W3C

- DRAFT -

WoT Discovery

06 Apr 2020

Agenda

Attendees

Present
Kaz_Ashimura, Kevin_Olotu, Michael_McCool, Kunihiko_Toumura, Christian_Glomb, Andrea_Cimmino, Zoltan_Kis, Tomoaki_Mizushima
Regrets
Chair
McCool
Scribe
kaz

Contents


Guests

McCool: only one guest for today, Kevin Olotu from Bosch
... any objections?

(none)

McCool: so would accept his participation

<McCool> Bosch, Kevin Olotu

Previous minutes

<McCool> https://www.w3.org/2020/03/30-wot-discovery-minutes.html

McCool: (goes through the minutes)
... pull request 4
... use cases
... issue 7
... and new issue about Vorto, etc.
... decision of acceptance after Christian is back

Meeting schedule

McCool: cancel the nect week meeting due to Easter

2 Phase model

McCool: maybe we should have some discussion, Toumura-san?

Toumura: ok

Prev minutes (revisited)

McCool: any objections to accept them?

(none)

McCool: accepted then

PRs and Issues

PR 12

McCool: assigned myself to this
... Intel guys are interested
... but it may extend the Charter though it's an interested topic
... anybody want to make a case?
... within the WoT discovery scope?

changes

McCool: (goes through the changes)
... objections to merge this?

Zoltan: usually different protocols for different discovery mechanisms

McCool: don't want to search over all the possibiities

Zoltan: it's not a discovery use case to be strict
... not for the scope of discovery

McCool: may raise big implementation limitation

Zoltan: agree

McCool: any other comments?... can merge this PR itself, and can have further discussion based on that
... (merges PR 12)

PR 15

McCool: Kevin made a PR but Bosch is not a Member yet
... so we can't merge this PR...
... let's look into it to see if it's normative or informative
... this is public informative and just informative

Kaz: this is not really a proposal for the spec but informative example from exisiting industry standards. right?

McCool: right

Kaz: so this is something like IETF ACE for wot-security

McCool: right

Kaz: in that case we can remove the title of "proposal" and add "this section is informative" instead
... but please let me check with Philippe and Wendy to make sure

McCool: ok
... also would like to suggest we add some note
... (adds a comment to the PR)

McCool's comment

McCool: tx for writing this up, Kevin!
... leave this PR as is until Kaz's getting clarification

Issues

McCool: need to go through the issues

Issue 5

McCool: have we discussed this?

Toumura: not yet
... this is Suzuki-san's trial implementation for 2-phase model
... he wanted to try this for the PlugFest

McCool: ok
... seems he used mDNS and DNS-SD
... also directories service by Mongo DB

Toumura: note that my first sequence diagram at https://github.com/w3c/wot-discovery/issues/5#issuecomment-598700491 is not correct
... please see the second one at https://github.com/w3c/wot-discovery/issues/5#issuecomment-599392658

McCool: we should include discussion of alternative DN-based approache
... for exaple the approach decribed in issue 5
... (adds that comment to Issue 11)

McCool's comment on Issue 11

Toumura: we can assume that authentication could be done on the network level?

McCool: might be too dangerous
... may need device authentication even with VPN setting
... let's discuss that later
... anyway Suzuki-san genrated Issue 5, and this point should be captured within the official document

Toumura: agree

McCool: (adds a comment to issue 5)
... can this (issue 5) be captured and put into an MD file and submitted to the repo?
... that way we are use what the current, "correct" description is

Toumura: can ask him to generate a MD

Issue 14

McCool: didn't get any comments so far
... can capture your comments now :)
... pros: integrate with systems and ecosystems that already support other metadata formats, e.g., OpenAPI
... cons: implementation complexity, interoperability downsde
... dont know in advance whta formats are supported
... any other comments?

Zoltan: interoperability with directory service?
... what would be the use cae for that?

McCool: we need to focus on supporting W3C standards
... if a service already exists to support other formats that can remain and be used as needed, whether or nt they share backends is an impleentation issue and need not be visible at the network API level
... any other comments?
... I myself don't have any opinion
... would like to ask the group for opinions

Zoltan: let's do some simple use cases first

McCool: ok
... (generates proposed resolution)

<McCool> proposal: (regarding issue 14) - a WoT discovery service should support only WoT TDs. Content negotiation, if supported, should only be used to select between different versions of TDs (when there is more than one...)

McCool: any objections?

(none)

RESOLUTION: (regarding issue 14) - a WoT discovery service should support only WoT TDs. Content negotiation, if supported, should only be used to select between different versions of TDs (when there is more than one...)

McCool: (and adds the above resolution to Issue 14)
... that's it
... aob?

(none)

[adjourned]

Summary of Action Items

Summary of Resolutions

  1. (regarding issue 14) - a WoT discovery service should support only WoT TDs. Content negotiation, if supported, should only be used to select between different versions of TDs (when there is more than one...)
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.154 (CVS log)
$Date: 2020/04/07 10:06:48 $