Meeting minutes
Agenda
McCool: we should talk about the next meeting too
Kaz: btw, for today, I can take notes
McCool: tx. should have a scribe list too
… also should clarify who/how to chair the call too
Minutes
McCool: (skims the minutes)
… several typos, e.g., "ThingModel" to be split
(fixed)
(approved)
Logistics
McCool: have created a dedicated GitHub repo, Wiki page and IRC channel for this liaison
… also the W3C Liaison table has been updated
Kaz: yes, updated with the liaison contacts
Next meeting
McCool: in two weeks, we'll have a F2F meeting in Munich
… so would cancel the NGSI liaison call
… the next meeting after that will be Dec 9
… what about the timezone?
Kaz: for the liaison discussion during WoT Week?
McCool: would have presentation from ETSI during the WoT Week meeting
… would expect a slot of half an hour or so
Franck: F2F?
McCool: can join the meeting remotely as well
… a Zoom call will be available
… the date will be Thursday/Friday
(Nov 28-29)
McCool: let's clarify the timezone
Franck: located in Munich. Thursday after 4pm
… Friday is OK. no constraints and can accommodate
Martin: all day Thursday
… not 2-3pm on Friday
McCool: would expect a 30min slot
… including 15min presentation and 15min discussion
McCool: (then shows the WoT Week wiki)
McCool: regarding the liaison call, let's cancel the call
… (put "Cancelled") on the NGSI-LD wiki
McCool: regarding the time for the presentation during WoT Week, we'll finalize the time on Wednesday (during the Plugfest call)
Chair
McCool: is anybody from you interested in chairing the call?
Kaz: given the mechanism of W3C calls is a bit complicated, I'd suggest we continue to ask you, McCool, to chair the call :)
McCool: ok
Deliverables and Resources
wot PR 1209 - Activate Liaison for ETSI ISG CIM for NGSI-LD
McCool: got a comment from Ege Korkan (from Siemens)
McCool: (adds some edits directly)
… we don't usually say "Things Descriptions"
… should be "Thing Description"
… Thing Model describes some class of devices
… if NGSI-LD behaves as a broker, it's kind of similar to the WoT Thing Directory
… so it's not just about Binding Templates
Martin: not 100% sure...
… TD basically can describe how the NGSI-LD can work with it
McCool: NGSI-LD itself is still a REST API
… here the description is a bit vague
Franck: there were two bullet points there
… and those points have been consolidated as a possible deliverable
Juanjo: the first point is describing what W3C could publish
… while the second point describes the deliverable by the ETSI side
… WoT Thing Model describes how a Thing Description can be derived to describe a device
McCool: the idea here is what the Consumers of WoT would be like, e.g., for Smart Cities
… I actually like the first part here
Juanjo: considering NGSI-LD as a form to handle devices
… a possible context broker could access NGSI-LD
McCool: should extend NGSI-LD by some standard protocol
… NGSI-LD broker needs to generate WoT Thing Description to consume a Thing
Juanjo: WoT Thing Description can be adopted
… then the NGSI-LD broker will handle actual devices
… using the WoT Thing Model mechanism
McCool: NGSI-LD is a proxy?
… Things have to be devices
… this idea itself is correct, but need clarification
… could do something like this...
Kaz: two comments
… 1. for today, we should add two sub sections, (1) expectation for W3C and (2) expectation for ETSI
… 2. we should invite the ETSI guys to the WoT Week meeting not only for the presentation but also for some more detailed discussion about our expectations for the possible deliverables
… probably, we should clarify some concrete use cases and also drawing a basic diagram of expected devices and applications (and brokers) would be helpful
Franck: adding information to the "Scope" section also would make sense
McCool: (adds "W3C WoT WG to develop..." and "ETSI ISG CIM to develop..." to the Scope section)
… W3C: NGSILD Binding that defines what infrmation is needed in a Thing Description to communicate with an NGSI-LD Context Broker.
Juanjo: not really sure about that
… Thing Description is a paper which describes Thing's capability
… Context Broker accesses Things and Entities.
McCool: think we're on the same page...
… WoT Consumer can read the Thing Description and handle the Thing
… we can leave out the "Alternatives" description within the "Delverables" section
… on the other hand, we can add clarification to the scope of the ETSI work
… describing how to use a WoT Thing Description for NGSI-LD model descriptions
Juanjo: don't know how we should express that...
Franck: (give comments on ETSI's deliverable)
McCool: Extend ETSI GR-055 or another document to include the use of WoT Thing Description for NGSI-LD model description
Juanjo: would see what is needed for the compatibility
McCool: extend existing specifications?
Juanjo: two things
Juanjo: 1. we'll extend the specifications
Juanjo: currently, some activity is ongoing
… then 2. understand how the WoT Thing Model can be used to rebase the NGSI-LD based systems
<TallTed> Outside (but maybe inside?) observation -- acronyms need expansion, at least for first use, at least at top of https://
McCool: let's see we can nail down what we expect
… maybe you can fill into the "Gaps" section with your expected gaps
Kaz: NGSI-LD and WoT Thing Description are similar to each other
… but it seems we have a bit different views and are using a bit different terminology to describe our expectations
… so would suggest we clarify our expectations based on some concrete use case including (1) Things and Consumers from the WoT side and (2) Context Brokers and Devices
McCool: that's possible but the "Simple Liaison" document should describe our high-level expectations for the liaison
Kaz: if we really want to focus on the high-level expectations, we should not dive into the details of the mechanism but should simply clarify expected document on the W3C side and the ETSI side
… I think we can merge this PR 1209 itself with keeping the "DRAFT" title and then continue that high-level discussion based on the resulted MD
McCool: that's possible but this PR 209 moves the liaison target document from the "proposals" area to the "liaison" area
Kaz: given the discussion today, we should rather keep the draft document under the "proposals" area with the title of "DRAFT" then
McCool: we're out of time, so let's continue the discussion next time
[adjourned]