Facebook's cloud isn't for the birds | Opinion

Facebook isn't exactly the toast of the town right now. In the world at large, the mammoth company's slapstick "oops clumsy me, I was trying to ogle some college girls at Harvard and I seem to have accidentally knocked over global democracy" schtick is wearing thin for many people.

Far more specifically in the games space, the firm's recent handling of Oculus has left plenty of people deeply nonplussed. The decision that deleting your Facebook account would also delete all your Oculus games, in particular, went down like a bowl of cold vomit -- which, in concert with the fact that many consumers and industry types alike view Oculus as the only thing of any relevance to games that Facebook actually does, makes it pretty unsurprising that the company's recent cloud gaming announcement flew pretty much under the radar.

I'm not sure it'll remain under the radar for long. What Facebook is doing -- rolling out a cloud streaming service for games on its Facebook Gaming portal on the web and mobile devices (though not iOS, for now at least) -- shouldn't be dismissed as just another tech giant hopping on the cloud streaming bandwagon. Yes, every Tom, Dick and Harry with a globe-spanning network of data centres and a few billion to burn seems to reckon they're going to be Netflix But For Games (and they're probably all completely wrong), but what makes Facebook's proposition interesting is that it isn't even remotely suggesting being Netflix But For Games.

What makes Facebook's proposition interesting is that it isn't even remotely suggesting being Netflix But For Games

Rather, it's targeting a completely different market segment -- one that seems certain to be much more receptive to cloud streaming games, not least because it's one that doesn't know or care what cloud streaming actually is.

Let me back up a second. Facebook Gaming is a thing a lot of people in the industry don't think about very much, and probably haven't thought about very much since the heyday of Farmville and Mafia Wars -- but it's still popular, just primarily with the kind of customers who don't generally venture beyond the solidly "casual" end of the market. That's "casual" in the weird sense of "casual" as a style of game, as distinct from "casual" in the sense of "not spending a ton of time or money." People do spend plenty of time and money in these games; it's just that being free-to-play games primarily aimed at non-core audiences makes them "casual" in a somewhat ill-defined and uncomfortably pejorative sort of way. Still, we don't have a better word for it, so "casual" it is.

For the past ten years, Facebook's gaming offerings have mostly been HTML5 games -- browser-based products not dissimilar to what you could achieve with the late, unlamented Adobe Flash plugin, making them very portable and available across almost any device you could imagine. HTML5 is capable of a surprising degree of complexity in terms of game development, but there are some pretty major limitations to what the technology can do nonetheless... which is where streaming comes in. Facebook's proposition is that, by setting up a cloud streaming platform, it can allow people to play significantly more complex, high-fidelity games in exactly the way they play Facebook games right now -- on their phones, on their janky old laptops, or whatever it may be.

This is a huge conceptual difference compared to something like Google's Stadia, or any of the other existing cloud streaming efforts -- and in the short- to medium-term I suspect that it's far more likely to succeed, finding an audience and generating actual revenues for itself. The crux of it is that Facebook is treating cloud gaming as a way to upgrade a currently low-fidelity experience -- its existing Facebook Gaming titles -- whereas Stadia and every other cloud service are essentially offering downgrades to an experience that's currently high-fidelity.

Moreover, Facebook is suggesting that it'll do this transparently, simply making cloud streaming one additional option for developers working on its platform. End users won't need any additional hardware or pay any additional fees, and will likely neither know nor particularly care that their game is being streamed from a data centre somewhere. One day they'll click on a game in Facebook's portal which will quietly start streaming in rather than rendering everything in their browser. In theory, the only thing they'll notice is that it looks better than other things they've been playing.

This is the first time a company in this space has proposed a use case for cloud gaming that's actually something consumers and developers want

That's a profoundly different offering than telling people who play games on PC or console that they can pay an additional fee to play the same games that are already available to them, but trading off a major drop in fidelity and reliability for the capacity to play in different places or on different devices. Actually, I'd argue that this is the first time that a company in this space has proposed a use case for cloud gaming that's actually something consumers and developers want -- better quality casual-style games in their browsers or on their phones -- as distinct from being a service whose existence was clearly reasoned backwards from a desire within the company itself -- namely getting people to pay money to do stuff with all these GPU compute units they've been shoving into cloud data centres at exorbitant cost.

The reason I'm so positive about Facebook's cloud gaming approach -- despite being a pretty consistent cloud streaming naysayer and no great fan of Facebook itself -- is partially because it shows rational thought about how to fulfill a consumer need, but also because it pretty much demands the creation of actually new, "cloud-first" games to take advantage of this format. The reason I've rarely been positive about cloud streaming isn't because the technology doesn't hold promise -- it absolutely does, though the infrastructure to realise some of that promise still isn't there -- but using that promise to run lower-quality versions of games you could be playing on your console, PC or even smartphone is simply squandering the potential of cloud streaming.

Personal computing power has never been cheaper. We all walk around with extraordinarily powerful CPU and GPU equipped devices in our pockets, let alone on our desks or under our TVs, so "hey, why don't you just use that as a dumb terminal for something running in a data centre" isn't a terribly meaningful proposition. "Hey, here's something genuinely new and different that's only possible because it's running on a centralised system and streaming to your device," though? Sign me up.

Like much of the free-to-play world, it'll likely be decried by a lot of the core audience

Existing PC and console games don't leverage cloud services' unique advantages and potential, and there's a huge chicken-and-egg problem here. The audience for cloud gaming is limited because it's all just existing titles playing in low quality, so there's little incentive for developers to explore the possibilities of cloud gaming because there's a limited audience... and so on. Unless a cloud company is willing to make an absolutely gigantic investment in getting game studios to develop cloud-exclusive titles which explore those possibilities, that's a hard loop to break out of.

What Facebook is proposing is quite outside that vicious circle; some developers may attempt to make existing console or PC type games run on the Facebook service, but the appeal will, I suspect, be very limited simply because it's the wrong context for that kind of title. Rather, developers are going to have to look at what actually works in the Facebook Gaming context -- a huge audience, but one that's quite distinct from the usual PC and console crowd -- and then think about how they can use the cloud gaming technology to build new types of game experience for those consumers.

Like much of the free-to-play world -- and Facebook is blunt about this being a free-to-play space -- it'll likely be decried and disliked by a lot of the core audience, but as a space for innovation it's a fascinating experiment in the making. The notion of paying $50 for an ephemeral right to stream a specific game from someone else's servers is a horrible Frankenstein's monster of a business model, to be honest; free-to-play, for all its well-explored problems, is a far more comfortable fit for what streaming actually is and does. Subscription tiers could potentially work too, but as a lot of media companies are discovering, trying to get consumers to pay a subscription for content is bloody hard if you're not Netflix or Disney, or their equivalent within a given medium. Facebook Gaming is certainly not that, and likely never will be.

I don't honestly know how much attention the broader industry will pay to what's happening here, given how many people will be turned off by the free-to-play nature of the whole thing -- but their lack of interest won't last long if this does turn out to be the one area where cloud streaming genuinely works and makes money in the coming years. Facebook's initiative sidesteps the biggest problem in the sector -- the fact that infrastructure isn't remotely ready for console-quality experiences streaming from the cloud, and that many of the biggest innovations speeding up other kinds of cloud services, like edge computing, aren't actually all that relevant to that kind of game -- and instead looks at the infrastructure and technology that actually exists today and figures out a way to apply it to improve an existing, popular platform.

In the process, it will push developers to think about the cloud not just as a "long tail" revenue option for games they'd really rather be selling for $60, but as a whole new platform whose advantages, disadvantages and unique possibilities demand to be explored. While we're all talking about the some-day, some-how potential of other cloud services, Facebook's cloud streaming games will likely be the most popular and profitable use of the cloud by the games industry for the foreseeable future.