The web lost to apps, but it can come back with VR

Society loves to cycle back and forth. Growth begets recession; boom and bust. As software developers, we oscillate between putting logic in the client or on a server. We wax and wane between rigid languages that ensure correctness, and with forgiving languages that require less ceremony. Hell, we even go back and forth on tabs vs. spaces. Each time we collectively change our minds, we do things differently, making new and different mistakes that eventually send us back the other way again.

One of these cycles is native apps vs. the web, a tired debate that has been handily won by apps in the past decade. People don’t care much about the debate anymore and aren’t paying so much attention, but I see a confluence of events that will likely send this status quo off balance and launch the web back into the forefront, albeit not in the way you might think.

By becoming the only two stores where people can buy apps on any major smartphone, Apple and Google have become dominant arbiters of industry. They’ve become the sole channels through which companies can sell to their customers, and that’s an enviable position to be in! If you’re a company like Facebook, Amazon, or Microsoft — this has you scared shitless. You’ve now been upstreamed, and it’s a matter of time before you’re disintermediated or sidelined.

That’s when Mark Zuckerberg tried the Oculus Rift. He presumably foresaw that as the technology got better, more and more people will own VR devices, and that’s an opportunity to wedge open a third app store; a third arbiter of industry. The problem was, by acquiring Oculus, Facebook set off an arms race where everyone is rushing to do the same thing. The race is on to own the next app store.

However, when everyone silos; when you get Microsoft, Oculus, Valve, Google, Magic Leap, and everyone else trying to be the next app store in the zero-sum model of the previous app stores, it’s tiresome for users and developers alike. As a developer, you want to be able to easily tell someone where to buy your product, without first learning what ecosystems they are part of. As a user, you just want access to all the apps and games you care about, and to be able to easily find new good ones.

In the meantime, Chrome and Firefox haven’t been sitting still at all. They have been incorporating WebAssembly, which embeds an engine that can run an assembly language at essentially native performance. They’ve all worked together on WebVR, which gives developers low level access to the VR headset. And they’ve been adding low latency fast paths for VR devices, which unlocks frame rates well beyond the web’s old speed limit of 60fps. All that while continuing to improve WebGL’s support for the latest features offered by graphics cards.

When this all ships (for Chrome on Android, this will ship as early as October!), high quality WebVR will reach several million Google Cardboard and Daydream systems, all of the desktop VR systems including HTC Vive and Oculus Rift, and will support new devices as they arrive, like the Microsoft Hololens. And it’s not all out in the future — if you’ve got a GearVR, you can try WebVR today using Samsung’s own built-in browser.

All of a sudden, WebVR is the best channel for small-to-medium sized VR content to be published if it’s intended to go viral and be shared with the most VR users.

In addition to the developments in the browser, there has been considerable consolidation in how these new VR experiences are created — they’re mostly built with Unity or Unreal Engine. While it’s far from a write-once-run-anywhere mecca, these engines do allow for a single codebase to target multiple platforms without much fuss. Teams realistically do target Windows, iOS, and Android all with a single codebase.

What this means is that browser vendors can work with these small number of popular engines to create optimized WebAssembly output, so that exporting for the web is as easy as it is now to export for any other platform. While I’m sure there will be some healthy developer skepticism, once it’s apparent that performance is reasonable, developers will find this very appealing.

Why? First, you don’t have to worry about getting users to update their apps. Assuming the export process sets up proper caching and invalidation (admittedly a big assumption) users will always have the latest. This means it’s so much easier to do alpha or beta tests, without having to get permission, approval, or deal with limits on numbers of testers by the app stores. Second, you can have a direct relationship with your customers, which is good for a whole host of business reasons. Third, it means you can tell potential customers one single place to go for your game: your website. Which in turn makes it a lot easier to close any viral loops and help your users share your app or game.

In fact it doesn’t have to be your own website. It’s entirely possible that we’ll see portals pop up to make it easier to publish to the web, maybe with some services or community to create an ecosystem. The YouTube of VR is still waiting to be built, and I don’t mean 360° videos.

When this really starts to get interesting, though, is when these VR devices really shrink down and start to mix reality in. When they’re the size of glasses and people start to wear them out in the world. This is when some of our other devices like smartphones start to fade away, and iOS and Android app development becomes less important. Now we’re working with 3D, and apps are built with game engines. But they’re distributed via, and members of, a vibrant web ecosystem. The metaverse.

If I’m right about WebVR, here’s how these next 5 years could go:

Now— First wave of consumer VR devices launches, along with the first wave of early WebVR browsers.

2017 — Siloed app stores are still the way we all get games and apps. Browser vendors are still working on the basics of WebVR tech, but we see the dawn of the first WebVR portals. “VR is dead” articles are written.

2018 — The second wave of consumer VR hits, broadening the VR install base dramatically. Browser payment API VR integration makes indie web business viable. WebVR portals are still niche for small experiences, but are starting to have enough users to use data to optimize. “VR makes a comeback” articles are written.

2019 — Competition intensifies among the app stores and ecosystems, as hardware features start to converge. Siloed app stores fight each other heavily with exclusivity deals, but are still where most tend to go for content. WebVR portals are growing, pushing larger experiences, and getting cozier with game engine companies, while building communities and value added service moats. “Is the VR fad over?” articles are written.

2020 — Third wave of VR devices hits, which are now getting very good and cheap (meaning lots more users), and which start to really mix virtual and real. WebVR portals integrate closely with game engines — now you can export directly from the game engine to their platform. WebVR is now being taken seriously as a quality common denominator. “VR (AR?MR?XR?) is having its iPhone moment” articles are written.

2021—WebVR and the new wave of portals is now the defacto way we share 3D interactive experiences, especially apps. Siloed app stores are used to distribute the biggest games, somewhat like how EA’s Origin is used today. “Smartphones are dead” articles are written.