W3C

WoT Use Cases

27 July 2021

Attendees

Present
Chinn_Hwa_Lim, Ege_Korkan, Kaz_Ashimura, Michael_Lagally, Michael_McCool, Rainer_Sciekofer, Sebastian_Kaebisch, Tomoaki_Mizushima
Regrets
-
Chair
Lagally
Scribe
kaz, sebastian

Meeting minutes

Guest

<kaz> W3C Patent Policy

Lagally: Rainer Schiekofer from Siemens is a guest for the OPC-UA collaboration topic today, and aware of the W3C Patent Policy.

Minutes check

July-13

Lagally: was about ECLASS

any objections?

no

Lagally: Christian Block provided the slides from last time

<mlagally> May-25

<mlagally> https://github.com/w3c/wot-usecases/pull/138

Lagally: lets check the minutes from May 25

any objections?

no

OPC-UA collaboration

Rainer: give presentation what is cover by OPC UA
… first of all OPC UA is not a protocol
… is a collection of technologies to ensure a secure exchange of standardized information from the sensor to the cloud
… OPCF membership 820
… why companies join the OPCF?
… guarantees early information to upcoming key technologies
… increasing number of companion specifications
… currently 63
… new specs addressing information models
… there is a huge number of liaison partners
… like with W3C

Rainer: shows an example of liaison partnership in the Process Automation domain

… currently working on OPC UA FX. First live demo at SPS in November 2021

Rainer: OPC UA has two communication paradigms
… client - server and pub / sub
… pub / sub also transport meta data
… also support end2end encryption

MC: can you also use different certificates than UA based ones?

Rainer: yes

Rainer: UA comes also with a subscription feature
… the information model is based on variables, object, data and reference type
… you can transform OPC UA to an ontology

Rainer: how to create a companion specifications

<mlagally> @Kaz, Ege - let's handle questions at the end

<Zakim> kaz, you wanted to ask about binding for different protocols

Kaz: we should think about a possible binding. UA can use different protocols. Is this kind of binding addressed?

Rainer: yes, we should address the different communication paradigm, the interaction patterns and the security

Ege: when I do a cloud communication, do I need a specific libraries?

Rainer: there are some activities that is called CASY(?)

Lagally: is there some plans to have Information Model as JSON?

Rainer: we had the plan to use JSON-LD, however, there is no ontology available

Lagally: there is end2end security, can you exlplain more in detail

Rainer: there is not defined which part of the JSON have to be signed

Sebastian: there is already a liaison between OPC and W3C
… very good time to work more actively
… the spec is getting stable
… the question is what should be the target
… e.g., how to specify OPC endpoint
… integrating applications
… how to describe the interface
… how to setup the OPC UA interfaces
… very clear to the clients
… see some activities on defining TDs compatible with OPC UA
… we can use the information to generate actual TDs
… mapping to the TD
… methods equivalent
… also think about communication information need to be provided
… for OPC servers/clients
… what kind of endpoint/node information to be transferred
… based on some specific TDs
… open questions there
… also security clarifications
… metadata in TDs
… possibly to be transferred to the OPC UA endpoints
… next slides (18)
… joint activity proposal
… OPC UA context to be referred to from TD
… metadata for UA security mode
… equivalent variables
… metadata for UA communications
… addressing scheme and URL patterns
… content types, e.g., for UA binary and XMLs
… local security setup
… custom UA options
… next
… Web of Things s bridging technology
… win-win situation
… industry specific semantic on the OPC UA side
… rich tool landscape on the WoT side

Lagally: remaining slides to be discussed next week?

Sebastian: I'm OK
… but what about you, Rainer?

Rainer: will try

kaz: wanted to suggest we discuss the next step, e.g., use case description and plugfest collaboration

McCool: wanted to talk about security but can wait until next week
… also thinking about possible binding based on the use case description
… but can wait until next week

[adjourned]

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