Main WoT WebConf/2023 WoT TPAC Agenda

From Web of Things Interest Group

Please find TPAC details here

TPAC Agenda

Presentation material is (should be) archived here:

Joint Sessions

Joint sessions held on Mon and Fri at the other group's room

WoT-WG/IG participants will visit the other groups' meeting roomss.

Joint sessions held during the WoT WG/IG Meeting on Thu-Fri

Related groups' participants will come to the WoT WG/IG meeting room, Azalea, Low Level.

  • Thursday (Sep 14) 17:30-18:30 (1h) Accessibility
  • Friday (Sep 15) 17:00-18:15 (1h15m) WoT CG (New Commercial Use Case Session)

Thursday (Sep 14) 14:30-18:30 CEST (4h) : WoT WG/IG Meeting - Day 1

Event link:


  • 14:30-14:45 (45m) Opening, Introductions, Joint Session/Breakout reports (Sebastian)
  • 14:45-15:30 (30m) Use Case and Requirements Process (McCool)
  • 15:30-16:45 (1h15) TD and Bindings (Sebastian, Ege, Koster)
  • 16:45-17:00 (15m) Break
  • 17:00-17:30 (30m) Discovery (McCool)
  • 17:30-18:30 (1h) Accessibility (McCool) - Joint

Friday (Sep 15) 14:30-18:30 CEST (4h) : WoT WG/IG Meeting - Day 2

Event Link:


  • 14:30-14:40 (10m) Opening (Sebastian)
  • 14:40-15:00 (20m) Outreach: WG involvement, CG cooperation, IIWoT Workshop, Digital Twins, SDO collaboration... etc. (Sebastian)
  • 15:00-15:30 (30m) Security and Privacy (McCool)
  • 15:30-16:15 (45m) Profiles (Luca)
  • 16:15-16:45 (30m) Architecture (Koster)
  • 16:45-17:00 (15m) Break
  • 17:00-18:15 (1h15m) New/Commercial Use Cases and Requirements (Kaz) - Joint
    • including visitors: PBG, MEIG, Second Screen, Devices and Sensors, WoT CG, WoT-JP CG, etc.
  • 18:15-18:30 (15m) Wrap Up and Next Steps (Sebastian)


  • Plugfest: Before or after the TPAC week


Meeting Topics

Each TF should collect topics they want to discuss and list it below.

  • Planning
    • Prioritize work items
      • Scope of each deliverable
    • How to handle versioning
    • Clarify relationship of Profiles and Binding
    • Decide on normative status of Architecture
  • Policies
    • WG/IG/CG Relationships
  • Liaisons
    • OPCF


Meetings afternoon on Thursday (Sep 14) and Friday (Sep 15): schedule confirmed

Note: Kaz has conflicts on Monday and Tuesday, and Wednesday is breakouts.

Joint Calls

Proposed Joint calls, with organizers.

  • JSON-LD (Sebastian/Ege?)
  • SDW (McCool)
  • Privacy (McCool/Sebastian)
  • Accessibility (McCool)
  • MEIG (Kaz)
  • WoT CG (Ege)
  • WoT JP CG (Mizushima/Toumura)
  • Web Agents (Ege)


  • Linked Building CG (Ege/Sebastian?)
  • Digital Twin (Kaz)
  • TAG (Sebastian/McCool)

Agenda Items

JSON-LD, RDF Dataset Canonicalization and Hash

Organizer: Sebastian and Ege

Monday 14:30-16:30:


  1. Introduction of Participants (10 mins)
  2. Short Introduction of WoT (10 mins)
  3. Signing and Canonicalization (30 mins)
  4. Degraded Consumption (20 mins)
  5. Additional Serialization (15 mins)
  6. Linting (15 mins)

Agenda Details:

  1. Canonicalization and Signing (Mahda)
    • How can we use the JSON-LD canonicalization and signing mechanisms to guarantee the authenticity of Thing Descriptions (in a TD Directory).
  2. Restricting prefixes for JSON (no JSON-LD) parsers (Ege+Luca?)
    • There are more and more (protocol) vocabularies that we want to include in the TD spec (as a registry). However, we have TD Consumers/Parsers who are not JSON-LD processors. Ideally, we want to restrict what prefixes should be used in the TD for a given vocabulary. Are there any mechanisms for that, any potential pitfalls?
  3. (related to the point above) Supporting JSON-LD without full JSON-LD processing (Ege+Luca?)
  4. YAML-LD (Mahda)
  5. Linting (Ege)
    • API Description Linting is a recent topic of relevance (see and since TDs are API descriptions of IoT devices, this topics will become important for us. Are there any mechanisms for JSON-LD or JSON-LD linting that would be of relevance to this topic?
  6. (Related to the breakout session below) How to work with a single source of truth for generating our spec? Currently, we use SHACL Shapes + handwritten template HTML and then we try to manually maintain a JSON Schema, JSON-LD context

Related breakout session:

  1. "You say Schemata, I say Schemas"

Others (maybe also under SDW):

  1. ?WebVMT - also related to geolocation, XML


Organizer: McCool


  • Invitation to Christine Perey sent discussing options.

Possible Topics:

  • Thing Description metadata
  • Dynamic geolocation data schemas
  • Combinations of the above: unification of static and dynamic data
  • Discovery ("Things near me", etc)
  • Privacy


Organizer: McCool/Sebastian


  • Invitation sent...

Possible Topics:

  • Classifications of personal data (e.g. how do we know a TD or a Thing should be treated as personal?)
  • Improvements to Discovery
  • Impact of proposed changes, e.g. geospatial data/location-based Discovery
  • Normative status of Privacy Considerations; testability


Organizer: McCool


  • Invitation email sent to Matthew Atkinson
  • Comments on agenda items and proposed times received from APA (added below)

Possible Topics:

Some possible times (proposed by APA):

  • Thursday: 1700-1800, or 1730-1830 if that is better for you

Discussion in WoT WG: Thursday 1730-1830 preferred

  • WoT Schedule is however full
  • Can we schedule a slot in the mornings during non-overlapping APA slots?


Organizer: Kaz


  • Contacted MEIG chairs and PBG Chair

Possible Topics:

  • MEIG agenda is packed but Chris Needham can join personally.
  • PBG can have a joint meeting on WoT for Publishing business.
  • A/V streaming, smart TVs, (data) publishing services (e.g. weather, traffic, etc), managing copyright, etc.


  • we should organize a use case session and invite MEIG/PBG members to this session


Organizer: Ege and Cristiano

The meeting slots of the CG are:

  1. Thursday 11:30-13:00
  2. Friday 11:30-13:00
  3. Friday 17:00-18:30

The agenda of the CG is available at . It is also copied below for convenience. The joint session is in Slot 2.

Slot 1

Time: Thursday 11:30-13:00

  • 11:30 - 11:45 Introductions of Participants
  • 11:45 - 12:15 Explaining Web of Things
  • 12:15 - 12:45 Hands-on
  • 12:45 - 13:00 Wrap-up and Discussions
Slot 2

Time: Friday 11:30-13:00

  • 11:30 - 11:45 Introductions of Participants
  • 11:45 - 12:00 Short Introduction to Web of Things
  • 12:00 - 12:15 Introducing WoT CG
  • 12:15 - 13:00 Discussion with WoT WG (joint session) (requires confirmation from WoT WG)
    • Binding Contributions from CG Participants
    • Feedback from CG Participants to Use Cases, WG documents
    • Overal agreement on scope
Slot 3

Time: Friday 17:00-18:30

  • 17:00 - 18:30 Alternative slot for WoT WG Discussion (requires confirmation from WoT WG)


Organizer: Mizushima/Toumura

Autonomous Agents on the Web CG (aka Web Agents)

Organizer: Ege

Link to Group:

Meeting Slot: Monday 11:30 – 13:00:

  • Kaz has a conflict at this time due to MEIG
  • Sebastian will be able to attend
  • (they also have another slot from 9-11)

Tentative Agenda:

  • [15 min] WoT intro
  • [15 min] WebAgents intro
  • [1h] Open discussion, overlaps, and common objectives:
    • In general, we are interested in incubation-level topics that can interest the Agent community. How such topics become standards at the W3C is currently out of scope. The topics are listed below
      • Consumer Descriptions and Proactive Things
        • The current WoT specs mostly target use cases where the Thing is mostly a reactive component like a server where clients (Consumers) send a request and the Thing responds. There are cases like MQTT using Things where we describe what the Thing is doing proactively but this is a very simplistic example. We believe this kind of thinking is necessary to start covering use cases where the Things are Agents (in the web). More specifically, what we do not have in the WoT:
          • We do not really explain how a Thing acts as a client to do something (like sending data to a (cloud) server) nor how it can be configured dynamically to send data to Cloud instances (see Webhooks). If the Thing accepts a certain description format to describe where it should send data, we can cover this use case. This kind of description format can be called a Consumer Description from one point of view.
          • We are not able to represent a system where there are controllers (Consumers) who act on the Thing and what requirements there are on the Consumers. This representation would be the second point of view for a Consumer Description.
        • There are possibly other use cases for this and incubation of use cases would be appreciated.
      • Manageable Actions
        • Action Affordance in the current WoT specifications is rather simplistic, where a Consumer sends a request and the Thing responds. Whether the response corresponds to the result of the physical action process or whether it is just an acknowledgment is unclear. Furthermore, we are representing an action that can be managed due to it taking a long time. There are examples such as a long-lasting dimming of a light or long-lasting movements such as robots, and cameras (panning, tilting). Given that the hypermedia community is also represented in the Autonomous Agents on the Web CG, we would benefit from such discussions and expertise.
          • Note: This topic is already in the scope of the next WoT WG charter.

Digital Twin IG

Organizer: Kaz


  • New Charter

Thing Description

Deadline for contributions: 2 August 2023 (Until the TD Call). Please think of a concrete Use Case for the topic you want to discuss to help with the discussion. We will update the agenda based on the relevance of the topics before the TPAC. In case a topic does not fit into TPAC, it can be discussed in the pre or post-TPAC F2F.

  1. TD in RDF: How well TDs work in RDF ecosystems? How rich is the TD in RDF (do we need manual work)? (Michael Koster)
  2. Data (Payload) mapping in TDs to protocol (Luca B.):
    • How to accurately describe abstract (Data Schema) and concrete data (Protocol Payload)
    • Which level of TD should describe what kind of data
  3. "Connected Protocols" where an initial connection needs to be preserved or recycled. (This can be made a general WoT discussion topic). This can be seen in practice in NodeRED.
  4. Relationship between forms of an affordance (subscribe-unsubscribe) and how to find the form you can use. This can be part of the normative consumption step but it can bring restrictions to how a TD is designed/created (e.g. do not put subscribe if there is no unsubscribe)
  5. Manageable Actions (Asynchronous, queriable etc.) which can be part of the Affordance Design discussion on how to make them more uniform. It can be interesting for the hypermedia community.
  6. Showing planned activity, i.e. TD Items:
    • Showing the direction that the next charter work is planning to increase (commercial) WoT deployments in real-world applications
    • Explaining the major version update (TD 2.0) and having consensus on this. This will/can include major refactoring.
  7. Bindings Registry (Ege+Koster)
    • Normative Bindings: document status (REC or not etc.), testability (how to do that, what is in our scope to mandate), what do we "guarantee" to the users of the bindings
    • Submission Process: who and where they do what
    • Experience with registry mechanism from other groups. Contacting editors to learn from them?

TODO: Check that all items are in the TD work items list.

TPAC is special because:

  1. First F2F since Covid
  2. More people than usual and people who are not usually in WoT calls

Thing Description Final Topics

  1. Summarizing planned activity, i.e. TD Items: 20mins (Sebastian)

    • Showing the direction that the next charter work is planning to increase (commercial) WoT deployments in real-world applications
    • Explaining the major version update (TD 2.0) and having consensus on this. This will/can include:
      • major refactoring
      • use case review
      • pain points of TD 1.1 for implementers
  1. Bindings Registry: 20mins (Ege)
  2. Manageable Actions (Asynchronous, queriable etc.): 5mins (Koster)
    • which can be part of the Affordance Design discussion on how to make them more uniform. It can be interesting for the hypermedia community.
  3. Q&A: 30mins (Sebastian)

Binding Templates

  1. Ontologies for Protocols. How well do they work? Fixed prefixes (Cris)


Organizer: ?

Duration: 2h

TODO: Add some detail to the below, links, etc.

  1. Review high-priority work items and use case/requirements analysis


Organizer: McCool

Duration: 30m (20m presentation, 10m discussion)

  1. Discovery Use Cases and Requirements (10m)
  2. Review high-priority work items (5m)
  3. Publication finalization for last charter (5m)
    • DID Ontology
  4. Discussion (10m)
    • How to update and document requirements?
    • What format, where to put them, how to cross-reference with use cases?
    • Work items should then reference requirements (implies requirements should be linkable, e.g. should have id anchors, etc.)
    • Should document both satisfied and unsatisfied requirements
    • New requirements


Organizer: McCool

  1. Reorganization and Refactoring
  2. Document purpose
  3. Normative status

Security and Privacy

Organizer: McCool

Duration: 30m (20m presentation, 10m discussion)

  1. Review Threat Model vs. Considerations (7m)
  2. Review Use Cases and Requirements (8m)
  3. Proposed Work Items and Reorganizations (5m)
    • In general:
      • How do we deal with security and privacy considerations?
      • Should they be normative or not?
      • How and when should we define best practices?
      • How does this overlap with existing mechanisms in protocols, etc.?
      • Deployments vs. specification features/design?
  4. Call for Action/Discussion (10m)
    • Need more people to work on the above...
      • Since security experts are valuable and their time is limited, can't do all the work
      • Instead, want to engage experts for guidance and review, with other tasks taken up by individuals and each task force
    • Proposal:
      • Align considerations and risks in S&P document
      • Consistent names, cross-references/link, etc.
      • For each threat, identify use cases (or categories of use cases) that are impacted by it, and document these in the use cases document
      • Since a given threat may apply to many use cases, categories (e.g. safety-critical, handles PII, etc.) may be a better way to handle this

Use Cases

Organizer: McCool

  1. Process
  2. Requirements Gathering
  3. Connecting Features to Requirements to Use Cases
  4. New Use Cases with focus on Commercial Applications


In Person

  • Sebastian Kaebisch
  • Ege Korkan
  • Kaz Ashimura
  • Kunihiko Toumura
  • David Ezell
  • Daniel Peintner
  • Hiroki Endo
  • Ryo Yasuoka
  • Hisayuki Ohmata
  • Cristiano Aguzzi
  • Mahda Noura


  • Christian Glomb
  • Michael McCool
  • Michael Koster
  • Luca Barbato
  • Jan Romann
  • Zoltan Kis
  • Tomoaki Mizushima