W3C

WoT Scripting API

03 May 2021

Attendees

Present
Cristiano_Aguzzi, Daniel_Peintner, Kaz_Ashimura, Tomoaki_Mizushima, Zoltan_Kis
Regrets
-
Chair
Daniel
Scribe
kaz

Meeting minutes

Minutes

<dape> minutes 19 april, see

Daniel: the call after the vF2F
… (goes through the minutes)
… we'll postpone the publication
… topics related to discovery
… issues related to TD, discovery and security
… requirements on the (possible) management api
… any objections?

(none; approved)

<dape> minutes 26 April 2021,

Daniel: joint discussion with the Security TF
… we need some kind of management api to handle security
… any objections?

(none; approved)

Joint calls

Security

Issue 315 - Security TaskForce related issues

Cristiano: well summarized

Daniel: any other comments? fine?

Zoltan: need to create some deliverable on runtime
… different from the script api itself

Kaz: implementation guideline, you mean?

Zoltan: a bit different

Daniel: we could have some kind of management api. right?

Zoltan: might need a different mode
… what is defined by the current Scripting API is definition for client APIs

Kaz: in that case, a separate document on the possible management API?

Zoltan: a document about runtime of Scripting API

Kaz: not really sure about the difference between an "implementation guideline" and a "runtime document"
… but if we're ok with generating an informative WG Note about that, that's possible
… if we want to generate a separate group Note about some topics, we should summarize what to be described by that document and bring the idea to the main call to get a group resolution

Zoltan: ok

Daniel: ok
… but still not really sure about what to be done there
… would like to get consensus within this Scripting API first

Cristiano: don't have any strong opinion at the moment.
… different information could be included under one specific repository

Daniel: how to proceed?

Kaz: we should clarify what we want to document first :)

<zkis> https://github.com/w3c/wot-scripting-api/issues/299#issuecomment-826823233

Zoltan: we can skim the issues listed here (=within the issue 315)

(issue 299 and 298 are included in issue 315)

Issue 299 - Chose a particular security schema for an ExposedThing

Daniel: we need to be careful about the API names

Zoltan: we can control the device remotely

Daniel: there is an MD file about our basic ideas

Zoltan: we should split the manager API topic from the Scripting API itself
… so we should move the content of "script-manger" area to another repository

script-manager area under wot-scripting-api repo

Zoltan: we need a separate namespace from the consumer discovery
… it requires a different level of permission

Daniel: the developer should have method to control it
… even though that is not the case with the node-wot implementation at the moment

Kaz: sorry but I still don't really understand your expectations...
… so would suggest you both clarify your expected features, phases, APIs, etc., first
… and then we can think about how to deal with those points by what kind of documents

Cristiano: exposedThing is almost a system-level API
… for system-level implementations, what is needed to be implemented?

Cristiano's comment on Issue 299

Daniel: this happens for exposedThing

Zoltan: we can take this forward but what security portion to be implemented?
… exposedThing has to be created
… and what to be provided by the runtime?

Cristiano: currently we don't do any validations

Zoltan: cat it include any kind of security specifications?
… wondering about what would be a better API definition

Daniel: we have a "wot" namespace already
… but what would be included in the runtime?

Zoltan: wot.runtime should have a different permission from wot.consume
… 3 different levels for consume, discover and expose

Cristiano: maybe we need those three levels

Daniel: wonder how to differentiate them

Zoltan: we can create a new runtime object for WoT
… but with a different permission level
… we have consume, expose and discover
… and what else?
… we can continue the discussion

Kaz: those three are rather different phases. right?

Zoltan: we use "phase" for provisioning, etc.

Kaz: in that case, maybe three different modes

Zoltan: need to continue the discussion

Discovery

Daniel: will we have a joint discussion with the Discovery TF today?

<cris> wot-thing-description Issue 1097 - Provide Typescript types definitions for Thing Descriptions

Zoltan: can be checked with McCool during the Security call

TD

Daniel: some description on the wot-scripting-api repo

Daniel's writeup

[adjourned]

Minutes manually created (not a transcript), formatted by scribe.perl version 131 (Sat Apr 24 15:23:43 2021 UTC).