Eric Romo and his cofounders at AltspaceVR want to make virtual reality as pervasive and accessible as the worldwide web. AltspaceVR is leading the charge to create a social VR experience, where people do in VR pretty much everything they do in real life.
Founded in 2013, the Redwood City, Calif.-based company now has more than 40 people. And it has raised $16 million since 2016. The crowds were thin at first, but on a given weekend night, you’ll now find parties with hundreds of people attending them. It’s now used in 150 countries.
I’m not sure what all this means, and why social events will be more fun in VR. So I caught up with Romo at the recent SVVR event in San Jose, Calif. Here’s an edited transcript of our conversation.
GB: How have you evolved over time?
Eric Romo: I started the company in 2013. The first year was me in a room, in my home office, tinkering with stuff. We went into closed beta by the end of 2014. We were showing people the product a couple of days at a time. And then in 2015 we opened up full time. At the time the only headset anyone had was the Rift DK2. Since then we’ve expanded to all the rest of the platforms. We’re on Rift and Vive and Gear VR and Daydream now as well.
Today we’re showing the advances we’ve made continually on our SDK. The first time we talked about that was probably September 2015. We’ve always talked about Altspace as this intermingling of game and web technology. We’re built on top of Unity, and we have a version of Chromium, the open-source version of Google’s Chrome, built into the product. Things you see and do in Altspace, often you’ll think—because of the way more VR applications are built, you would think of them as game things built in a game engine. But in reality they’re websites, effectively.
A way to think about this, all the things you’ll see in the trailer over there, or if you get a chance to try the demo station—everything you experience is literally a website. It’s written in A-Frame, which is a markup language like HTML. The objects and things in the room, in space, are pulled down in real time from servers, just like images on a website in your normal browser.
We’ve created this bridge between the web and things that game engines are good at. Game engines are good at high framerate rendering, good at integrating controller interactions, good at physics. We’ve given developers who use our SDK the ability to use the elements of a game engine when they write what’s effectively a VR website. We’re trying to create the best of both worlds in the VR web experience.
GB: What are some of the most interesting things people are creating or finding a way to do?
Romo: Most of the things they’re showing in this trailer are literally just the newest ones this week, which is a lot of fun. The things most users experience in Altspace, what developers have created, are things like party apps. Great places for people to hang out, listen to music, play card games, and interact. On an average weekend we’ll have multiple parties with hundreds of people showing up in Altspace.
Somebody built a boxing gym, where you get gloves on your hands and people are dancing around the ring. There’s a climbing application, so you can climb around the virtual space. The one before that was a miniature version of the room you’re in as part of the space, almost like a dollhouse, and if you put your hand in that miniature version, a giant hand appears in the room with you.
People are experimenting with the functionality of this bridge between the web and a game engine. We’re excited about just giving people this canvas. From a developer’s perspective, some things are very challenging about VR. Being cross-platform is tough to do. Unity helps make that easier, but if you survey the market and look at what apps are on all four major platforms, there are very few. We tell developers, “If you develop something once to work in AltspaceVR, using A-Frame or our SDK, it’ll work on every platform.” Developers can build one thing and obviate the need to figure out the store situation. They just go directly to all users on all platforms. That’s a valuable thing.
GB: How many people do you have working for you now? Did you raise money recently?
Romo: We have about 40 people at the company. The last time we raised money was about two years ago. We’ve raised about $16 million so far.
GB: How do you keep advancing this and stay out of the way of the likes of Facebook?
Romo: As we set out our focus was, we don’t know how long the market’s going to take to evolve. Mark Zuckerberg talks about this as a five to 10 year plan, right? While we wait for the market to catch up, we need to build integral pieces of technology. Our SDK is a great example of that. It’s core to the way we think the VR web will expand, this marriage between game engines and web technology. We’ve built really hard things that we think are valuable and useful irrespective of when the market arrives. What’s exciting to us is you just have tons of innovation in the space, including from Facebook.
GB: Is there some kind of standard? Does WebVR come into this in some way?
Romo: It’s related. All the stuff here is built with A-Frame, which is a cousin or brother or something like that to WebVR. WebVR is literally the standards that make VR things happen in a web browser. It allows you to use your HTC Vive with Google Chrome, for example. But what we’re building on top of is all the power of the 3D web. This is things like WebGL, Three.js, and libraries for people who build 3D things that now we’re allowing to render themselves in a virtual space, in Altspace.
With A-Frame we give developers a markup language that’s relatively simple and human-readable, like HTML, whereby they can then lay out the environment people go into and the things they do there. That code right there is the code for the space where all these things are happening on the screen. All the signs and the objects and the pictures, all that’s written in markup code that fits on that one sheet.
We think it’s important that the experiences you have together in virtual spaces—lots of people are going to want to build those, just like lots of people build websites. In order for that to happen it needs to be accessible, and A-Frame is something that helps make it accessible. At the same time, it needs to have the power of a game engine. The things that browsers don’t do well, game engines do. That’s the bridge we create.
GB: On the PC we’re seeing things like Linden Lab’s Sansar, which looks very nice with the tools they’ve provided. Do you want to get to that kind of visual quality?
Romo: The great thing about the approach we’re taking is we’re embracing the web. We see this as, just like the web started as a file transfer protocol used for document viewing and eventually progressed to images and video, we’re seeing VR as part of that evolution of the web. We’re focused on building that VR layer on top of the web. There’s really no limit to what the quality can be.
What you see in things like Sansar, you’re basically saying, “This will work on a small subset of machines that can render this type of imagery.” Our approach has been to hit a broader swath of the market – things that work on mobile, on Daydream and Gear VR. It’s a difference in focus. Should we hit the top end of the market or a broader spectrum? But there’s no particular reason why a developer couldn’t create something that looked similar to that high-end stuff. It would just only work on a few machines.
GB: Who else seems like they’re in a related space? Is Roblox somewhat similar?
Romo: I suppose so. The difference is, they’re embracing user-generated content, a democratization where lots of people will be creating content. We think that’s a way to go. We’re focusing on the professional creator market a bit more, people who are going to write code to create these experiences versus something where I’ll go into a space and work with my hands to create an object in that space. There’s a place for that, certainly, but we’ve been more focused on people who build websites, people who build games. Those are the folks we’re targeting with the SDK.
GB: How are you monetizing this?
Romo: We’ve been focused on getting the quality of the experience to a place where we could consider charging people for things. The last thing we want to do is have an experience that somebody pays for and they think it’s not worth it. We have a big focus around these live events in Altspace you might have seen. We did the partnership with NBC around the election. We’ve done shows with Reggie Watts where we have thousands of people showing up. The average users spends almost an hour in some of these events. There aren’t many pieces of content in VR people will spend an hour with.
There are avenues to monetizing to that content, but the reality is that our best events are thousands of people. If you go to Twitch right now, you’d have to go down to the hundredth most popular stream on the League of Legends channel alone to get to a thousand people. It’s early days for thinking about monetization. What we see as a trigger is if the engagement is there. If I can spend a user to spend an hour in something, there’s eventually a way to transact with that user.
GB: Does your traffic generally match the popularity of all the different platforms? Is Gear VR first?
Romo: We’ve remained relatively steady at about half Gear VR, yeah. Or half mobile, including Daydream, and then half Rift and Vive. We find the same thing a lot of people find. Session times on mobile are shorter. Average session time on Rift or Vive might be 40 or 45 minutes. Average session time on Gear VR or Daydream might be about 20 minutes. They stay for less time, which makes sense, because of the lack of controllers, the lack of tracking. They can’t move around as much. Users on mobile are a bit less sticky because there are so many of them. They’re trying to figure out how to use this free thing they’ve got.
GB: Are you more hopeful for the PC or for mobile?
Romo: Personally, at least, I’m a big believer in the direction mobile is headed, but not necessarily in mobile as it’s currently incarnated. It’s hard for me to imagine my aunt or uncle getting a PC and a Rift or a Vive and setting it up. Not that they’re not technical, but the user experience is challenging if you’re not a gamer. At the same time, only a subset of people have the right phones for mobile. Mobile lacks things like head tracking that are really prerequisites for a great VR experience.
There’s an interesting direction emerging in the stand-alone headset market. I have a thesis that that’s what will win commercially in the space, eventually. Things like Qualcomm’s VR 835 that they showed at GDC, the Intel Alloy project, the Facebook Santa Cruz project, these are things that you imagine would have the price of a tablet, roughly. You buy one box, take it out of the box, turn it on, and it works. From a consumer perspective, that’s much easier than trying to figure out how to set up a computer and so on. It’s obviously early days in the entire VR market, but that’s our thesis.