Users who want access to two operating systems are power users since they know about the boot process of an OS. Average users don't know. You could argue that the number of power users is growing, but absolutely, its very small and will likely remain so.
BS. Windows 8 was released with a $40 upgrade for Windows 7 users. Obviously there was tons of hardware out there on which it could run.
From an Engineering standpoint its not that complex and typically this is the only standpoint Intel takes. From an end use standpoint, dual-boot is adding unnecessary complexity to a single OS. It's much like Windows 8 which has a metro mode and a classic mode.
Windows 8 is complicated so is dual boot. People will stick with straight Android tablets, mostly ARM based and the richer ones may get iPad. In the tablet space, the tone has been set by the iPad which is a very easy to use product. A novice user can start using one in minutes. Microsoft seems to overlook this fact with their design of their new OS.
+The sales numbers have spoken. It looks like Microsoft has a veritable hit on their hands. With the intro of Haswell, some of the criticism around this PC will get addressed. Also, this leads the way for core series chips to be put to use in tablets which is probably a win for core series sales volumes. If I was Qualcomm or NVidia, I would be very scared about competition for tablet chips in this form which neither has a chance of competing with at least on the performance front.
Windows 8 it seems is set to take off as a touch OS albeit a little slowly contrary to earlier reports.
Looking at the big picture, there now exists competition for Intel in servers. That's bound to drive down ASPs and ARM based startups like Boston don't need much to survive in a race to the bottom. As startups, they don't have market capitalization to worry about like the old established Intel.
Intel is an out of favor company at the moment. Whoever comes in needs to change that. Not an easy task for such an old and massive company. There's a lot of inertia working against you.
In exchange for paying foundries and ARM licensing, they have significantly lower R&D costs and don't have to invest in their own fabs. The barrier to entry is therefore a lot lower. That's why there are 3 or 4 big ARM licensees that Intel is going up against alone.
That's my estimate based on knowledge of CloverTrail battery life and Snapdragon and the changes since for both chips. Note that for Clover, you have to add the battery drain of the Baseband chip which is not integrated like the Snapdragon 600.
This is the magazine's opinion. Gehlsinger also had several recognized successes preceding Larrabee so one failure doesn't wash away the past.
Intel CEO is a challenging position though not very enviable. Not sure Gehlsinger would be interested. He's smart enough for the job though and that's all you need for Intel.