W3C

– DRAFT –
WoT Profile

28 May 2025

Attendees

Present
Ben_Francis, Cristiano_Aguzzi, Kaz_Ashimura, Michael_Koster, Sebastian_Kaebisch, Tomoaki_Mizushima
Regrets
Daniel, Ege
Chair
Ben
Scribe
mjk

Meeting minutes

Minutes

<kaz> May-14

<kaz> (minutes approved)

PR review

PR # 419

<kaz> PR 419 - Explicity state contentType of application/json in Forms - closes #141

Ben: How does a TD consumer select from multiple forms?

<kaz> related Issue 141 - how to handle multiple forms

Sebastian: We don't need to explicitly specify applicatoin/json in the form because it is the default
… TD 2.0 will also have global definitions that might help with this issue

<kaz> Diff: 6.2.1.1 readproperty

<sebastian> https://w3c.github.io/wot-thing-description/#table-default-values-of-vocabulary-terms-that-are-used-when-the-terms-are-not-present-in-a-td

Ben: The TD spec is not clear enough about the behavior of default content types

Sebastian: Here is a link to the TD spec section

Ben: (reviews the algorithm for content type selection in the profile document )

Sebastian: (explains the content selection algorithm commonly used by WoT clients)

Sebastian: A client does needs to know which content type it needs and need to look in the form

Ben: there are some cases that won't work

Kaz: we should make an issue for TD and go ahead with Profile 1.0

Ben: I have made an issue for TD 2.0
… will leave the issue #141 open for discussion

Sebastian: to clarify, the redundancy doesn't hurt, and we can improve the definition in TD 2.0

Cristiano: It seems appropriate for Profile clients to look for application/json content type

<kaz> Ben's comments for Issue 141

WIP PRs

Ben: can the WIP PRs be closed or merged??

PR #334

<kaz> PR 334 - WIP: Provide example in the new introduction section

<kaz> Ben's comment on PR 334

PR #271

<kaz> PR 271 - WIP: Common constraints - identifiers

Ben: I don't think using urn:UUID is necessary

Sebastian: agree
… it it's important to use UUID URNs you can do that

Ben: Does anyone disagree that this isn't necessary?

Sebastian: We should leave this for Lagally review

PR #314

<kaz> PR 314 - WIP: Allow auto security scheme for other permitted security schemes

Ben: For this one we should wait for McCool to return

PR #330

<kaz> PR 330 - Cloud Events Message Format

Ben: there is disagreement about cloud events
… it was for Profile 1.1, should we keep it open

Sebastian: Cloud Events has been used in the market but I'm unsure about the formal specification
… We should discuss more

Cristiano: if there is no activity, we can close the PR and re-open if we need more discussion

Ben: OK to close the PR with comments?
… OK, will close

PR #417

<kaz> PR 417 - A start on use cases and requirements for Profiles 2.0

Ben: This was to drive consensus on the requirements for Profiles 2.0
… We could leave it open to continue the discussion

Ben: any opinions?

Kaz: It was resolves that a task force can work on the use cases and requirements

Ben: It's important to have a central place for use cases but also a separate place for each specification

Issues review

Issue #255

<kaz> Issue 255 - Remove or re-work section 6.1.2 Links

Ben: The idea is to constrain the use of link relations for interoperability
… It's difficult to make testable assertions
… Is fixing this important for TD 1.0?

Cristiano: Is the concern stability or interoperability?

Ben: The concern is tetestability

Cristiano: agree that is is difficult to test

Ben: There are a couple of ways we can improve this
… We can remove it or we can fix the text
… What do we need to do for the Profile 1.0 note, considering testability is not as important for a note?

Cristiano: The content is useful as informative to have in the document, and will improve the note

Kaz: I agree, and suggest we add a clear explanation in the text as an editors note

<kaz> Ben's updated comment

Issue #258

<kaz> Issue 258 - New section on Webhook message format

Ben: This proposal for webhooks is aligned with SSE
… the metadata is in HTTP headers instead of the body

Ben: It avoids the need for a content wrapper
… does this need to be fixed for 1.0?
… It is needed for interoperability

Cristiano: Should we remove this section?

Ben: We might define our own payload format or use the header approach

Cristiano: The webhook is easier to use with Profile 1.0
… because the payload is described in the data schema

Kaz: This discussion should include the binding group
… Webhooks don't need to be resolved for Profile 1.0

Ben: Agree thet 2.0 is the place to resolve this, the issue is that there is a specific requirement in Profile 1.0

Kaz: minimum effort is the guidance

AOB and Adjourn

Ben: adjourned

Minutes manually created (not a transcript), formatted by scribe.perl version 244 (Thu Feb 27 01:23:09 2025 UTC).