W3C

– DRAFT –
WoT Profile

07 May 2024

Attendees

Present
Ege_Korkan, Kaz_Ashimura, Luca_Barbato, Tomoaki_Mizushima
Regrets
-
Chair
-
Scribe
kaz

Meeting minutes

Editors

Luca: (adds himself to the Editors list on the Profile wiki)

Kaz: we need to see who would like to be part of the Editors again

Minutes

Apr-23

Luca: fine with the minutes

(approved)

Quick PRs

PR 411

<luca_barbato> PR 411 - Update references to wot-thing-description with wot-thing-description11 - closes #410

Luca: fixes the reference to 1.1 version

(merged)

PR 409

PR 409 - Update readme with the link to mastodon

(merged)

PR 408

PR 408 - Make sychronous member of ActionAffordance mandatory - closes #405

Luca: would ask Ege for opinions

Ege: mandating synchronous action
… that's good direction, I think

Kaz: let's merge this PR, and fix possible issues separately later

(merged)

PR 407

PR 407 - Add sse subprotocol to SSE example TD - closes #370

Luca: looks good

Ege: ok

PR 399

PR 399 - Add a diagram of the current profiles

Luca: this PR is an old one
… fixing the diagram of the profiles

<Ege> https://w3c.github.io/wot-architecture/#wot-profiles

Kaz: as long as the content of the diagram is correct, I'm OK with merging this

Luca: ok

(merged)

PR 381

PR 381 - Adding missing member "alternate" to the table

Luca: should not be controversial

Ege: the PR itself is OK
… but the table there is not really useful...

table defining th link relation types

<Ege> https://w3c.github.io/wot-profile/#common-constraints-links

Ege: would be better to have the definition on the TD side

Kaz: also OK with the PR itself
… but we need to think about further spec refactoring

Luca: ok

(merged)

Issues

Luca: we have many issues to be discussed

Issues to be discussed

Luca: what kind of transformation can be done with node-wot?

Ege: SSE can be handled by binding templates
… maybe outside of node-wot itself
… Webhooks are more fundamental, though

Luca: one of the main concern is duplication of parsing
… need to clarify the mechanism
… you have completely implicit operations
… if you have an "invoke" that is asynchronous, you have a default query action
… to cancel the whole remaining actions
… if you see an "invoke" with the protocol...
… for asynchronous actions
… cancel the query

Ege: whenever we see the case, runtime save the ID inserted
… when the query comes

Luca: you can get away and can say
… the best you can do is invoking the action
… but if you don't have explicit cancellation mechanism, you can't do anything

Ege: need to talk with Daniel and Cristiano too

Luca: if you handle async actions, need to know how to cancel it
… it's not explicit but implicit
… they can't do anything incorrect
… would be great to have descriptions handled by the TD 2.0 spec

Ege: in that case, publishing the WoT Profile as is would be the only way

Kaz: we need to clarify what the WoT Profile spec should handle
… also we can think about possible Plugfest to see what is really needed for industry applications based the WoT Profile spec
… can think about the possibility of June virtual F2F, TPAC in Sep and/or physical F2F in Nov
… so far all the meetings are not fixed yet, though

Luca: sounds good

Kaz: btw, I think it would be easier to handle the discussion about which way to go by dedicated meetings
… we can have an additional meeting
… or can have discussion during the possible June virtual F2F

Luca: ok
… anything else for today?

(none)

Luca: Ege, please talk with Cristiano, Daniel, Sebastian, etc.
… you can ping me if you want

[adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 221 (Fri Jul 21 14:01:30 2023 UTC).