot/issues/938

JS: I'd like to make some clarification on the use cases

Next steps

Use cases review identify those with a11y component

MMcC: Any other agenda?

JS: This may be a best practice.. but want to declare specific eco-systems that a device may sit in.

MMcC: We have a security best practices doc

We may need an accessibility best practice for IoT

MMcC: Will add

MC: Please look at this URL

<McCool> https://github.com/w3c/wot-usecases

We have been organising use cases

<McCool> https://github.com/w3c/wot-usecases/tree/master/USE-CASES

This is a template

You can fill in owner, device data etc

There are sections for security and privacy and we should add accessibility use cases also.

JS: We will happily work on that.

MMcC: I'll work on that now.

There may also be i18n and security

JS: Experience has taught us this is the case.

MMcC: I'll add

<McCool> https://github.com/w3c/wot-usecases/tree/master/USE-CASES/processed

We have also created sub dir with several MD files.

MMcC: <Gives overview of process>

<McCool> https://w3c.github.io/wot-usecases/

The use cases are in the index.html file

We have an editors draft - still organising.

We should go thru all the use cases and consider the accessibility use cases.

There are verticals like agriculture and horizotnals

Some accessibility cases evolved out of MMI

JS: Interesting

MMcC: We can consider things like farmers with disabilities, or drivers.

We need to expand on these.

MMcC: <Gives overview of accessibility use cases>

Your audio visual system as IO device

MMcC: Gives overview of Smart Home use cases

There are others that have accessibility components./

We also have the Linked Data meeting coming up - that relates to hotels, and airports etc

Regarding Smart Buildings - how can we make the use of similar interfaces happen in buildings?

BG: What about electronic keys.

JS: We have a deep history in that..

Not in W3C but there was an ISO group that was looking at this - how to control devices in public spaces

There were use cases captured.

The ideas may be useful here.

Greg Van was leading that.

MMcC: Do you know the ISO number?

JS: Not smart keys, more generic than that.

BG: There is an ISO spec we looked at when I was at IBM.

Credentials relating to door locks - it can carry over into the lock for security reasons

MMcC: there was a comment about professional users.

<jasonjgw> https://www.iso.org/standard/60821.html

MMcC: THere is a transportation use case

Logistics and deliveries

Package deliveries etc

No contact delivery a la COVID

Lockboxes etc

<McCool> https://github.com/w3c/wot-usecases/issues/64

MMcC: Publishing etc we need to create.. we had a discussion with Singapore and Smart cities

JS: For any kind of UI there are a11y issues.

These are familiar for those working in this area

JS: There are vertical integration needs that we should look at.

When trying to integrate.

JS: So we have controls but we may need APIs.

We should identify these spots as we go thru these lists.

<becky> bg: aside: iso compliant driving licence identity and integrity validation overlaps with some use cases. ISO/IEC 18013-3:2017 (https://www.iso.org/standard/72366.html)

JOC: One of our concerns was with middleware right?

JS: Yes.
... Companies will want to brand this and keep you in your ecosystem

but interoperability benefits e'one.

<Zakim> Joshue, you wanted to remember the feedback APA game on existing use cases



JOC: You have seen the APA feedback

MMcC: Thanks for that, apologies

JS: We have been looking at this for a while - remembering WoT F2F in Berlin.

We are happy to engage now.

MMcC: There is more to do, esp around semantics.

JS: Sounds good

MMcC: I want to mention publishing

JS: Great - we are meeting them on Thurs.

<McCool> https://github.com/w3c/wot/blob/master/PRESENTATIONS/2020-10-joint-wot-pub/2020-10-WoT-Intro-Pub.pdf

MMcC: I need to write these up. We have brought up the eBook as an IoT device

There are affordances such as remote control

and with hybrid device used to interact with content

Receipe books

MMcC: Will right that up.

Semantic extensions

MMcC: We have talked about this not a lot of progress

<McCool> https://github.com/w3c/wot-testing/blob/master/events/2020.09.Online/TDs/Intel/intel-ocf/Intel-OCF-button1touch.jsonld

<walks through example of using thing descriptions and JSON-LD>

At plugfest we looked at Geolocation data

And one data model - there is a Zigb and ZeeWave project looking at new semantic models

As defined by One data liason and annotating what to do with them

<states and capabilities are defined in the data model>

States can be mapped on alternative output devices

There could be affordances for blind people

These can be mapped for various sensory modalities.

We did realise that not all switches are controlable

However, they may be read only or controllable.

This isn't always clear

There is not indication of what the UI looks like as such

A light may or may not be there, to indicate state.

So some semantic annotation is needed

JS: Some of the older work will help

JW: I've found that spec and put it in IRC
... It has different components

It also mentions the discovery protocol

MMcC: I'll add that to the use cases doc

JS: We may even get Gregg V to talk about this.

MMcC: I'd like to have an RDf context file that gives the standard terms for annotating IoT devices

JOC: The annotation semantics should be generic and used with other arguments or logic to determine implementation and serving various modalities

MMcC: If something is accessible may need less work.

Remapping may be between custom UI, browsers etc

JS: If we get this right, this may facilitate other things we have not thought of

The kerb cut effect - build something for one group and it benefits others.

Proven to be successful.

Feeds into Machine Learning etc

MMcC: Regarding i18n - we are working here also.

<gives i18n example and need for base directions>

This can't be done easily in RDF.

Right now it is inferred - some languages may be in opposite directions.

The work around is to start in the language that you want.

There may be impact in text to speech systems

JS: I'm thinking of our Personalization spec

<becky> Personaliation video: https://www.w3.org/2020/10/TPAC/apa-personalization.html

JS: People with cognitive disabilities use symbol sets.

They are learned from school and are proprietary

MMcC: We are also talking about using sort of HTML hacks to indicate direction of language.

Could symbol sets be added like this?

JS: We can dive into the overlay approach that we are looking at now.

Not sure how it will apply to your use case, but think there is a nexus.

MMcC: I will capture this - we will do use cases, and then we will boil these down to requirements

Requirements go into our architecture document which is normative.

MMcC: We are also working on a new spec for discovery.

imagine you could access location of a11y services and ad hoc connect

and find IoT services that you need.

JS: This is great - there are many connections.

MMcC: Anything else?

One other point is maybe we can infer UI requirements from higher-level semantics

I will write this up

We should plan to meet again after incorporated

We have work to do.

It would be nice to have by the end of the year.

Lets meet again in December.

JS: That works
... We can do this Weds - even at this time, our regular call time.

MMcC: Mid December is good.

BG: 9th of Dec is good

JOC: Works for me

9 am Boston on the 9th for follow up

<Zakim> kaz, you wanted to say thank you very much to all (later)

<kaz> kaz: note 9am Boston on Dec 9th is not the WoT main call but the WoT PlugFest call

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version (CVS log)
$Date: 2020/10/14 14:03:00 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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/wot-joint-attendees:/present:/
Succeeded: s/ecosyste,m/ecosystem/
Default Present: janina, plh, CharlesHall, Chris_Needham, jeanne, Francis_Storr, Lauriat, Nigel_Megitt, becky, mikecrabb, jasonjgw, MelanieP, Joshue, SuzanneTaylor, KimD, jib, martin, Ken_Ogiso
Present: Kaz_Ashimura Janina_Sajka Josh_O_Connor Michael_McCool Becky_Gibson Cristiano_Aguzzi Sebastian_Kaebisch Jason_White Kunihiko_Toumura Martin_Alvarez_Espinar Tomoaki_Mizushima Ken_Ogiso
Found Scribe: Joshue108
Inferring ScribeNick: Joshue108
Agenda: https://github.com/w3c/wot/issues/938

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 14 Oct 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.)

[End of scribe.perl diagnostic output]