Re: [voiceinteraction] VoiceXML platform builtins

Dear all,

unfortunately, I will have to skip the meeting today. However, I made 
some changes to the document and you may want to have a look at that

  * section 3 new graphic for minimal architecture
      o new component to track context requires an update in other
        architecture drawings
      o plan to use this as a basis to introduce detailed subcomponent
        drawing per component section
      o end with a general overview before the walkthrough (now right
        below the simplified drawing)
  * section 3.2.2.1 table for dialog strategies (ws a list)
  * section 3.2.5 new table for core dialogs
  * section 6.1 (new) added a section for abbreviations

Thank you

Dirk

Am 06.10.2020 um 23:55 schrieb Deborah Dahl:
>
> During our last call we talked about generic capabilities that could 
> be included in any Intelligent Personal Assistant platform, and it 
> seemed like we could get some ideas from VoiceXML 
> (https://www.w3.org/TR/voicexml20/) 
> <https://www.w3.org/TR/voicexml20/)> built in events.
>
> I took an action to look them up.
>
> They are:
>
> Help
>
> Nomatch
>
> Noinput
>
> Error
>
> These could be used in any VoiceXML document and a VoiceXML-compliant 
> platform would be required to provide some kind of handler, that is, 
> not crash or hang, if they occurred. It was considered to be poor 
> design on the developer’s part to let the platform default handler 
> actually handle these events (by not providing overrides), but they 
> had to be there.
>
> In our architecture, the analogy could be that these capabilities are 
> always part of the IPA Service.
>