W3C

- DRAFT -

WoT Scripting

08 Jun 2020

Attendees

Present
Kaz_Ashimura, Cristiano_Aguzzi, Daniel_Peintner, Tomoaki_Mizushima, Zoltan_Kis
Regrets
Chair
Zoltan
Scribe
zkis

Contents


<scribe> scribe:zkis

Prev minutes

<kaz> June-1

Past minutes approved

issue#216

<dape> --> https://github.com/w3c/wot-scripting-api/issues/216

Daniel: fetching in node-wot was quite cumbersome
... it was difficult to deploy applications
... discovery isolates the fetching, we get the TD and it's fine
... so this looks better, but we need to experiment with it

Zoltan: it seems we have consensus on the discovery approach

Cristiano: we'll also see how it goes in the Discovery calls

PR 218

<dape> https://github.com/w3c/wot-scripting-api/pull/218

Zoltan: Added InteractionData instead of any in the handler callbacks
... added the missing subscribe and unsubscribe handlers
... to ExposedThing
... added the missing algorithms in ExposedThing and adapted the existing ones for using InteractionData

Daniel: we need to update node-wot and get experience in plugfest

Cristiano: so this updates ExposedThing - will check it soon

Daniel: if we read a Property, we can pass along a formIndex
... for single reads
... but how to deal with InteractionOptions and formIndex for readMultiple or readAll

Zoltan: we have a Form with an op=readmultipleproperties
... so I don't see a problem there

Daniel: for readMultiple we don't need a Form

Zoltan: if we cannot do readmultiple with single network request, there is no point for it

Cristiano: agreed, this is tricky to implement and we should raise an issue with TD spec

Daniel: we could use multiple Forms for it

Zoltan: but then we need a way to say to use ALL the Forms, since now we only use only one Form

Daniel: let's indeed start with an issue in the TD spec, will create the issue

<kaz> 8.6 Handling requests for reading multiple Properties

Daniel: the current algorithm for ExposedThing side makes sense to me but not on the ConsumedThing, which required a single Form
... it would be the same for looping readProperty in the app
... no, since it's done inside the implementation

Cristiano: if there is no single network request, there is no point for readMultiple

Daniel: looping would make sense since it's well balanced between client and server

Zoltan: but that should be specified that way in the TD spec

Daniel: OK, we'll discuss it in the TD issue

Cristiano: MODBUS supports reading multiple properties

Zoltan: it still needs a Form to spec how to make the request
... let's check the same for writeMultiple
... we could also cover it by an Action to read/write multiple properties

Cristiano: yes, in the past we had that, but this is a convenience

Zoltan: we are missing the input from the Action in this case

Cristiano: right

Zoltan: Daniel, could you please also include an agenda point for the F2F to discuss this?

Daniel: OK

[adjourned]

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version (CVS log)
$Date: 2020/06/09 07:25:01 $