Under the bonnet of the Internet

“Chrome Web Lab”, an interactive exhibit at the London Science Museum, presents the invisible fabric of the Internet as embodied in the world. Rory Hyde spoke to its creators, and discovers a version of where the Web may be headed next.

In Jacques Tati’s 1958 film Mon Oncle, Monsieur Hulot, played by Tati, visits his sister’s bright-white modern home. He’s baffled by the appliances, which pop and fizzle with cold electronic automation, and seems completely out of place with his hat, pipe and 19th-century demeanour. The house is ordered, rational, clean and — as the film more than implies — soulless. By contrast, Hulot’s own home in a ramshackle building in some old corner of Paris is brimming with personality. He makes his way up, ducking under laundry hung out to dry, past an old lady sweeping, through what seems to be a neighbour’s living room, to reach for his key trustingly left above the door.

Created at a time of great transformation in postwar France, these two scenes capture the distinction between the technological and the human; a distinction that has remained largely persistent in our imaginations. However, as technology is further embedded into all aspects of our lives today, the sharpness of this distinction is becoming increasingly blurred.

Matt Cottam, CEO of interaction design studio Tellart, tells a story that highlights this contemporary blurriness. His dog had gone missing in Connecticut while he was away; startled by gunshots, it had dashed off into the woods, leaving his family in a panic. Then, a few days later, Matt got a call on his mobile phone while on a tram in Amsterdam. It was the dog catcher, who had scanned the dog and found an RFID chip embedded in its neck, connecting it to a database of owners. As Matt explains, “I was already amazed this lady had figured out how to dial an international phone number, but it was just incredible that she could scan the dog’s neck, see an interface on him, where my phone number was registered, and all these service providers could connect us up to bring my dog back.”

View of the hall containing a series of Sketchbots for the sand portraits. Photo by Andrew Meredith

Here, Tati’s distinction between the cold and electronic, and the warm and human, is dissolved by a dog, which is possibly the furthest thing away from the glowing rectangles we stare at all day. In a few short decades, computing has become intimate and personal. The Internet has insinuated itself in the stuff of daily life completely, becoming indistinguishable from life itself. While this possibly sounds sinister, the experience of it is completely normal, unremarkable, banal even.

But of course, it is completely remarkable. So remarkable, and yet still so invisible, that it requires a physical exhibit in order to reveal and explain it.

Enter “Chrome Web Lab”, a large-scale public exhibition at the London Science Museum, integrated with an online platform. It features a series of “experiments” that each expose a different aspect of the inner workings of the Web, and explain the computational magic that keeps it all humming along. It’s the product of many people and many companies, foremost among them Tellart and Google Creative Lab.

Details of the physical installation of "Chrome Web Lab" at the London Science Museum. The aesthetic is much like the Google homepage: clean, white and simple, with bold highlight colours. Photo by Matt Cottam

Steve Vranakis, Creative Director of Google’s London Lab, explains, “I think all too often the Web is ubiquitous; it’s a more or less invisible part of our lives. It’s not that we necessarily take it for granted, but sometimes we forget what this thing actually is. So we wanted to lift up the veil and expose some of this stuff.” Or, as Cottam puts it, “If you push the Internet until it’s smoking, this is what it looks like.”

This smoke is generated by the five experiments that comprise the “Web Lab”, including robots that will draw your face in sand; a series of Web-connected instruments that can be played collaboratively; Teleporters that let you control a real-time view into another location; and a Data Tracer that shows you where an image is stored on the Web, and how it gets to you. All this is tied together with Lab Tags, unique computer-readable ID cards that keep track of your interactions.

Each of these experiments demonstrates a particular aspect of the Internet’s architecture — real-time collaboration, data compression, programming languages and databases, packet switching, machine intelligence — and, by pushing it really hard, the ability of the Google Chrome browser to handle it. Indeed, the mandate of Google’s Creative Lab, as Vranakis explains, is to “demonstrate by example the performance and possibilities of the platforms that Google produces”. So, ostensibly, the “Web Lab” is an elaborate advertisement for Google’s Chrome browser, the kind of everyday product the SuperNormal series addresses, but this aspect is thankfully downplayed. As Vranakis says, “This is a public space, a museum, a place where people are curious and eager to learn. The aim is to teach people about what you can do online, how it works, and bring that to life. Then perhaps we can inspire the next generation to become interested in computer science.”

Details of the physical installation of "Chrome Web Lab" at the London Science Museum. Photo by Andrew Meredith

What’s different, perhaps, is the way this “teaching” is articulated. Just like Cottam’s digitally augmented canine, the Internet is presented here not as something that is only peered at through a computer terminal. The “Web Lab” overturns the metaphors we’ve used to describe the Internet in the past — coral diagrams of links and nodes, the cascading green numbers of the Matrix, the global brain, the “cloud” — and instead presents it as embodied within physical objects in real space. It’s as if a corner of the Internet’s invisible fabric had suddenly revealed itself in the physical world.

The reason that this small corner looks and acts the way it does can largely be attributed to the designers at Tellart. Tellart sit in the ambiguous space where industrial design and computing overlap, which has yet to congeal into a fully autonomous discipline. Cottam first describes Tellart’s work by leaning on a whole raft of buzzy tech terms — interaction design, the Internet of things, ubiquitous computing, big data, smart cities and physical computing — but ultimately concludes that what they do is “simply 21st-century industrial design”. It’s an intentionally prosaic account, one that reaches back to the company’s roots in Providence, Rhode Island, where the can-do, sleeves-rolled-up, “New Englander” attitude has left its mark on their particular approach to design and making. They conceive of “data as just another material alongside wood or metal or glass”, adopting a hybrid and multidisciplinary approach to design.

The user interface for the Universal Orchestra. Photo by Matt Cottam

The Web Lab project began inauspiciously in 2009, when Tellart were approached by an ad agency on behalf of Google to produce a memorable interactive installation that might compel one to switch browsers. A few months down this track, the project was on the brink of being cancelled due their agency client having a conflict of interest. But instead of letting this opportunity slip away, the Tellart team with key members of the Creative Lab and Chrome Marketing devised an entirely new proposition. This concept was captured by Tellart in what they simply refer to as The White Film.

Using only simple white card models, stop-frame animation and projected coloured light, The White Film managed to capture in a number of minutes the essence of what would become the “Web Lab” exhibition. It’s a terrific piece of “unsolicited” filmmaking, a pro-active pitch that pulled the project back from the edge, and ratcheted it up from six months into a two-year endeavour.

As an aside, there’s a lesson here for designers on how to get a great commission. Great projects are rarely the result of a phone call out of the blue from a wealthy client with a perfectly formed brief, but require strategic vision and risky tactics to shape the job you have into the job you want.

Sketching the user interface for the Universal Orchestra. The blobs represent notes to be played by the robotic instruments. When a note is changed, it stretches into place. The time it takes for the blob to move to its new position depends on how long it takes to communicate the change with the computers in the museum. Photo by Jasper Speicher

The White Film also established the core experience concept for the “real” exhibit that would be realised two years later. To get to this point, a number of other partners were brought in for the final stages to design and build the experience with Tellart and Google Creative Lab. While it was a thoroughly collaborative process, the various roles can be roughly attributed as follows: Universal Design Studio/MAP led the exhibition design and collaborated with Tellart on the industrial design, B-Reel developed the branding and designed and built the website, graphic designers Bibliothèque worked on the visual identity of the in-space signage, and programming wizard Karsten Schmidt collaborated with Tellart on the design and coding of the Lab Tags and Lab Tag Explorer.

The completed result is a far cry from the Tron aesthetic we’ve come to expect when presenting technology. Instead the palette is simply comprised of a white background and primary colours; clean, tidy and unintimidating, perhaps like an inhabitable version of the clean and white Google homepage. The experiments are built of white powder-coated open steel frames, sitting on solid grey plinths, with bright-yellow wiring and cable trays tracing through the space to reinforce the connection to the Web. All is set off by the highlight colours of the Lab Tag graphics and robotic instruments. The effect is “like a playground with kids running around”, Vranakis says. “The structures even look like monkey bars, because sometimes technology can be a bit cold and scary.”

Calibrating the Lab Tag Explorer, a two-axis, computer-controlled whiteboard marker, which shows where online visitors are in the world. Photo by Seth Snyder

The other “space” of the “Web Lab” is the online experience, as all of the experiments can be explored through the associated website. More than just a virtual simulacrum, this is an integrated synthesis where the physical and the digital depend upon one another. “If you take away one, the other goes silent,” says Cottam. In an impressive exercise in consistency, there is little to distinguish the online experience from actually being there. The interfaces to control the experiments online are the same as the touch screens one encounters in the museum, and you can even watch your “performance” in real time via an array of cameras.  

In this sense, the “Web Lab” represents a form of hybrid space, designed to be experienced both in the flesh and as mediated via the Web, and presents a particular challenge to how we have traditionally approached design. It collapses Finnish architect Eliel Saarinen’s neatly nested scales-of-design operation; no longer can this space be “considered in its next largest context — a chair in a room, a room in a house, a house in an environment”, etc. Instead it leaps from an interior in London to the global scale of Google’s network, as a digital portal peered into at any time from any place. And yet this potentially jarring shift in scale feels seamless, due to the consistency and coherence of the digital and physical design in the “Web Lab”.

The massive robotic components of the exhibition were fabricated and tested in Providence, Rhode Island, then packed and shipped to London. Photo Seth Snyder

However, it is precisely the success of this online-offline integration that has posed one of the more thorny challenges to the “Web Lab” team. As Vranakis says, “The only thing that we continue to refine is the believability. People just don’t believe that when you are playing an instrument that there is someone online in a different country playing right next to you.” Even for Cottam, who perhaps knows what’s happening more than anyone else, there remains a sense of wonder. “Going past all these instruments and seeing ‘Vietnam’, ‘Australia’, ‘California’, and realising that no matter how many times in your life you connect objects with the Web, it’s still magical to know that people are sitting in their houses, or in their offices, right then, looking through cameras at you and playing the instrument.”

This word “magic” comes up again and again with those I speak to about the exhibit. Indeed, “The Magic of the Modern Web” was an early slogan for the show. But of course the Web isn’t magic at all. It’s about the furthest thing from it; it’s science, technology, design, business and a lot of really hard work. The slogan was dropped, and instead efforts were made to reveal the invisible reality of what’s happening and to communicate it clearly. This occurs on the literal level of explanatory videos accompanying each of the experiments, but also more subtly, such as by leaving the status bar visible on the touch-screen interface in the museum, demonstrating that it’s all being run from inside a browser. Part of this is presumably marketing, to allow Google to announce proudly that “it all runs on Chrome!” But it also speaks to the reality we occupy, and what’s possible today with the incredibly powerful tools most of us can access through our computers.

Details of the physical installation of "Chrome Web Lab" at the London Science Museum. Photo by Matt Cottam

And that’s the point of it all really — to show what’s possible today. The experiments themselves are pointless in a way; I can think of few practical uses for a line drawing of your face in sand. The sand-drawing experiment shows that a computer can, in a matter of minutes, take your picture, recognise your face, crop it, adjust the levels, apply nodes to the edges, create a vector pattern, convert that information to code, send that to a robotic arm in a museum in London which draws it as lines in sand, have it all filmed in HD, upload it to YouTube, and send you the link. This is impressive to be sure, but the fact that I don’t need the aid of a robot to draw a face in the sand only seems to emphasise this pointlessness. I can’t help but think that if all this whiz-bangery were applied to something of more substance, perhaps it might do more than just inspire the next generation of computer scientists, but inspire them to do something useful.

But why sand? As an exhibition that can be “visited” from anywhere in the world, the attendance figures are impressive. In the five months since opening, the “Web Lab” has attracted 200,000 visitors to the museum, and an additional 4.3 million visitors online. This is a physical exhibition operating on the virtual scale of Google, demanding particular strategies for coping with these numbers. Where the Google approach to scaling up capacity is simply to throw more servers at a problem, to accommodate this scale of data in the real world requires a clever idea. Drawing all of these faces on paper would be a terrific waste — 9,000 reams of A4 by my maths — whereas the sand is simply wiped flat, ready to be reused for the next drawing.

Online interface for the Universal Orchestra and Teleporter experiments. Image courtesy of B-Reel

Wiping the sand flat is also a neat metaphor for “forgetting”, something that Google is increasingly sensitive about as it tries to preserve public trust in how it uses the massive amounts of personal information the company holds on many of us. Despite this, the “Web Lab” team were still keen to enable visitors to be able to keep track of the things they create. The normal way to do this would be to require some kind of log in, with an email address or a social network. But as a large proportion of the audience are school children, too young to be compelled to sign up to Google+ for instance, this option was off the table. The solution is the Lab Tag, a unique and anonymous identifier composed of two parts: one human-readable, what Cottam refers to as “hieroglyphic”, and the other that can be read by computer vision, like a QR code. What the visitors produce is recorded and associated with that tag. At home the tag can be read by a webcam, pulling up everything you’ve made in your visit. It’s a brilliantly simple bit of design, deftly sidestepping issues of privacy and creating an opportunity out of a constraint.

Online interface for the Universal Orchestra and Teleporter experiments. Image courtesy of B-Reel

Similarly, the idea of using music was also driven by the constraints of the audience and privacy. As Cottam explains, “We wanted people to be really social with each other and collaborate in this learning process, but we couldn’t just let people online talk to kids in the museum, because there’s a big risk that people will act inappropriately, and one of the great ways that people can interact non-verbally is through making music.” Indeed, the Universal Orchestra — both online and offline — is a real highlight of the exhibition. The idea that a handful of people from all over the world can come together online and produce music, with no pre-defined idea about what it ought to sound like, is just fantastic. But what’s even more remarkable is that it actually sounds good.  

Not all the experiments are as successful, however. Despite their interactivity, the Teleporters — swivelling periscope-like viewfinders that allow a peek through a Web-connected camera positioned elsewhere in the world — felt familiar and pedestrian, like Skype or a webcam.  

Similarly, the Data Tracer — which shows on a world map where an image is stored online, and how long it takes to retrieve it—didn’t seem to push the technology sufficiently or produce much insight. These are small quibbles, and perhaps I’m harder to impress than the target audience of school-age children. But then again, when it comes to the Internet, it’s likely they know more than I do.

Online interface for the Universal Orchestra and Teleporter experiments. Image courtesy of B-Reel

The Web Lab is at once recognisable as a physical manifestation of the Internet in the real world, and yet also something unique and distinct. It’s projective in that sense, as it helps us to imagine what the Internet could be, representing a more optimistic and creative expression of where our digital lives may be headed.

The challenge remains, however, for this and other endeavours like it, to overcome the “believability issue”, to convey the technology of the Internet as science, not magic. The elegant execution, design consistency and online-offline coherence of the “Web Lab” is laudable, but, like a magician’s sleight of hand, it conceals the mechanics of the trick underneath. This is a compliment to be sure, but I can’t help but think that with a few exposed “seams”, and a little a more purpose and openness, it might be easier to imagine how this technology could be applied in other circumstances.

A conceptual sketch of the Web Lab project by Matt Cottam. © Matt Cottam

But walking around the “Web Lab”, one can still feel a little like Monsieur Hulot in his sister’s modern kitchen — intrigued, impressed, a bit confused, and yet ultimately aware that all these seemingly useless novelties will soon change the world. Rory Hyde (@roryhyde), architect, researcher and broadcaster