Shaker, a new start-up backed by TechCrunch's Michael Arrington, is a Facebook-powered virtual chatroom environment that is, in the words of its founder, "like a Second Life, except you’re yourself.” It's a 2.5D simulated space where your Facebook profile literally floats above your avatar's head, and where you're able to chat and interact (dance, buy drinks) with other Facebook users, with your Facebook-defined common interests acting as an icebreaker. (As pictured here: Maria, Maria, I just met an avatar girl named Maria... and we both like Obama.) Unsurprisingly, given Arrington's backing, it's getting a lot of buzz in the Silicon Valley tech community. (My friend Liz Gannes has a really good write-up on All Things D.)
My own initial take? There's a lot of potential in linking avatar-based chat to Facebook profiles. City of Eternals, the Facebook-based MMO I helped develop a couple years ago, also had that feature, and even in its early Beta stages, generated some really interesting social interactions. And while it's also early days for Shaker, I think it has a steep road to succeed -- here's why:
There's no room-based chat in Shaker, as far as I can see (it's all player-to-player interaction), no user-generated content, no gaming systems or game objectives. The avatars themselves are totally generic, unappealing, and undifferentiated. Zynga's YoVille linked Facebook profiles to avatars and chat two years ago, and was pretty successful for awhile, but has been dwindling since. (Not a good sign.)
What's more, I don't think Shaker's developers understand why Second Life has succeeded (at least on a profitable niche level): A big appeal of SL is not having to link your real identity to your avatar, freeing you up to be more creative, take more chances, publicly express parts of yourself that are important but that you generally keep to yourself. (Also a big reason why pseudonymous user-generated sites like DeviantArt and Reddit do so well.)
A more relevant comparison to Shaker is Habbo Hotel, since they're both web-based, and not full 3D like SL. Habbo's doing pretty well (now with about 15 million monthly uniques I believe), but again, Habbo has succeeded with features that Shaker doesn't have: User-generated content, pseudonymous identities, gaming systems. (All of which Turntable.FM, another new virtual social chat startup, does have, which is probably why it's getting good traction.)
Again, this is just my initial take of an early Beta product, so it's not fair to write it off just yet; I'm very curious to see how Shaker develops. But for now, I'd say this: Shaker probably needs to shake out some of its underlying assumptions, to really succeed.