I’ve had my Google CR-48, the reference implementation notebook for Chrome OS, for a week and a half now. You can read my earlier reviews here and here.

The pilot program asked that you use your CR-48 as your primary computer. Although my employer’s privacy rules would restrict me from doing that for my day job, it was my full intention to do so for all other use, which includes personal computer use as well as writing fiction.
It turns out that the browser has some shortcomings as the primary user interface, and the CR-48 has some hiccups. As I mentioned in my last post, two issues include the lack of a taskbar/dock/start menu, which makes quickly launched desired apps and getting notifications awkward, and the lack of fixed menu/buttons in web applications, which makes every invocation of functionality a cognitive task, rather than a muscle memory task. (On most web pages, menus and buttons float with the content, rather than remaining in a fixed location.)
Add to that two other issues: The CR-48 is slow. Slow enough that even when I plan to do a browser based task, if it’s a big enough task, I find myself thinking that maybe I’ll just run downstairs and use my MacBook Pro. It also does very poorly on YouTube. (Hello, last I checked, YouTube was a Google property, yes?) And Pandora. And sometimes the simplest page refresh takes an awkward amount of time – as much as 20 or 30 seconds.
The other, lesser, issue is that resume is slow. Sure, the screen is on, but the notebook is still negotiating the wireless connection for another 10 or 15 seconds. By that time, I’m already engaged with the keyboard trying to get to whatever app I want, only to get an error message because the Internet isn’t available yet.
On the plus side, it’s still extremely portable, and has a stunning battery life. I went about four days over the holidays without plugging it in. Granted, I wasn’t using it that much, but in that time, even just standby mode on my MacBook Pro or a Windows PC would drain the battery. That portability and battery life makes it very convenient and tempting, so I do reach for it first – even over my smartphone when around the house. But the slow speed is a barrier to any serious use.
During the Chrome OS presentation two weeks ago, they said the Chrome OS development team had been using these as their exclusive computers for months. I can’t imagine how. Am I just missing something? (And how would you do software development in a browser?)
On the topic of the trackpad, I have figured out by trial and error that by seperating my fingers by at least a quarter of an inch, I can reliably right-click and scroll. However, I still haven’t figured out how to reliably click-and-drag, an operation that is terrifically easy with a mouse, possible with a Mac trackpad, and seemingly impossible with Chrome OS. 
Despite the negative aspects of this, I am still hopeful that some of the performance issues can be addressed with software updates. And hopefully by the time Chrome OS comes to market, it comes with a higher powered CPU than an Atom processor.
In the meantime, it makes a great second notebook, and fulfills the classic role of a netbook. But I’d like it to be viable as a primary computer, which is what is needs to be in order for it to have strategic value to Google.