I turned off JavaScript to see if the site still worked: it did. I then navigated to different pages to see if it was slower than the "blindingly fast" the blog post promises: I couldn't detect any difference.
In summary, I have no idea why you would use React to build a static site, but I sure hope they had fun doing it.
Another datapoint. Difference in speed is obvious to me coming from a Comcast pipe. There's an actual page load when JS is off vs instant page change when it is on.
I love what you guys have done. I'm interested to know what is the rationale for this? Are you finding that it helps improve conversion or something? Basically help me convince the business side of my company so that I can use this for our site too. :)
The performance wasn't really the main requirement for the project, it just fell out of the choice of using React.
We basically wanted to manage internationalised static content effectively. That meant reusing markup, but also making sure that you're linking to pages in the correct locale and not trying to link to pages that aren't available in your locale.
React is obviously good for templating and its been quite straightforward to implement the internationalisation logic too. Since it isn't much overhead at all to render on both the client and server, it seemed a no brainer.
In summary, I have no idea why you would use React to build a static site, but I sure hope they had fun doing it.