W3C

WoT Scripting API

09 Nov 2020

Attendees

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

Contents


<scribe> scribe: zkis

Oct 19 meeting minutes review

<kaz> Oct-19

Approved.

TPAC minutes

<kaz> TPAC consolidated meeting minutes

Zoltan: we should discuss to include in the scope (or not) of the semantic API (suggested by M. Koster) and simpler API (like Dave Raggett's object API)

Daniel: the semantic API can be done easily on the top of the current API
... we can publish a separate Note about it

<cris> +1 also from my side

The Scripting API part of the TPAC minutes approved.

Issue#280 https://github.com/w3c/wot-scripting-api/issues/280

Daniel explains the issue

Connected to https://github.com/eclipse/thingweb.node-wot/issues/333

Zoltan: explains the node-wot issue discussion

Cristiano: if we tag a property as read-only, scripts should control emitting notifications than the runtime

Daniel: the ConsumedThing you get from expose() will also have problems writing read-only properties

Cristiano: the emitPropertyChange() method solves that, since it can update the clients about the updated Property

Zoltan: the ConsumedThing returned by expose() is a local object (looks the same but have different binding)

Cristiano: explains why this is good

Daniel: we don't need any change, we can do this already
... explains relevant node-wot code example

Cristiano: we might have an internal DB and then writeProperty() would not be useful
... we should separate the issue of inheritance and emitPropertyChange()

Zoltan: scripts could implement the convenience methods writeProperty() etc themselves
... we should have separate definition of a LocalThing from ConsumedThing

Cristiano: that is confusing since it looks like ConsumedThing but works differently

Daniel: we do have internal containers in node-wot and we could expose that in the spec perhaps

Zoltan: I would first remove ConsumedThing and then experiment and see how to standardize the convenience

Daniel: I would also allow LocalThing to write even readonly properties, which is why I wanted different method names
... OK, I agree we should start with removing ConsumedThing, at least that is portable

Cristiano: agreed

Daniel: we need to involve Ege and his students for feedback

Zoltan: please comment on the issues

Cristiano: agreed

[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: 2020/11/17 06:48:39 $