Oh, it's the obnoxious Apple guy on youtube again.
As a Windows Phone user, I'm used to my devices getting outperformed in benchmarks pretty much all of the time. But people tend to take performance benchmarks as an indicative as the whole platform and that's just missing the point. It's what you can do with the device with that performance that matters. If you're so simple minded to judge a device and a whole OS just by one of the many things that it can perform, well go ahead, keep on chasing that carrot.
For me, I don't really care about Office being on the Surface. What is awesome for me about the Surface is that I can consume content on this device almost as fast as on my desktop computer due to it's multitasking model, user interface, swipe gestures, and snap modes. What slows me down a bit compared to the desktop is no site-specific zoom levels, scrolling down long pages take longer compared to a scroll wheel, not being as fast at opening multiple tabs, and websites with a lot of small touch targets. Even with those, I don't find myself wanting the Desktop to consume content because I can bring my Surface almost everywhere and touch is still better for speed in general navigation and to navigate Metro.
Being able to consume content this fast on a device meant for touch is awesome. So how about you benchmark the speed of switching apps and multitasking in the Surface vs. other tablets? Would you judge a whole device by that then?
Edit: Oh yeah, another thing that really helps on content consumption is that I don't need to transfer content from my server or my other computers. All content I have is accessible through the network and I never have to think of how to transfer that content over to my tablet device. This is definitely not applicable to everyone, but this makes the Surface not another device I have to independently manage. It just needs to play mkv files though...