Alexa is the perfect example of a product that could have been made by Apple but wasn't, and which ticks all those boxes.
And not just those. Apple's big strength isn't hardware, it's ecosystems. The hardware was always just a front end to a network of unique systems, markets, and services.
The latest Apple offerings seem to have forgotten that. Watch is primarily a shiny widget, not the thin client the iPod and iPad were. The latest Macs are heading in a similar direction. There are more services, but they're less appealing to users because the UI designs have become clunky, and there's no sense of forward momentum among many user groups.
With Alexa, Amazon has proved that it can play the consumer game too. Alexa includes links to content providers, links to hardware manufacturers, and links to other software services.
I wouldn't be surprised if there's a crash R&D program to create Apple's own version of Alexa. (I would be surprised, but maybe not completely, if Siri Home appears on the 27th. I'm guessing next year is more likely and 2018 isn't impossible.)
But it may be too late. Amazon has a first mover advantage, and it can only drop the lead by allowing the platform to stagnate.
Amazon may well do that. Bezos seems to suffer from occasional new product ADHD, so he may not follow through.
But there's certainly huge potential to take Alexa much further, and if Apple is going to compete any R&D project has to aim for where Alexa could be three to five years from now, not where Alexa is today.
Which doesn't change the fact that Apple could have made Siri Home first. The hardware was always available. The only thing missing was Cook's lack of vision and drive.