W3C

WoT Discovery

20 September 2021

Attendees

Present
Andrea_Cimmino, Christian_Glomb, Farshid_Tavakolizadeh, Jan_Romann, Kaz_Ashimura, Kunihiko_Toumura, Michael_McCool, Tomoaki_Mizushima
Regrets
-
Chair
McCool
Scribe
jkrhb

Meeting minutes

minutes

<kaz> Sep-13

McCool: created a PR for issue #193
… we can discuss it today

McCool: There is a typo under issue 185
… no other problems, however
… any comments?

There are no objections

McCool goes over today's agenda

PRs

PR 220

<kaz> PR 220 - Fix assertion in section 5

McCool: Let's deal with PR #220 first
… this resolves issue #200
… There is another PR which might cause conflicts, but let's merge this PR first
… adds consistency by using assertions in certain sections
… IDs and classes look good, we can go ahead and merge after looking at the preview
… exploration and introduction should be clarified in the "Introduction Mechanisms" section

McCool: Not an issue with this PR, however

McCool: Are there any objections to merging this PR?

No objections, McCool proceeds with merging

PR 220 has been merged

PR 221

<kaz> PR 221 - Update intro to introduction mechanisms

McCool explains the issues with the current "Introduction Mechanisms" intro

McCool: text said result of introduction can be a URL or a TD but it is always a URL at this stage

McCool: There is currently a conflict that has to be fixed before merging

McCool resolves the merge conflict

McCool brings up the rendered preview for a walkthrough

McCool: "Discovery" is replaced with "initial contact"

<kaz> Preview - 5. Introduction Mechanisms

McCool: intro clarifies that multiple introduction mechanisms are possible
… the two phases are now seperated in a better way
… merging of results is now being mentioned

McCool: There is a similar issue in the following (sub)sections that requires clarification
… mostly in "Direct" and "Well-known URIs"

McCool: Any comments on this PR?

McCool: I suggest merging this PR and then do more clean-up later

There are no objections to merging the PR

McCool adds a comment to the PR about future clarifications and proceeds with merging

McCool: With the merging of #221 we can consider issue #200 resolved
… issue #193 has been resolved as well

McCool closes issues #200 and #193

<kaz> Issue 200 - Implementing a DNS-based approach, are DNS-SD and mDNS mandatory?

<kaz> Issue 193 - Assertions removed by mistake for Introductions?

PR 222

<kaz> PR 222 - Discoverer

McCool: I opened this PR with regard to brownfield devices that don't support discovery natively
… the PR introduces the concept of a separate "Discovery"

McCool: Any objections to merging this PR?
… not sure if it is the right place but we can always refactor it later
… I am not hearing any objections

McCool proceeds with merging PR #222

<McCool> https://github.com/w3c/wot-discovery/issues/216

McCool: With the merge of PR #222 we can close issue #216 now

PR 223

<kaz> PR 223 - Add object types to actions

Farshid: This PR solves the issue Ege was raising about the input

McCool: We also need a note explaining the type, looks like an empty object to me

Farshid: We have the content-type describing that it is a TD

McCool: The output depends on your schema
… can also be a string

McCool: Maybe we should first get some feedback

McCool adds comments to the PR

McCool: Maybe the type should be oneOf object, number, string, ...

Farshid: The changes actually aligns it with the specification
… we are waiting for the specification of JSONPath, however
… in JSON path some clarification is still required

McCool: I assume schema is optional in TD spec
… type: "object" probably does not describe an empty object as additionalProperties is always true
… therefore I assume this is probably okay

Farshid: Everything is optional in DataSchema

McCool: Probably not a problem then
… this should probably be explained, though
… in the text

Farshid: In the text it should be clear

McCool opens the preview for reviewing the changes

Farshid: In "Creation", for example, there is a paragraph regrading the validation of TD objects

McCool adds a comment to the PR saying that there should be an additional sentence regarding the output schema of the action

Farshid: It could also be part of a description

McCool amends his comment on the PR

ft will clean up the PR accordingly

McCool adds a label to issue #182 saying that a PR is available

<kaz> related issue 182 - TDD TD: Don't Actions need an input field

DID

<kaz> wot issue 982 - Joint call with DID

McCool: I created a joint call for DID
… there is a dedicated label for DID, some issues could probably be closed
… some of them are older than a year
… an extra PR might be needed for clean-up and clarifications, for example with regard to issue #201

https://github.com/w3c/wot-discovery/issues/201

McCool: DID could be useful for distributing keys
… is also related to signing TDs, but there has to be a signature mechanism defined before
… DID could also be used to resolve encryption for secure local transport

McCool: We should talk to the DID guys and discuss the possibilities for enabling local HTTPS
… there are some issues that have to be resolved before actually being able to recommend using DID

McCool: In general, there are a lot of issues that are probably able to be close, I will go through them and add "Propose closing" labels

Cancellations

McCool: Plugfest and F2F are coming up
… normally we cancel meetings but there is a lot to do at the moment
… I propose going to 2 hour meetings
… the meetings next week and the week after are cancelled due to Plugfest and F2F
… October 11 will be a 2 hour meeting then
… October 4 will be a 1 hour meeting, however, to prepare for F2F

… October 18 will also be a 2 hour meeting

McCool: I will update the cancellation section accordingly
… October 25 will be cancelled as well

<kaz> [adjourned]

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