Ultimately, I suspect that the "tablet vs. notebook" race will shake out such they both continue to exist serving different applications. There will be some overlap, certainly, but tablets, at least for the near future, can't completely fill the job of a notebook.
The popularity of tablets is not a reflection of the uselessness of notebooks in general. It's a reflection of the fact that notebooks don't cover all application, just like desktops don't cover all. It's not yet time to stop advancing the state of the art in notebooks.
I don't think this is to be dismissed too cheaply, there are plenty of Tablet users who are yeaarning for a handheld which can do more and not be chained to Apples diktats on what you can have on your device.
I think the key to making a tablet killing ultrabook is the right combination of energy efficient CPU and graphics paired with a fast solid state drive. Which, as another poster noted, is basically what the Macbook Air series is all about. Again, Apple has seemed to find that balance in their second generation machines. Why can't other companies duplicate that combination at a lower price point? Maybe these new Intel microprocessors will do the trick.
Complex instruction set processors will always use significantly more power than RISC processors using the same process technology.
Simplified operatin ssytems (Android, Chrome) are the natural platforms for mobile devices. As the role of complex operating systems diminishes in mobile devices, so will the role of CISC machines.
Tablets and moble devices will become 'smart' terminals to distributed complex systems running on the cloud of servers.
Intel is trying too hard to impress upon low power chips. They have been late and importantly missed the boat for the tablet segment but they can come up fast and strong. True that not many would want a high processor energy gulping tablet which needs to be plugged in every few hours.
I had an Windows Based Pocket computer afforded me by the company I worked for between 2001 and 2005. It had full WIFI access and all the MS-Office compatible stuff you could shake a fist at.
It barely got used. There were others at my company who would say the same thing.
When it comes to open architecture, Microsoft whipped Apple for volume of sales from the 80s through all the way to this decade. Largely because of open hardware and software architecture.
I've seen Windows CE based stuff that persisted in being MS-Klunky well into this decade and I don't see them changing their rigidity.
I'm betting that MS-Windows variations remain MS-Kluny and Apple-Apps remain Apple-Snobbish and the Droid Open Architecture continues to grow like crabgrass across the cloud and non-desktop market.
This is going to go against the industry momentum towards centralizing the computational heavy lifting through cloud computing. Intel is going to have to come up with a reason to have all of that power at the edge if this is going to fly. Otherwise it will be tough to justify the extra cost and battery use.