W3C

– DRAFT –
WoT Architecture

23 June 2022

Attendees

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

Meeting minutes

minutes

May-24

any objections to publishing?

Lagally: hearing none, will publish

issues to close

Lagally: two; duplicate assertions, and hubs

Issue 735

<kaz> Issue 735 - Duplicate Assertions on Protocol Bindings

Lagally: issue 735, duplicate requirements
… two assertions say the same thing
… decided to leave it alone, one is a superset of the other
… will close

Issue 724

<kaz> Issue 724 - Define and discuss hubs

Lagally: issue 724
… define and discuss hubs

McCool: not super happy that I did not get around to this, but no time now, not a blocker

Lagally: ok, let's close

McCool: we can deal with in the next round

publication blockers

Issue 681

<kaz> Issue 681 - Review architecture document consistency

Lagally: issue 681, document consistency
… is a checklist, four items, one dealt with
… are three other related issues
… arch-bundling seems to also be dealt with
… arch-id-correlation seems to have been dealt with - assertion removed
… finally, section structure, assertions, and issues

Lagally: ege re-opened, said many issues still open

<kaz> i|issue 646|related Issue 646 - [Index] Section structure, assertions, and issues|

Lagally: but seems that only one other open issue is linked

Lagally: suggest closing this issue and asking for other individual issue to be opened

McCool: agree, these omnibus issues are difficult

Lagally: let's look at arch-schema

McCool: btw directories are a counter-example where a detailed data schema for a response is not possible for query results

McCool: however, it is a should, so exceptions are allowed

McCool: think it is fine for architecture to define some general requirements
… but we have to watch out for contradictions

Kaz: I think it is fine to have general requirements in arch

Lagally: agree, so let's close the issue and keep the assertion

Issue 681

<kaz> Issue 681 - Review architecture document consistency

Lagally: issue 681
… seems all the sub-issues were dealt with, can close

McCool: and if there are any smaller sub-issues to deal with, let's open small issues, not the omnibus issue

Issue 627

<kaz> Issue 627 - Chapter 10 uses non RFC assertions

Lagally: issue 627
… non-normative, uses must, may, etc. in multiple places

McCool: so most of this seems to be about the runtime, scripting API, etc. but this is a non-normative deliverable

Lagally: suggest we leave this alone, not formal assertions

McCool: agree with leaving it alone
… especially since we might want to make this normative in the next version

normative changes

Issue 689

<kaz> Issue 689 - Reorder chapters in Section 7

Lagally: issue 689
… reordering sections, but needs more detail to be actionable

Lagally: will mark as nice-to-have, but will close

Issue 676

<kaz> Issue 676 - Move "orphaned" annex B.1 from "binding templates" to architecture

Lagally: issue 676

McCool: thought this was done?

Lagally: will close

Issue 673

<kaz> Issue 673 - Review Discovery Spec, ensure consistency of architecture and discovery

Lagally: issue 673

McCool: can close for now, if I notice anything in my next review I will bring it up

Issue 617

<kaz> Issue 617 - Review Chapter 6.7 in relation to Binding Templates 4.1

Lagally: issue 617
… think this is done, will close

McCool: if there are any specific issues left they can be raised

Issue 608

<kaz> Issue 608 - Relationship between "8.7 Protocol Bindings" and "9.5 WoT Binding Templates"

Lagally: issue 608

McCool: WoT Protocol Bindings and Binding Templates
… I think Binding Templates is an implementation of the general requirement stated in Protocol Bindings

McCool: adding one sentence to relate the two
… would help make this clearer

Sebastian: so one issue is that protocol binding is more specific, just the protocol, not the data formats, etc.

Sebastian: issue was about understanding the relationship

Sebastian: should be about both the protocol and the serialization

Kaz: ok with any kind of definition as long as it is clear
… both protocols and media is find
… suggest changing "protocol binding" -> "binding" and clearer definitions

Sebastian: would just add an intro sentence to separate sections

McCool: I was going to suggest changing Protocol Binding to just Binding also...

Lagally: although it is used in lot of places

McCool: I could live with just leaving it and just accepting that it includes more than just the protocol itself, but also the data

Lagally: agree, let's do that
… sebastian has volunteered to work on this

by CR transition

Issue 753

<kaz> Issue 753 - Clean up new TLS S&P Considerations

Lagally: issue 753, clean up TLS

McCool: still on me, want to mention specific version of TLS
… will try to get to it today

Issue 684

Lagally: issue 684
… missing terminology defn for Trusted Environment

<kaz> Issue 684 - Introduction misses WoT Discovery

Lagally: issue 684
… missing Discovery in list of deliverables

Issue 706

<kaz> Issue 706 - create terminology entry for "Trusted Environment"

Lagally: issue 706

McCool: will also try to do soon, review in security call Monday

Kaz: we're planning to have joint discussions with JSON-LD and DID during TPAC, so I think it would make sense to have some discussion about "Trusted Environment" there.

McCool: definitely

new PRs

Lagally: TD Server, will merge

PR 779

<kaz> PR 779 - Editorial fixes

Lagally: editorial fixes, #779

McCool: would like to look at this first

Lagally: ok, will keep open for now

<kaz> [adjourned]

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