Let’s forget business models and monetization -- just for a brief moment. Instead, we’ll focus on one key issue: the interface, the way you access, browse, spot, save relevant information. The interface is pivotal. A good one will allow you to rope in your readers / viewers, and make them loyal to your brand, your contents. Pouring money and resources into an editorial effort, striving to get the best out of your team, buying the best contributions, pictures, multimedia features available… All of this is pointless without an effective interface. With this in mind, let's see what's lies ahead of us in the interface world.
Last week, I spent a couple of days at Microsoft’s Redmond campus, near Seattle. I was part of a small group of foreign journalists given access to Microsoft Research scientists and top engineers. Every year, in Redmond, they hold their Techfest reunion, a private, by invitation only gathering. This is the venue where they show off their work and exchange their findings.
In a way, Microsoft Research functions more like a university where a group of 900 PhDs is encouraged to publish in science journals or to speak at conferences. These are not product people, they’re more like scholars in disciplines such as oceanography or molecular biology. Those fields can be quite far away from Microsoft core business – even though the proportion of hard core computer scientists is significant. Engaging such people in discussion is an exhilarating experience. I'll come back to it in an upcoming Monday Note.
Of my many meetings, a notable one entailed a visit to the Microsoft Live Labs. The group aggregates about 80 people, two thirds of them engineers, on the 12th floor of a building in Bellevue, ten minutes away from the Microsoft campus. The Live Labs are a kind of intermediate layer between research people and product teams. They focus on transformative web experiences (read their manifesto, here). Their work stems from three technologies : Photosynth, which allows the user to stitch digital photos into 3D models; Pivot, a stunning way to organize large collections of data (to get an idea, watch this excellent TED's talk by Gary Flake, Live Labs founder). The third pillar is Seadragon a technology acquired by Microsoft in 2006 and refined into an actual product now integrated into some Microsoft services, something anyone can play with.
The most spectacular integration of Photosynth and Seadragon can be seen on the latest version of Microsoft's Bing Maps (if you connect from the US). As I visited with the Bing Maps group in Seattle, they showed me its newest features: a mash-up with the huge Flickr digital photo library and, even more spectacular, the prospect of integrating live video into the navigation experience. Go to this newly released TED presentation by Blaise Aguera y Arcas, mind blowing.
Let's go back to our topic du jour: new ways to navigate news contents. Note that I'm merely discussing a browsing experience here — exactly as you do when you flip through the pages of a publication, in a random, non sequential fashion, which is actually the best way to graze our daily information fix.
Seadragon is based on a simple concept: infinite zooming. To jump from one element to the next, instead of navigating through links and pages, you zoom in and out. To grasp the power of Seadragon, just look at the image below.
This is called a gigapixel image. While your digital camera typically captures a 10 million pixels image, this one is 2.6 billion pixels big, 260 times larger that the one you'd shoot staying at the same Sierra Nevada vantage point. Translated into the physical world, obtaining such resolution (i.e. being able to see the white Jeep) would require a 25 meters wide image.
This is what Seadragon is about: it lets you dive in an image down to the smallest detail. All done seamlessly using the internet. The Seadragon deep-zooming system achieves such fluidity by sending requests to a database of "tiles", each one holding a fraction of the total image. The required tiles load as we zoom and pan. And because each request is of a modest size, it only needs to cover a fraction of our screen, the process works fine with a basic internet connection. You can actually try on this map + photos of the Yosemite National Park.
To understand what it means for media, Bill Crow, manager of Live Labs group and Beatriz Diaz Acosta, senior engineer, showed me what lies ahead. In a prototype, they used a set of 6400 pages of the final editions of the Seattle Post Intelligencer, the local daily that folded few months ago.
Let's picture this: a one year of a daily newspaper entirely shown on one screen. 365 days x 50 pages of newspaper on average, that is about 17 800 pages to navigate. At first, this collection is represented using a series of thumbnails that are too small to be identified.
One click breaks up the stack by month, another click organizes it in a much more manageable set of weeks. Now, I pick up an issue and dive in. The only tool I'll use is my mouse’s scroll wheel, or whatever system allowing me to go deeper. It could be a trackpad, it could also be the stunning Microsoft Surface interface, a sort of coffee table covered by a large 30 or 40 inches glass top allowing multi-touch manipulations.
Unlike the hyperlink system I use when going from one page to another, in the Seadragon-based interface I'm not leaving my "newspaper". I'm staying inside the same zoomable set of elements. As I land on a page of interest, again, I can zoom in to a particular story (which, in passing, reconstructs itself in order to avoid the “old-style” jump to the article’s continuation on another page).
Back to monetization and business models: A key byproduct of this innovative browsing experience is its ability to reinvent the online advertising. As I mentioned in a previous Monday Note (see The Web Design Problem), online advertising suffers from an inherent flaw: its main purpose is to take the reader away from the content page. Imagine a TV commercial forcing you to switch channels to see the ads. Seadragon’s resolution yields the ability to zoom in down to the fine print of an ad. From there, the same add is blown up to the size of a billboard. This breeds really new ways to advertise in the Web. As Bill Crow takes me through the navigation experience, the endless zooming can be used to display more layers of information such as rates or detailed offers that become discernible only if you zoom deep enough. See this example of the Yosemite map, with the enlargement of the box in the lower right corner of the map.
Microsoft's Live Labs Bill Crow emphasized the demo is not a product, it's a prototype. But it relies on technologies already available and implemented elsewhere. And, in an indisputable way, it shows how much room for improvement the internet browsing experience still offers. The web as we know it? It’s just the beginning.