By Rob Knies, Managing Editor, Microsoft Research
Have you seen the one about …?
From time immemorial, mankind has maximized its ability to share information orally. Whether it is passing along a recounting of a movie, recapping last night’s sporting event, or simply telling a joke, that tradition continues today. It’s easy, it’s expedient, it’s socially engaging.
Spotlight: Blog post
Eureka: Evaluating and understanding progress in AI
How can we rigorously evaluate and understand state-of-the-art progress in AI? Eureka is an open-source framework for standardizing evaluations of large foundation models, beyond single-score reporting and rankings. Learn more about the extended findings.
But, as we all know, there are drawbacks to this sort of information exchange. We might not be as good at explaining a movie as enjoying one. Perhaps, in the retelling, we forget a key play in the game. Maybe our joke-telling skills need a bit of polish.
Maybe zCast can help.
zCast, a research project out of Microsoft Research’s lab in Cambridge, England, is exploring the possibilities of using datacasting—sending various associated files along with broadcast signals—to provide services to mobile devices. Tim Regan, a research software-design engineer with the Socio-Digital Systems group within that lab, is fascinated by the potential such a technique could offer.
“zCast is looking at how to get media content to handheld devices,” he says. “What kind of media would people want on the move on their mobile phones? When and why? How do we get it to them?”
The answers to those sorts of big questions could, before long, enable you to share a movie trailer, send a game clip, or e-mail a late-night talk-show anecdote to friends or colleagues as easily as if you were telling them about it—without having to burnish your skills as a film critic, a play-by-play announcer, or a professional quipster.
The latter scenario is one that, Regan’s research has revealed, could have particular social value.
“People will download a funny video clip to their mobile device,” he says, “and then, when they meet their friends in a bar later on that night, they’ll say, ‘Have you seen this?’
“What they’re doing is they’re demonstrating that they’re slightly ahead of the curve, because they’ve seen the content that you haven’t, and that they’re socially cool, because they know what will make you laugh. And then they can give it to you. I can Bluetooth you a copy of the joke. That kind of use you don’t get on standard TVs.”
The datacasting concept, of course, is far broader than a joke-delivery mechanism.
“There are two sides of the research interest,” Regan says. “There is the human-computer interaction side: What kinds of user experiences does this enable over different networks? And there’s the underlying network side: If you do have different ways of getting stuff to devices, how do you manage those different channels, and how do you balance across them?”
Such is the interest in zCast that the project actually turned the usual research model on its ear.
“One of the things that’s really intriguing about the project is it breaks the model of what usually happens in research,” Regan says. “My experience with industrial research is you have a cool idea, you research it in some depth, and then you go to the product groups.
“This was very different. Product groups, including eHome, Windows Media™, and devices, were looking to do a project in this area. It sounded so interesting we got involved. It was really kind of product pull, which was very nice.”
The zCast project began in 2003, when Regan and colleagues received permission to use a portion of the digital radio spectrum in the United Kingdom to experiment with datacasting.
“The kind of applications we were thinking about,” Regan recalls, “were additions to radio. You might be listening to a radio program, and you could get additional information without having to make a point-to-point connection across the Web, because you’re browsing files that have been downloaded to you already.”
In the process, the researchers became fascinated with what they termed an “asymmetric user experience,” in which lots of information is pushed to a user, but little or none is sent by a user.
“The archetypes of that are things like TV, where it’s all one way,” Regan explains, “or things like buying or voting, where you need a lot of information to make up your mind, but once you’ve made up your mind, you need to pass very little information back the other way.”
The model could work in one of two ways. The information could be streamed so that everyone receives it at the same time, such as TV, or it could be delivered via a carousel: A number of files are downloaded, and a user could check the carousel for a desired file.
“That’s what makes it interesting,” Regan says. “The market hype would be that everyone wants this all the time, whereas more realistic people might say, ‘No, that’s not true; there are some times when it might be useful and plenty of times when it isn’t.’ From a researcher’s perspective, that’s what I wanted to understand in some depth: When is it that things like this would be useful?”
He learned that video on a mobile device could prove a common usage.
“There were a bunch of scenarios where the way people will want to consume video on a mobile device is analogous to watching TV at home,” he says. “If you want to catch the next episode of Lost, and, when it’s on, you happen to be on a train, it would be really nice if you could just watch it.”
The project is now in its third phase. The first included platform-level work done by Microsoft’s eHome team, which was interested in the potential of datacasting.
“We did some work to determine the key scenarios to show to people to make them understand what this is all about,” Regan says. “One was the augmented radio service, where you have the ability to vote on the track you were listening to and move it up the playlist of the radio station—or buy the track.”
Other first-phase work included designing a movie-download service and an enhanced electronic program guide.
The second phase concentrated on showing the technology to governmental and industry leaders and educating them about the innovative potential of alternative uses of the broadcast spectrum.
“It’s important,” Regan says, “that regulators realize that the spectrum usages they’ve undertook in the past are not the ones that will be useful in the future, not necessarily.”
Apparently, the effort paid off.
“That was phenomenally successful,” Regan says. The broadcast industry, in particular, has been impressed by the project’s potential.
“A lot of them are excited about the mobile space,” Regan says, “so they’d love to get a deeper handle on how their content and their model of the consumer reaches beyond their current platforms out into the mobile.”
Indeed, a leading communications provider has tested the technology within London’s M25 motorway and is planning to launch a mobile broadcast entertainment service based on zCast later this year. The service, which broadcasts using the digital-audio broadcasting network, is navigated through an electronic program guide on a mobile phone. Virgin Mobile has agreed to act as the mobile-operator partner for the effort.
Meanwhile, phase-three work continues to refine the technology. The U.K. digital-TV standards body has devised an alternative datacasting standard, and the two will be compared in a Cambridge network. Scenarios are being tested among users, such as ways to enhance events.
“It may be that when you’re at a baseball game, there’s related content being delivered to your mobile device,” Regan says. “But it doesn’t have to be at a live event. It could be a fun contest associated with a momentous TV program that we’re all going to watch, or it could be a soccer game that we all choose to view on a big screen at a bar. There are a number of those social settings where content to enhance a shared event would be really nice to understand.”
The potential for such enhanced, shared experiences are what Regan relishes about the project.
“I’d like to see the use of media—the way I watch a bit of TV or listen to a bit of music—made a part of my social experience, made much more easy for me to share that with my friends, made part of the barter and banter between us.”
And, he says, it’s gratifying to see his work start to take off.
“We’ve been able to influence not just industry, but the fabric of European regulation,” Regan says. “Being able to do that and, at the same time, having written the code that does it … There are not many companies where you get to do all of that. I love that.”
Graphics and multimedia 
Search and information retrieval