I was speaking more towards the granularity of functionality with voice recognition, e.g. MS possibly may not have to have the voice app respond to "Do I need an umbrella tomorrow?" to be competitive.
Sure, the implementation isn't as "neat" but if the end functionality is on par or there's other functionality [see post that you quoted me from] that supplements for the lack of such design then who knows it could still work out.
Of course, people generally like all that glitter and glam huh?
Glam, hmm, hadn't thought of it that way. Some google searches don't turn up what I think they should, and if different phrases or words are used, its spot on... this is where I think siri shines:
"Is jfk closer or lagurdia?" Wouldn't work at all for android or telme, the results would be interesting on siri, and nkot glam at all. Planning a trip via voice searches in stalled traffic would be a great example of how this is a good implementation.
I'm just starting to use voice features on android for sending emails / txt / directions to people, but with more kids introduced to it I can see a completely different landscape of profiles... instead of glowing faces staring down at their phones, they tell it to do something, and it just works.
Lots of possibilites, just like we didn't imagine free driving navigation a few short years ago...it goes on.