The immersive internet

Karan Singh
5 min readApr 5, 2017

--

A quarter of a century ago, I did a doctoral thesis on human figures in virtual reality (VR). It was 1994, and I was fortunate to have worked on the first virtual space tele-conferencing system at a futuristic lab, in rural Japan. The participants were decked out like deep-sea divers, immersed and tracked in a digital ether, where they interacted with proxies of themselves and artefacts, powered by room-sized computers.

At the time, even the strongest proponents of VR acknowledged that this was merely a vision executed as an engaging, expensive experiment. It seemed unimaginable that in the near future there would be the content, connectivity or computing power for it to be anything more. I graduated soon after, happily put away my VR goggles, and pursued a career in computer graphics and interaction, until 2013, when James McCrae came up with a curiously mind-bending prototype.

Back in the 90’s, web content was mostly interconnected text and images, with occasional video thrown in. 2D monitors and a format, the Document Object Model or DOM, naturally served up such content as linked documents. VRML was a representation for 3D aimed at the web, somewhat disparate from 2D, and perhaps before its time. The internet and VR share an interesting parallel in their evolution timelines. The resurgence of both every few decades could well be a complete coincidence, but one might argue that advances in computing, connectivity, display and sensing hardware, combined with the build-up of spatial content, have precipitated watersheds of development. As people increasingly engage with one another through phantasmagoric representations in virtual worlds, it may be time to rethink the fabulous, but near quarter century old internet browser design of Mosaic and its successors.

Simply pasting a current generation browser in front of your face in VR is not a good idea for three major reasons:

1. Long periods of focus on flat webpages designed for monitors and mice, are fatiguing in VR.

2. Windows with tabs and scrollbars unless spatialized, make poor use of VR’s expansive visual real estate.

3. Web content has evolved, and now finds itself unnaturally incarcerated within flat documents.

Social feeds today are peppered with panoramic images and 360 videos. Medical visits produce CT, MRI and other holographic data. Our watches and cars create elaborate GPS trails. And 3D animations are as common as live actors in our video content. This content cries out for us to blur the boundaries between our digital and physical realities, thus far the realm of science fiction.

Imagine instead, a webpage as an immersive space, and a link as a portal, a tear in space-time that you can magically open as needed, and walk right into another space. If you think of teleportation as a means to transport you from one space to another, a portal is a means to transport a space to your current context. Now imagine that as you walk through the internet, you can interact and collaborate with other people, simply by virtue of being at the same internet address. Imagine today’s online forums completely transformed into live conversations between strangers in virtual locations of common interest.

Re-imagining webpages as webspaces is not about 2D versus 3D. It is simply about enabling all content regardless of dimension to be presented optimally, and the ability to interact with this content with an interface that is as natural, or better than the way we interact with content in the physical world.

JanusVR: webpages as spaces, links as portals.

At JanusVR, we have spent the last three years chasing this vision!

The immersive internet to us, while being embedded in the three dimensions we perceive, is not about dimensionality at all, but the conception of a compelling interface to interact with multi-dimensional content. Therein lies the fundamental difference between our vision and that of those in pursuit of simply bringing 3D to the web.

Change, even radical change, must address the disruptive problems of compatibility and adoption. We don’t expect or want current browsers or 25 years of legacy web content to disappear. But we do champion a continuous transition from the documents of today to the spaces of tomorrow.

“Taking the internet of today into the internet of tomorrow”:

A large number of current webpages exist optimally in their current 2D layout, readily embedded as surfaces into webspaces. Armed with a semantic understanding of a webpage’s code however, it is not difficult to imagine image collections like pinterest.com and imgur.com being turned into art galleries, youtube.com and vimeo.com laid out as theater complexes, ikea.com and steelcase.com with their 3D models transformed to design showrooms, and geotagged tweets posted like stickies on a globe you can spin at will.

Once imagined, transforming these webpages spatially is a simple matter of deconstructing the existing content into its component assets and reassembling them. In other words, existing webpages can be recreated as meaningful VR spaces automatically or with minimal human guidance. Webpages embedded as surfaces are also palettes of content to interactively populate webspaces, allowing webpages to continuously evolve from documents to immersive spaces.

imgur.com in JanusVR
youtube.com in JanusVR

“Bringing the internet of tomorrow into the internet of today”:

There is a level of technological depth, hard won familiarity and user trust in current browsers, worthy of respect and complete inter-operability. Just as we are able to embed a webpage into a webspace, we should be able embed a webspace within a web document viewable in an existing browser.

Fortunately, current browsers allow us to do exactly this, through an evolving standard called WebVR, that we support using our browser within a browser, JanusWEB.

A webspace embedded in a webpage using JanusWEB

There is an embarrassing wealth of immersive content trapped unimaginatively in flatland. At JanusVR we are researching, building the representations, tools and interface to set it free amidst us. The future is undoubtedly moving towards an augmented world, where the URL pointing to bits of storage physically lodged deep in the bowels of a decentralized cloud, has physical meaning when set within an immersive internet of things. Beyond an augmented world, where reality can subtracted, abstracted and transcended, digital and physical reality can blur with dreams.

I hope to teleport there some day.

--

--

Karan Singh

Professor of Computer Science at University of Toronto, co-director of graphics and interaction lab DGP. CTO & co-founder of JanusVR.