I don't see how the preview box is any faster than conventional tab usage:
result miss case: read the preview, decide you don't like it, go back (no difference)
result hit case: read the preview, decide you like it, click "yes I want to read this in the window" (less work by one step)
I'm also not sold on the point of ubiquity -- it seems to me that punching most of this stuff into google already provides what ubiquity promises. Directing this traffic at search engines gives users more opportunity to select the implementation they like (say Yahoo search does something better than Google). Pulling it away from the default search provider also means they're directing less traffic at Google, meaning less (potential) income for the project. This clearly shouldn't be a primary consideration but I don't see a win in any direction.
Reichenstein also states that tabs could not be the web’s next biggest thing as tabbed browsing isn’t such a big deal in many people’s minds.
Tabs are six+ years old (or so). There's probably no more easy growth. People that are not proficient enough with computers don't want them or care about them: my friend who lives on a farm and believes every pop-up that appears on his screen will never benefit from them, and that's fine. However, it would be entirely expected to find out that the take rate for tabs with the 12-20 set is dramatically higher -- kids use computers differently than their parents do and aren't encumbered by the same analogies that their parents use to understand what is going on on the screen.
I contend that people will grow to the tools if you provide them. I'm not convinced anything but the tree-tabs touched on here are actually forward progress for the browser.