Three things on that review were somewhat misleading.
The screen is terrible so this is not a good product.
Yes, it is well known that the screen is not hi resolution but it has also been stated that it is not for consumers to buy and only a dev version. With everything, software support is key and getting the groundwork laid now is critical for its success. They have stated many times that consumers should not buy this version and the consumer version will be at a minimum 1080P. A preview that states the product is bad because of this is missing the point. If people that are unfamiliar with the Rift are being informed to wait for the consumer version that is one thing but to say it is outright bad and can't be fixed is just not accurate. In the end most of us are willing to sacrifice some resolution for more immersion overall.
You need 60 fps minimum and need to render twice as many frames to get 3D.
First, this would not be the first item or game that needed a higher end system to fully support it. Having said that this is also an untrue statement that these are required for the rift. 60 fps is recommended for true immersion but it works below this and many people have reported running it much less with no issues. So to say that if it ran at 60 fps for the whole game but a few times it dips to 40 is a deal breaker is simply wrong.
As for the performance impact this is no different than 3D that is available today without the rift. With that there are two modes you can use. Z Buffer renders a a single view and then modifies it for the second eye. This has on average has a 10% hit to performance and is a viable option for people running a resource intensive game on a lower end system. For less demanding games or ones on higher end systems you can use geometry based instead which renders both eyes separately. This is optional based on the customers system. Turning down the options but getting more immersion would likely be worth it as a worst case.
No positional tracking (6 degrees of freedom)
This again has been mentioned will be in the consumer version. I've used Track IR from the beginning and I remember when it only had 3 degrees of freedom as a production product. Rift will have 6 at launch.
Having said all of that, should everyone go out and buy a dev kit now? No, probably not. It's missing several key things that will make it a comercial success. These are all planned to be fixed and there is no reason to think they wont. Following the mobile phone tech curve is a perfect way to continue to improve the product every year or so which tells us it will keep getting better and better.
A lot has been said about the overhyping of the rift but the truth is it is real people trying it themselves and experiencing a level of immersion that hasn't been possible before even with the less than perfect screen. Imagine how good it will be once those items are improved before launch.
The great news is the dev kits are here and almost all engine companies have committed to support it. If we can get the support in now then when the production equipment is ready we will be set to move into the most exciting thing to happen in gaming in a while. A viable, affordable and fully immersive VR experience that we all have dreamed of.