W3C

– DRAFT –
WoT Profile

29 June 2022

Attendees

Present
Kaz_Ashimura, Michael_Lagally, Michael_McCool, Tomoaki_Mizushima
Regrets
-
Chair
Lagally
Scribe
McCool

Meeting minutes

Agenda

Lagally: review of 06/22 minutes

Lagally: note that I cleaned up and merged some things we decided on last week

Minutes

<kaz> June-22

Lagally: reviewed a lot of issues at a high level, will drill down on some of these a bit more today

McCool: also, for the agenda, suggest we don't spend too much time on length limits, other things are higher priority

Lagally: agree

Lagally: publish if no objections?
… hearing none, will publish

PR 235

<mlagally> https://github.com/w3c/wot-profile/pull/235

<kaz> original PR 198 (closed due to merge conflicts) - WIP: event payload format

Lagally: event payload format, has a lot of merge conflict, have created new PR 235 PR 235 - CloudEvents as message payload format for WebHooks
… note there is however a lot of useful discussion under the old PR
… will close, but will link back to it from the new PR

Lagally: issue with multiple devices talking to the same cloud endpoint, single listener, multiple "speakers"
… message has to convey device; also timestamp, affordance (event source), data payload, etc.
… naive approach would be to define a JSON schema for the message format
… but if you look at what cloud companies do, there is already a format for this purpose

McCool: is this the *only* format being used?

Lagally: not aware of any other public standard used broadly

McCool: note this constrains not onl the Things but the cloud endpoints

Lagally: correct, this is for webhooks also

McCool: a little worried about the formality of the cloud events spec, also versioning

Lagally: is versioned

McCool: ok, but I think we should also have enough information in our spec to be clear even if the cloud events spec goes away

McCool: ok with merging if others are

McCool: what is the minimum implementation?

Lagally: less than ten attributes in a message

McCool: actually, count only 5 attributes that are mandatory

Lagally: if no other questions, will merge

Issue 233

<kaz> Issue 233 - Define a constrained set of link relation types

Lagally: constrained set of link relation types

McCool: also think we may want to limit the media types

Lagally: and also make the type mandatory

McCool: also, tm:* relations are allowed? Are TMs covered by profiles?

McCool: also, TMs are definitely going into the TD spec now

Lagally: not clear about manifest
… which refers to a UI

McCool: ben may have some input on, WebThings tends to return HTML for affordances unless requested otherwise

McCool: if we DO support manifest, I think we should be specific about the type is points to

Lagally: but we can leave it out for now

McCool: agree

<kaz> TD draft - 5.3.4.1 Link

Lagally: also "alternate" has a number of examples of different representations

McCool: my preference would be to insist on JSON-LD serializations of TDs in profiles

Lagally: so "alternate" does not make sense
… let me make a note about that

Lagally: anyway, these six are the suggested types
… I will do a PR if there are no other volunteers

issue 232

<kaz> Issue 232 - Define minimum set of supported languages

Lagally: minimum set of supported languages

Lagally: could look at locales supported by Java

McCool: might also want to look at browser support, take the intersection
… suggest we ask i18n team, and can also label, tag @aphillips, etc.

cleanup

PR 227

<kaz> PR 227 - Update Implementation Report

Lagally: merged impl report, that is done

PR 207

<kaz> PR 207 - updating JSON Schema Validation reference

Lagally: PR 207 updates the JSON Schema ref

McCool: just make sure this is not changing from the version used in the TD spec
… I'm not even sure we need this in profiles if TD specifies it

<kaz> [adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 185 (Thu Dec 2 18:51:55 2021 UTC).