Welcome to the metaverse ?

There's a recent piece in Wired which is so awful I refuse to repeat or link it here. Basically it says that Facebook's determination to push ahead in becoming a "metaverse" company is akin to lavishing praise on a "rich white kid". That is, developing virtual reality is apparently racist because they should be building schools and hospitals instead.

A much better piece on the metaverse can be found here. Basically, the concept is much like that Futurama episode where they surf the web in the year 3000. Instead of reading text and watching YouTube videos, it's a vast, sprawling 3D simulation of a city. Websites are buildings, stores, etc. - it's a "real" place you can wander around in and talk to people as you would in real life. Zoidberg - sorry, Zuckerberg's vision is basically this, after a fashion.

So, is this (a) technically achievable in five years as per Facebook's vision, (b) desirable in principle, and (c) seriously, does it have to be Facebook ? Here's my take as a proudly Oculus-wielding VR acolyte who sincerely wishes Mark Zuckerberg wasn't a thing.

To the first one my guess is yes but with major caveats. Of course, no-one is really envisioning that walking around the internet as a literal simulated city would be in any way practical. Rather, imagine something more like a super-advanced version of Microsoft's HoloLens : the ability to bring up well-rendered, 3D graphics wherever and whenever you are, using data from across the internet. Essentially, portable augmented reality on demand. 

The technical state of VR is looking pretty rosy. Headsets are getting lighter, more powerful, and cheaper. Standalone headsets can do some pretty impressive things, but couple them to a PC and the graphical quality improves dramatically. Given developments like the Steam Deck, I don't think it's at all crazy to postulate affordable, entirely standalone (that is, a single purchasable unit), wearable, PC-quality VR within five years. Maybe it won't be run by the headset itself, but by a Deck-like device in a backpack, or streamed wirelessly from a remote server. 

(Actually I'm a bit surprised that there's been so much focus on developing more powerful headsets rather than just giving them dedicated streaming equipment to let a PC do the heavy lifting, but the point is there are a variety of solutions to this part of the challenge.)

My expectation of a device needed to make the metaverse feasible would be something much closer to a beefy pair of glasses than the current helmet-like googles. Something weighing 1-200g or so, that's no more difficult to put on and no less comfortable than regular glasses. It may or may not have a separate accompanying wearable device to do the main computations. The glasses themselves would include high-resolution colour cameras to provide a 3D display of the real world, together with a LIDAR scanner and of course accelerometers. This is not too far off the capabilities of current high-end smartphones.

The two key features are the need to be wireless and to provide real-world visuals, both of which VR can do today but in a severely limited fashion. With a wire, all other improvements are almost neutered, since you're still tied to a clunky external device : convenience is destroyed, and climbing into the metaverse would be something you'd have to dedicate special time and effort to. This is fine for games, but useless for regular productivity. Currently, the wireless PC streaming works pretty darn well on the Quest, but not all routers meet the ideal capacity and not everyone has or needs a high-end gaming PC, hence the need for a standalone package.

The requirement for real-world visuals is that a big problem for VR is the space requirement. Again, it's fine for dedicated gaming if you have to set aside an area, but not much fun for productive work if you have to clear out a space every time you want to use it (especially in small, crowded offices). Having a simultaneous clear view of reality - I mean one that's good enough that you could tolerate it for hours at a stretch - would eliminate this need. Being able to overlay graphics* means you could use real surfaces as an advantage rather than a hindrance (and using a screen instead of clear pane allows the possibility to switch back and forth between VR and AR as required). This is only just about possible right now, and the impact is not even close to being fully realised. But in five years, with Facebook's resources ? Probably possible.

* And they have to be high-quality graphics. Currently things are sufficient for immersive gaming, but not for reading the large quantities of text that actual work requires. There are people who swear by VR apps for productivity, but my experience thus far has fallen far short of what I need.

So would this be desirable in principle ? Yes, but as to whether it will really be something widely adopted in five years, I'm very skeptical. It might be technically feasible in five years but it will take a lot longer before the levels of reliability and adoption become close to what Facebook hopes. VR devices are threshold-limited : get anything slightly wrong and everything becomes worse. None of the technical requirements are outlandish, but they all have to be done well or it just won't work. Getting to the required level of quality and especially price in five years would be a big challenge, given the rate of development of Oculus devices.

Similarly, would people really be likely to adopt this en masse as Facebook hopes ? Here the Wired author has an abysmal lack of imagination. Suppose you could bring up technical documents on a whim, arranging them in three-dimensional space around you however you like. Instead of needing multiple expensive monitors, or arranging multiple desktops, your whole environment would become a giant screen. But then also imagine bringing in collaborators, who could walk around your virtual world inspecting documents as they please, annotating and drawing, making physical gestures... this is a huge change. Would you ever want a conventional monitor again ? I doubt it. At most you'd want a physical keyboard for the tactile experience of typing.

I think the full implications of a true 3D AR interface are only going to become apparent when we've got one. We'll start with reproducing existing GUIs, for familiarity, but from there things will diverge in unpredictable ways. 

The convenience of this is key. If it's an expensive piece of equipment that's laborious to set up, that makes the experience qualitatively different from the case of being no more burdensome that putting on a pair of sunglasses. Sure, if my main task is managing spreadsheets, a clunky, pricey headset would offer only disadvantages. But if it's so easy to put on that it's no burden, and if it lets me do all the same stuff I do in conventional interfaces and more, why would I ever opt out ?

Likewise the potential everyday uses are manifold. I go on holiday, I see something interesting - I instantly bring up information with a wave of hand, with no need for bringing out a phone and typing out a query. I remain fully immersed in the environment I've come to enjoy but with a richer experience, not a diminished one. I no longer walk along the streets glued to a phone and bumping into lamp-posts. I bring up instructions on cooking, maps, DIY... I can examine objects and animals, history and science in a way I could never do otherwise. I experience more of life than I would otherwise, not reduce myself to a tiny screen.

But don't mistake my meaning here. This is not some glorious technocratic solution to all the ills of the world. Those who aren't interested in a topic won't become more interested just because it's in VR. It'll just make things better for those who are, and provide convenient, more effective assistance in certain situations. Making good into great is a hell of a lot easier than making bad into good.

So yeah, I can buy in to the potential of a genuinely revolutionary AR/VR "metaverse" that would break out of the current niche appeal of VR and be used routinely in regular office jobs; it needn't be a Utopian solution to all problems to still offer huge advantages. But just as the technical achievements will be bottlenecked by the weakest link in the chain (to mix metaphors), so too do the details of the interface matter. Which brings up the third and final point : does it have to be Facebook ?

Or perhaps it would be fairer to envisage the ways this could go wrong. The scope for privacy violations is huge. A bad interface could make organisation next to impossible, and transitioning from the familiar displays to something so wholly new won't be easy. There will be a lot of tasks that don't benefit from AR/VR at all. It could be overwhelmed with notifications, adverts.... to say nothing of the even greater potential for misinformation and manipulation. It won't bring about the dystopian real world that everyone cheerfully brushes aside in Ready Player One, but it could potentially have equivalently awful consequences.

There is no question to me that a VR metaverse is desirable. Equally, a metaverse could be awful. It doesn't follow that all forms of metaverse are wonderful any more than it follows that they're all abhorrent. They are neither. VR, to me, is a wonderful and enriching technology, but no purely technological safeguards can prevent misuse. I think it's utter cynical, knee-jerk reactionary garbage to claim Facebook shouldn't work on this because they could do something else instead, but equally, we've already seen what happens when we're cavalier about social media technology. If we don't learn the right lessons about regulating information, any new technology will only make things worse.