TU Delft pioneers Internet TV

Tribler, the most advanced p2p search engine available, and developed by TU Delft, is the technological basis of the project.

With the advent of broadband, people are spending more time surfing the Web, and in the process abandoning their one time best media friend – the TV. With millions of people now online, businesses are looking for the Holy Grail on the Internet. In their never-ending quest to reach and maintain customers, social networking sites and major movie and TV production companies alike are now seriously thinking about the next big thing: Internet TV.

TU Delft has joined forces with a diverse range of partners in a project that intends to fundamentally define the future of Internet TV as we know it today. The project, dubbed ‘P2P-Next’, is specifically focused on developing a Europe-wide ‘next generation’ Internet television distribution system, based on peer-to-peer (p2p) technology. The project has recently been awarded a €14 million research grant by the European Union.

Other major partners in the project include the BBC, the European Broadcasting Union, Lancaster University (UK), the film production company Markenfilm, Pioneer, and one of northern Europe’s largest technological research organizations: VTT.

Tribler, the technological footing of P2P-Next, is the current end result of the world’s largest university driven p2p research project. For the past few years, a group of researchers at TU Delft’s Faculty of Electrical Engineering, Mathematics & Computer Science have been working on the Tribler project. Tribler . an advanced p2p file sharing and search engine . helps the user find and share film, TV and music through a peer-to-peer network.

“Imagine millions of viewers all wanting to see the same football goal at the same moment in high quality on the Internet. The current Internet Protocol TV (IPTV) infrastructure is just not ready for that,” says Jacco Taal, valorization manager of the Tribler Project.

The current Internet infrastructure isn’t suited to simultaneously transmitting live events to millions of people, because such large numbers of users would congest the Internet. Taal: “P2P technology is scalable and able to tackle this problem in a very cost-effective manner.”

This is essentially because the Internet infrastructure is currently predominately based on methods (unicast, multicast) where the data transferring capacity is concentrated in a single or a small number of servers. But in a p2p network, each single computer – apart from being a host – also serves as a small server.

A p2p network is structured such that for each megabyte a computer downloads, it uploads a megabyte to another peer. In this way, each single computer, with its processing power and broadband connection, contributes to the total bandwidth available in the p2p-network.

“The insight that the Internet is not a simple mesh of computers and connections give us great opportunities to improve it. For instance, some computers are connected to the Internet-backbone, while others have a simple ADSL connection,” Taal explains. “This enables us to systemize the Internet power-house and use it to the benefit of faster streams.”

Tribler does exactly this. It automatically finds the best peers with the fastest connections for a particular download. Data is subsequently spread over the globe in such a way that slower peers can also benefit. Another advantage of a p2p network like Tribler is that it lacks a single point of failure.
Copyright

The P2P-Next project runs for four years. The project plans to conduct a large-scale technical trial of new media applications running on a large number of different consumer devices. Taal explains that these trials will occur somewhere midway in the project: “The kind of trial we’re thinking about is to use the technology to transmit a European or large national media event. But also to run it on consumer devices, such as cell phones and setup boxes.”

By eventually making the P2P technology applications possible in such devices, P2P-Next will set the stage for a new range of media services, unthought-of as yet today. Taal: “For instance, an eye witness to a major disaster could, by uploading his cell phone-made video file to a p2p-network, make it available to millions of people, almost instantly.”

The overall objective of P2P-Next is to induce the development of a wide range of new social content and Internet services; for instance, by allowing audiences to build their own virtual communities around their favorite content. The project has therefore chosen to share its results: all core software technology that is developed in the coming years will be made available as open source.

Taal says the added value of P2P-Next, compared to front-end sites like Youtube, is that P2P-Next will provide people extensive possibilities to express themselves: “People want to choose their own forum and post what they want. P2P-Next will give people the possibility to interact with and augment content that is provided by traditional broadcasters.”

But P2P-Next will also benefit industries. The advertising industry will benefit from the interactivity and flexibility that P2P-Next will enable. Taal: “Currently, web-based advertisement is not very effective. Nobody knows what future advertisements will look like, but many people think that the deployment of more interactive ads, which also provide viewers real added value, is the way ahead.”

P2P technology is, however, not new and has a somewhat dubious reputation, primarily because illegal file sharing sites, like Kazaa, are associated with the technology. Critics often rightfully refer to this, and P2P-Next therefore also specifically considers outstanding legal, regulatory, security, business and commercial issues. According to Taal, the copyright problem is ultimately a problem of institutional path dependence.

Current contracting and regulatory frameworks are built with traditional broadcasting and print standards in mind. They are not suited to the Internet, which is inherently less controllable. Taal: “To cope with this problem, new copyright models can be developed. You can, for instance, choose to grant rights that refer to the quality of the material, the period that the content is made available to viewers, or the number of times the viewer may watch the content.”

With the advent of broadband, people are spending more time surfing the Web, and in the process abandoning their one time best media friend – the TV. With millions of people now online, businesses are looking for the Holy Grail on the Internet. In their never-ending quest to reach and maintain customers, social networking sites and major movie and TV production companies alike are now seriously thinking about the next big thing: Internet TV.

TU Delft has joined forces with a diverse range of partners in a project that intends to fundamentally define the future of Internet TV as we know it today. The project, dubbed ‘P2P-Next’, is specifically focused on developing a Europe-wide ‘next generation’ Internet television distribution system, based on peer-to-peer (p2p) technology. The project has recently been awarded a €14 million research grant by the European Union.

Other major partners in the project include the BBC, the European Broadcasting Union, Lancaster University (UK), the film production company Markenfilm, Pioneer, and one of northern Europe’s largest technological research organizations: VTT.

Tribler, the technological footing of P2P-Next, is the current end result of the world’s largest university driven p2p research project. For the past few years, a group of researchers at TU Delft’s Faculty of Electrical Engineering, Mathematics & Computer Science have been working on the Tribler project. Tribler . an advanced p2p file sharing and search engine . helps the user find and share film, TV and music through a peer-to-peer network.

“Imagine millions of viewers all wanting to see the same football goal at the same moment in high quality on the Internet. The current Internet Protocol TV (IPTV) infrastructure is just not ready for that,” says Jacco Taal, valorization manager of the Tribler Project.

The current Internet infrastructure isn’t suited to simultaneously transmitting live events to millions of people, because such large numbers of users would congest the Internet. Taal: “P2P technology is scalable and able to tackle this problem in a very cost-effective manner.”

This is essentially because the Internet infrastructure is currently predominately based on methods (unicast, multicast) where the data transferring capacity is concentrated in a single or a small number of servers. But in a p2p network, each single computer – apart from being a host – also serves as a small server.

A p2p network is structured such that for each megabyte a computer downloads, it uploads a megabyte to another peer. In this way, each single computer, with its processing power and broadband connection, contributes to the total bandwidth available in the p2p-network.

“The insight that the Internet is not a simple mesh of computers and connections give us great opportunities to improve it. For instance, some computers are connected to the Internet-backbone, while others have a simple ADSL connection,” Taal explains. “This enables us to systemize the Internet power-house and use it to the benefit of faster streams.”

Tribler does exactly this. It automatically finds the best peers with the fastest connections for a particular download. Data is subsequently spread over the globe in such a way that slower peers can also benefit. Another advantage of a p2p network like Tribler is that it lacks a single point of failure.
Copyright

The P2P-Next project runs for four years. The project plans to conduct a large-scale technical trial of new media applications running on a large number of different consumer devices. Taal explains that these trials will occur somewhere midway in the project: “The kind of trial we’re thinking about is to use the technology to transmit a European or large national media event. But also to run it on consumer devices, such as cell phones and setup boxes.”

By eventually making the P2P technology applications possible in such devices, P2P-Next will set the stage for a new range of media services, unthought-of as yet today. Taal: “For instance, an eye witness to a major disaster could, by uploading his cell phone-made video file to a p2p-network, make it available to millions of people, almost instantly.”

The overall objective of P2P-Next is to induce the development of a wide range of new social content and Internet services; for instance, by allowing audiences to build their own virtual communities around their favorite content. The project has therefore chosen to share its results: all core software technology that is developed in the coming years will be made available as open source.

Taal says the added value of P2P-Next, compared to front-end sites like Youtube, is that P2P-Next will provide people extensive possibilities to express themselves: “People want to choose their own forum and post what they want. P2P-Next will give people the possibility to interact with and augment content that is provided by traditional broadcasters.”

But P2P-Next will also benefit industries. The advertising industry will benefit from the interactivity and flexibility that P2P-Next will enable. Taal: “Currently, web-based advertisement is not very effective. Nobody knows what future advertisements will look like, but many people think that the deployment of more interactive ads, which also provide viewers real added value, is the way ahead.”

P2P technology is, however, not new and has a somewhat dubious reputation, primarily because illegal file sharing sites, like Kazaa, are associated with the technology. Critics often rightfully refer to this, and P2P-Next therefore also specifically considers outstanding legal, regulatory, security, business and commercial issues. According to Taal, the copyright problem is ultimately a problem of institutional path dependence.

Current contracting and regulatory frameworks are built with traditional broadcasting and print standards in mind. They are not suited to the Internet, which is inherently less controllable. Taal: “To cope with this problem, new copyright models can be developed. You can, for instance, choose to grant rights that refer to the quality of the material, the period that the content is made available to viewers, or the number of times the viewer may watch the content.”