He's actually almost 100% right. I disagree with the us vs. them narrative though. There a lot of different ways to look at what has happened in recent year. One way would be to argue that FLOSS killed the desktop by showing that there's no more money to be made with it. Another way is just to look at it in a more down to earth fashion: while a lot of talented hackers were pushing FLOSS to world dominance, another bunch discovered that world dominance had already been achieved and it was called a "browser" and all that was needed was Ajax to co-opt decades of established application silos. And guess what, they were right.
So, for better or worse, there will never be a "year of the Linux desktop." But we are most definitely, and will continue to be for the foreseeable future, in the time of the Linux server. We've also most certainly entered in the decade(s?) of the Linux mobile. But both in the case of server and mobile, FLOSS is taken for granted as a building block for far more richer applications. Android, for instance, bares little to no resemblances to a Linux desktop, yet it runs the Linux kernel.
I personally feel that somehow becoming webish is so far away from what Gnome does today that it's like trying to make an elephant fly. However, I would see how Gnome could revolutionize the mobile world. The fact of the matter is that Android is only open in the marketing sense. And the other Linux-based efforts are *way* behind Android. Gnome could most definitely be an attractive platform for pads and, with some zeal, for phones. But that would require rethinking some of Gnome's core paradigms. One thing that Gnome has in its favor, though, is that it's already started with a minimalistic mindset (recall Linus' take on Gnome's interface from a while back ...)
I know Gnome has already tried to make itself relevant in the mobile world, but that really went nowhere ... Now is a good time.