Of course, I want an Apple TV and an iPhone, but I may actually wait a bit.
Quentin has a good point about the Apple TV. Without a DVD player it does actually complicate use cases for the time being. Although a DVD player is just $20-40 these days, I personally don't have an easy way to mix two composite video inputs to the second hand CRT TV we use these days.
I'm disappointed that there isn't (yet) a hard-disk-based iPhone with 120GB or so, so I can watch videos on it, and maybe stream the videos to the Apple TV, and use it as a remote control. I also would have liked a non-cell-phone version with just WiFi, and we're tied to Verizon for a while now, so I may wait for gen 2 or 3 to get it. I liked their comment that the killer app for cell phones is calling, and their focus on making that easy. I hope they really did. I think if they were to combine it with a better text input method (and I've seen one that fits the bill) it could be awesome.
But, my main concern is actually about extensibility. While I think extensibility is asking for stability trouble, and can really hurt your reputation unjustly, they have to allow for an ecosystem here if they want to succeed in a major way. Apple can provide a great phone without it, but there are so many 3rd parties that could add value to the system that they'll make another lose-to-MS mistake if they don't allow 3rd parties to develop for it. Even developing for it in significant ways, changing the nature of the device, since there are so many cool and useful things that can be done with camera/video input, storage, connectivity, a great I/O screen, speakers, processing power.
How they should do it is another story. Creating an ecosystem and creating great integrated products haven't fit together traditionally. Maybe they should have two products, one for hackers (read 3rd party developers), and one for the consumer, and only gently offer the bleeding edge stuff to the mass market. Maybe users should need to attach a discrete hardware doohickey to the iPhone to allow it to run non-certified software.
But, I think the best outcome for Apple would be if the iPhone, like the iPod, allowed 3rd parties to create new solutions based on it as a platform, e.g. building a GPS device with very low R&D costs by extending their platform, etc., following the Handspring and iPod model.
I do hope, however, that we'll eventually see businesses and business models succeeding with much more open solutions, where hardware and software design and implementation details are all open sourced.