Windows Phone devices already worked like that before Cortana. But in a more pre-scripted fashion. Something like
-- "You've got a new message from Paul, read it or ignore?"
-- Read it.
And so on (including the dictation and sending of the reply)
That's not how I think of 'worked like that'. I have no interest in learning pre-scripted voice commands. Once you have everything scripted, implementing an interaction as I posted is almost as trivial as a CLI. That's why I put 'natural language' as the first criteria, which renders the second task a lot more challenging.
Comments
Windows Phone devices already worked like that before Cortana. But in a more pre-scripted fashion. Something like -- "You've got a new message from Paul, read it or ignore?" -- Read it. And so on (including the dictation and sending of the reply)
That's not how I think of 'worked like that'. I have no interest in learning pre-scripted voice commands. Once you have everything scripted, implementing an interaction as I posted is almost as trivial as a CLI. That's why I put 'natural language' as the first criteria, which renders the second task a lot more challenging.