W3C

– DRAFT –
WoT-WG - TD-TF

19 January 2022

Attendees

Present
Cristiano_Aguzzi, Daniel_Peintner, Ege_Korkan, Jan_Romann, Kaz_Ashimura, Michael_Koster, Michael_Lagally, Michael_McCool, Sebastian_Kaebisch, Tomas_Jaeckle, Tomoaki_Mizushima
Regrets
-
Chair
Sebastian
Scribe
kaz, sebastian

Meeting minutes

new member

Bosch joined the WoT group yesterday

Thomas Jäckle representing Bosch in WoT

Thomas: hi everyone
… I'm contributing to the Eclipse Ditto project
… currently working on integration of WoT in Ditto. Using the Thing Model approach e.g.

minutes check

check minutes from Dec 15

https://www.w3.org/2021/12/15-wot-td-minutes.html

McCool will check this issue https://github.com/w3c/wot-thing-description/issues/1211

any objections?

no

next minutes from last week

https://www.w3.org/2022/01/12-wot-td-minutes.html

any comments, objections?

no

Webhook

<kaz> Issue 1323 - Missing event/notification affordance or operation

Lagally: happy to have Bosch onboard

Lagally: let me give the background again
… everything what has an interface a TD can be designed
… its mainly works for IP based connections
… a servient is a combination of a consumer and provider
… everything is a Thing and the TD describes the Thing's interface
… we have properties, actions and events
… for event affordance we get notifications

<ML shows the place of the EventAffordance definition in the TD spec>

https://w3c.github.io/wot-thing-description/#eventaffordance

Lagally: a consumer Thing do a subscription
… consumer gets notify if values are changed
… if there is a battery powered device it can not open the network the full time
… a consumer register a listener at a Thing
… we have TD for the consumer which is part of the event mechanism
… we need a kind of callback functionality for the event affordances
… it's an abstract model without specific protocol
… it would be good to have a notify in the event affordances

Ege: I agree with the use case
… however, it is very protocol driven use case
… TD needs 2 kind of consumer

Lagally: sounds to complicated which seems not needed
… only one TD is needed

Ege: with notify it is pushed to the listener that was registered

Lagally: the registration can a TD which would be a solution

Lagally: TD spec is silent where I have to provide the listener information, e.g.

McCool: What I understood so far, there is missing information to initalize and execute Webhook based interaction

Lagally: shows the PR with a proposal

https://github.com/w3c/wot-thing-description/pull/1329

McCool: looks ok for me this kind of approach

Ege: Im wondering how the forms look like
… would be good to have TD example

Lagally: ok I will take this action

Lagally: dataresponse is the response of the data notification
… I need to go

McCool: lets continue this discussion in the issue and do discussion next week again

<kaz> diff version from PR 1329 - 5.3.1.5 EventAffordance

Seb: Im ok with the dataresponse, however, not sure about the op type "notify"

Chris: I'm not sure about the use case which is ML looking for
… by definition I'm not a consumer of the TD any more as a consumer, it is also a producer

<Ege> +1 for kaz

Kaz: We need a detailed use case description with concrete device setting, data transfer and interaction protocols, and should ask Michael Lagally to work on that. Then we can talk about what is needed based on that.

McCool: I'm concerned about duplicated definitions if TD is used to define the target resources
… we need to understand more a practical use case
… however, I understand this main intention of the use case

Chris: If you understand this use case why we cannot use a simple protocol for this?

McCool: we need only an action that uses the notify op
… however, issue came very late, needs more clarification

publication plans

2nd WD

Sebastian: made resolution about publication the 2nd WD
… in the main call

Kaz: will work on the publication. would expect it'll be ready next week., however, depends on issues which may arise

Updated schedule

<kaz> proposed updates

Wide reviews

Sebastian: I will start to do the wide review process until end of this week for the TD spec

check PRs

#1282

https://github.com/w3c/wot-thing-description/pull/

decided not to merge

#1317

https://github.com/w3c/wot-thing-description/pull/1317

will be addressed until CR transition

Change log for the 2nd WD

Kaz: btw, we do need a change log section from the previous WD for the 2nd WD publication

Sebastian: good question. need to talk with Taki

#1319

https://github.com/w3c/wot-thing-description/pull/1319

decided to merge. Has no impact on the TD spec itself

#1327

https://github.com/w3c/wot-thing-description/pull/1327

I need to check this PR before decided to merge

Daniel: we should be careful about this fix namespaces

add discussion label

#1328

https://github.com/w3c/wot-thing-description/pull/1328

Kaz: this file was used for publication in the past
… is not needed any more

decided to merge

#1342

https://github.com/w3c/wot-thing-description/pull/1342

needs further discussions

TD Issues

#1349

Clarifications on readmultipleproperties and writemultipleproperties / writeallproperties

https://github.com/w3c/wot-thing-description/issues/1349

#1357

<kaz> https://github.com/w3c/wot-thing-description/issues/1357

<kaz> (continue the discussion next week)

Binding

Sebastian: let's discuss wot-binding-templates issue 144 next week

<kaz> [adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).