W3C

– DRAFT –
WoT-WG - TD-TF

16 June 2021

Attendees

Present
Cristiano_Aguzzi, Ege_Korkan, Kaz_Ashimura, Michael_McCool, Sebastian_Kaebisch, Tomoaki_Mizushima
Regrets
-
Chair
Sebastian
Scribe
kaz, sebastian

Meeting minutes

check minutes from last time

<kaz> June-2

any objections?

no

<kaz> approved

publications plans

New TD WD was published last week

to be aligned with the charter we have to submit the CR on July 15

one week before we need the resolution from the group (July 7)

<kaz> +

<kaz> updated publication schedule

to be realistic the dates are too optimistic. We have some open white gaps from testing. In addition Discovery and Profil also need more time.

we should think about an extension

Kaz: the maximum extension is 6 month, so deadline to complete the charter is end of June, 2022

we have also consider the summer vacation time

<kaz> a/2022/2022. however, we should remember we need to discuss the rechartering procedure and also possible comments from the wide reviews, etc./

<kaz> kaz: so we should look at the current draft again, and seriously think about what is really missing.

so far, we would be ok to have a 6 month extension. However, let's clarify this on Monday in the vF2F meeting

vF2F TD topics

I provided a proposal which kind of topics we can discuss for the TD session

I think the TestFest topic we can remove since this is coverd by the generic TestFest session of the first day

another proposal would be about definitions about properties/actions/events

and TD RDF round-tripping (JSON-LD 1.1 releated)

Kaz: points out there is a possibility that it's not necessarily we need to modify the TD spec itself but might be that we should clarify how to write TDs and map it with the other IoT standards using the Binding Templates. For that purpose, we need concrete best practices.

McCool: we do need best practices on modelling to apply TDs to the other standards
… our examples within should align with those best practices if possible

Sebastian: we have to clarify the usage of property/action/event
… maybe we should gather best practices about that
… for that purpose, we need a leader for that direction
… maybe Matsukura-san himself could do that
… and Ege could help
… would be great if Ege could work with Matsukura-san to generate some concrete proposal
… how to use actions and proposals, etc.

Ege: ok

Kaz: thanks a lot!

Sebastian: what about signature?

McCool: major security topic

Sebastian: ok
… next, clearer description for Thing Model

McCool: what is go into TD
… and what the limitations are

<Ege> https://github.com/w3c/wot-binding-templates/issues/113

Cristiano: what can be done and what can't

McCool: validation for TM and TD as well
… how restrictive should the validator?

Sebastian: note we don't support multiple inheritance at the moment

McCool: ok
… what would be alternative for multiple inheritance then?
… need to discuss combination of multiple models anyway

Sebastian: ok
… anything else for the TD session?

McCool: summary of the change logs

Sebastian: thought had generated a list already

Kaz: you can add the latest changes after the WD publication
… to the change logs within the WD

Sebastian: ok

Ege: could we talk about assertion clarification too?

McCool: validation extension is still a bit vague
… we should think harder about extensions
… maybe using the same context for different content types
… anyway assertions should be self-contained and validatable

Ege: wondering about assertions for TM as well

Sebastian: regarding security for TD, is everything covered?

McCool: 30 mins for security; 20 mins for signature and 10 mins for other topics like body, uri, ...

Sebastian: what about coffee break?

McCool: we do need a coffee break

Sebastian: ok
… next PRs?

PRs

PR 1151

PR 1151 - WIP: TD Signatures

McCool: changed this a bit

McCool's comments 14 days ago

5.3.3.2 Signature withing PR 1151

McCool: (goes through the updated text)
… (specifically review the table defines the table at section 5.3.3.2)
… (the detail of "alg" parameter)
… (and the "jku" parameter)
… (and then describes the signature verification process, 1-8)

Sebastian: any idea about the dependency on platforms?

McCool: there are different kind of signatures
… visits 5.3.3.1 SignedInfo

5.3.31 SignedInfo

McCool: referenceType values, i.e., jsonpointer, jsonpath and xpath are enumeration

Sebastian: do we have examples?

McCool: not yet

PR 1157

PR 1157 - Fix ContentEncoding and ContentMediaType

Cristiano: need to fix the schema and regenerate the HTML

McCool: can merge this PR itself, and add further update?

Sebastian: (adds comments)
… small fixes on the list of "7bit", "8bit", "binary, "quoted-printable" and "base64" to be examples
… and provided by the string type

Sebastian's comment

PR 1158

PR 1158 - Proposal: use version field in schemas

Sebastian: (merges)

PR 1129

PR 1129 - More fixes to canonicalization

McCool: let's discuss this next week

Sebastian: (approves the latest updates)
… (adds comments)

Sebastian's comments

Sebastian: (and then merges the PR)

PR 1163

PR 1163 - add success to addResp and refactor addResp into definitions

Sebastian: (merges)

PR 1167

PR 1167 - fix: fix RDF triples example from Thing description json-ld 1.1

Sebastian: need more discussion during vF2F
… PR 1168 and PR 1169 as well

Issues

Issue 1173

Issue 1173 - Unclear assertion on TMs

Sebastian: (would add clarification)

Issue 1172

Issue 1172 - Unclear assertion on security

Sebastian: (goes through the implementation report for the TD REC)

implementation results

Ege: also Issue 1171

Issue 1171 - Fixing Thing Model assertion ids

Sebastian: (adds a label of "PR needed")

Issue 1165

Issue 1165 - Minor spelling and grammar errors

Sebastian: to be updated

Issue 1174

Issue 1174 - Rename / retitle op for Things

Kaz: think TD itself is a data model and adding an alias name should be handled by some external mechanism of TD, e.g., Scripting API or the processor side

McCool: right

Sebastian: would like to look into this issue a bit more

AOB

<Ege> https://github.com/w3c/wot-thing-description/issues/1154

Ege: got a response from Henry about the issue above
… still need some more discussion, though

[adjourned]

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