W3C

WoT-WG - TD-TF

03 February 2021

Attendees

Present
Christine_Perey, cperey, Cristiano_Aguzzi, Daniel_Peintner, Kaz_Ashimura, Michael_Koster, Michael_McCool, Sebastian_Kaebisch, Tomoaki_Mizushima
Regrets
-
Chair
Sebastian
Scribe
kaz, sebastian

Meeting minutes

review minutes

Jan-27

<kaz> (Sebastian goes through the minutes)

any objections to bring them to public?

no

WoT binding

there is a PR available about the Modbus work to provide a prototype document

wot-binding-templates PR 109 - Refining Modbus protocol binding

so far, the PR have an TD example provided

there needs some still some work on different topics as mentioned in the PR

<Ege> MQTT in RDF

there was some discussion about the structure of the new document

it was recommended to follow the approach as shown in chatper 5.3.1 of the binding template docuement in combination of a set of examples similar as shown in chapter 4.3.

<Ege> uRDF Store

<kaz> WoT Binding Templates Editor's Draft - 4.3 Interaction Affordances

Ege: shall we work on the binding template document?
… let's wait when we have the Modbus binding prototype ready and get agreement on this. Proposal to copy all protocol specifcs to seperate document

Ege: do we need section 4.3?

MK can support on the Modbus work

<cris> https://github.com/eclipse/thingweb.node-wot/tree/3e9ccd44072a097d0452b5c306819c6d8fb04e12/packages/binding-modbus#modbusrange

<McCool_> (sorry I'm late, had a conflict)

<Ege> https://github.com/w3c/wot-binding-templates/issues/110

Chapter 4.3 will be updated by Ege

TD - PR 1042

<kaz> TD PR 1042 - Add additionalResponses to Form

Sebastian: shows the preview of the table at "5.3.4.2 Form" including "response"

<kaz> TD preview - 5.3.4.2 Form

<Ege> Hue developer page (need login)/

<Ege> a successful response is like:

<Ege> [ {"success":{"/lights/1/state/bri":200}}, {"success":{"/lights/1/state/on":true}}, {"success":{"/lights/1/state/hue":50000}} ]

<Ege> example failure message in hue api

<Ege> [ { "error": { "type": 7, "address": "/lights/1/state/on", "description": "invalid value, 123,, for parameter, on" } },

<mjk> https://tools.ietf.org/html/rfc6570

<mjk> URI templates

TD - PR 1034

<kaz> PR 1034 - Init information about WoT relation types

<kaz> IANA Link Relations

<kaz> RFC 6573 - The Item and Collection Link Relations

(PR1034 has been merged)

TD - PR 1024

<kaz> PR 1024 - WIP: Topics around Thing Model

Sebastian: shows Cristiano's 2nd example at https://github.com/w3c/wot-thing-description/pull/1024#issuecomment-768966909

TD Issue 972

TD Issue 972 - JSON Schema for Thing Model

Sebastian: shows his latest comments at https://github.com/w3c/wot-thing-description/issues/972#issuecomment-772565826

Kaz: Thing Model is a template to generate actual TDs
… so we don't have to make all the possible Thing Model valid JSON, do we?
… maybe we could simply think about some preprocessing (to convert {{WATER_MINIMUM}} into a valid TD parameter) before applying some schema validation

Sebastian: if we wrap {{booleanKey}}, etc., with "" (double quotations), we could apply usual JSON Schema validation

Kaz: yeah, that's fine but we still need quick check of the template notation, i.e., {{booleanKey}}, is a valid expression as a Thing Model fragment
… so my question is, does a Thing Model really need to be a valid JSON?

McCool: think it should be

Kaz: ok, if the answer is yes, we should go for it :)

Sebastian: (adds some more comments to Issue 972)

Sebastian's new comments for Issue 972

Ege: btw, do we want to maintain two separate schemas?

Kaz: to be strict, Thing Model has a bit different syntax from usual Thing Description instance, so we still need to have two (a bit) different schemas, one for Thing Model and another for Thing Description

Sebastian: right

[adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 127 (Wed Dec 30 17:39:58 2020 UTC).