People here need to decide what they are discussing, because they are mixing things up.
The one question, if the Surface is going to suck or not, or if it's going to be successful or not, is a side issue, because it's an issue of execution, timing, current state of technlogy and the capability of the marketing and sales teams.
The other questions, and that's what the headline of the article suggests is the main issue, is this:
There is 'a strong possibility' Apple needs a Surface-like device
A Surface-like device is a device that converges two product categories. This has a tradition. e.g. smart-phones are convergent device of what once was called a PDA and a cell phone, and then modern smartphones are convergent devices of anything from FM radios, GPS navigation devices, point&shoot photo cameras, Fliq-like video cameras, PDAs, and mobile phones.
At every turn, any marketer worth his salt would try to keep product categories apart for as long as possible in order to be able to make two rather than one sale. One of the reasons Apple was so disruptive in the market, because they were the ones who agressively brought these categories together, crossing the strategy of the rest of the industry of selling many iterations and generations of evolutionary change to maximize profit, by being the new kid on the block who offered a short cut to the future.
The convergence of tablets and laptops is pretty much a given, because making computers faster and more power efficient means making them smaller, which means eventually they will all look like an iPad in the sense that the computer is a chewing gum sized stick somewhere in a case otherwise filled with batteries. The question is only when that stick will offer enough compute power to be the equivalent of a laptop.
Look at the development of the desktop computer: things got smaller and smaller, until eventually (witness the iMac) the screen became the computer. If you have a laptop in which the screen becomes the computer, you have in essence something like an iPad, plus a (detachable/optional) keyboard, and that in turn is a Surface-like device, in that it's both a tablet and a laptop, depending on how you're going to use it.
How usable such a device is, depends on how well the GUI adapts to the current mode of usage. The old WinCE showed how not to adapt a desktop OS for phone/tablet. iOS showed how to adapt what is in essence Mac OS X to a phone/tablet. Nothing in software forbids to have an OS to have multiple GUI personalities. Just as the iPhone/iPad can adjust to horizontal vs. vertical screen orientation, or iPhone vs. iPad screen size, it could adjust it's entire GUI depending on whether the device is in laptop or tablet/phone mode.
The whole post-PC device thing is a marketing slogan. It draws attention to a particular aspect of technology in such a way as to create in the mind of consumers two distinct product categories that makes them think they need to own two devices, and Apple would be stupid not to do this, for as long as they can, because it helps their bottom line.
Jobs was known to create market categories that made his products look like a winner, so much so, that back in the day when the excellent NeXT workstations had trouble getting adopted due to the chicken&egg problem with users&software, people were joking that NeXT was the world leader of black education workstations.
Personally I think that the Surface is not ready for prime time, because the ARM based one will lack compute power for real laptop/desktop apps, and the x86 based version is too thick and heavy and will likely still have too short of a battery life to be a decent tablet. But just because something is too early, doesn't mean it's the wrong concept; just technology isn't ready yet.
It's also clear why MS is gambling on doing this now: it's one way they can hope to take the lead again by creating a category in which Apple looks like they are playing catch-up, and that if successful may force Apple to end a product strategy that makes people buy both a laptop and an iPad, which would shrink their profitability quite a bit. They risk, however, that their product is so much premature, that it becomes a joke for its flaws, which would allow Apple to ride on its current strategy even longer, and raise their long term profitability, by delaying the point when they have to offer a convergent device to the point when Android/ChromeOS competition may force them to enter that market, at which point they'd have the chance to show once again "how it's really done".
So it's not a matter of if, it's a matter of when, and of who gets to claim to have the leadership role in the segment. Some may remember the reports from quite some time ago, that Apple supposedly internally tested MacBook Airs based on ARM chips, and Apple supposedly being surprised by how well OS X performed on an ARM CPU. You can bet, these experiments were made in view of the time when convergent devices were pushing onto the market, knowing that x86 CPU power consumption would be a big stumbling block. So for the very reasons MS does now an ARM based version of Win8, Apple must have an ARM based version of OS X in their labs. The whole one OS, multiple CPU types scenario NeXT/Apple has played out multiple times, because NeXTSTEP/Darwin/OS X/iOS and it's apps ran on a variation of m68k, m88k, RISC, MIPS, SPARC, x86, and ARM, and at almost any given point it had the ability to create universal binaries that ran on more than one CPU platform.
So don't be surprised when in the not too distant future MacBoook Airs have quad-core ARM chips instead of dual-core x86 chips, double their battery life and retain roughly the same performance, while coming down in price by a couple of hundred bucks to light the fire under the Surface. Might as well have a detachable/flippable touch screen that doubles as iPad at that point. But of course, Apple is in no hurry to compete with itself to lower its own profits. So as long as the Surface and its imitators don't become a major force in the market, Apple will do its best to convince people that tablets and laptops are two things that shall never meet...
...and by most of the comments here, the Apple marketing department is doing a good job; for now. (And they will do an equally good job at convincing everyone why when they introduce such a device, none of what is said now matters anymore...)