There seems to be an obsession with bandwidth in the IPTV world.
Of course, if you're unicasting (ie sending an individual stream to an individual viewer), then there is a cost of delivery both at the source of the stream and the delivery end. Increasingly, there may also be carriage charges for carrying the packets of data between these two networks over peering points.
With the demise of Net Neutrality the existing system will be thrown into commercial chaos.
There are a couple of technologies that get around this problem: multicasting is similar to traditional broadcasting, whereby the same picture is sent to all viewers on a network. However, this looses some of the advantages of IPTV such as the ability to target advertising to individual viewers or to produce viewer generated channels.
Another alternative is peer-to-peer networks, where there are hundreds of competing products allowing users on a network to use other users' bandwidth when available. Probably the best known is Kontiki, now part of Verisign. It's a little like Bit Torrent - you get various bits of your bandwidth from different providers. I've been evaluating these technologies over the past few months and the results are patchy. Again, the technology doesn't work as well with live, scheduled or dynamically delivered content. Another problem is security (despite claims made by the vendors, you're opening a port on your computer to unknown data and offering it as a service to third parties); there is also the fact that a user may be paying for another user's bandwidth use if they're on a metered plan.
Another problem is asynchronicity - most broadband customers have the capability to receive far more data than they can send out (hence Asynchronous Digital Subscriber Line - ADSL). This also hits products like the Slingbox, where you make use of the data coming out of your home.
In the US, where competition is intense and major corporations such as Verizon and AT&T are rolling out fibre in the major urban areas it's easy to pick up good synchronous providers at very high data rates.
With this rollout comes a hybrid solution to the bandwidth problem - that service providers manage bandwidth within their own networks and redistribute it from one customer to another - this overcomes some of the security and performance issues of existing peer-to-peer services. With Net Neutrality teetering, this would be one genuine and defensible way of offering added value within a specific service provider, and possibly, dare I say, charging for it.
Of course, if you're unicasting (ie sending an individual stream to an individual viewer), then there is a cost of delivery both at the source of the stream and the delivery end. Increasingly, there may also be carriage charges for carrying the packets of data between these two networks over peering points.
With the demise of Net Neutrality the existing system will be thrown into commercial chaos.
There are a couple of technologies that get around this problem: multicasting is similar to traditional broadcasting, whereby the same picture is sent to all viewers on a network. However, this looses some of the advantages of IPTV such as the ability to target advertising to individual viewers or to produce viewer generated channels.
Another alternative is peer-to-peer networks, where there are hundreds of competing products allowing users on a network to use other users' bandwidth when available. Probably the best known is Kontiki, now part of Verisign. It's a little like Bit Torrent - you get various bits of your bandwidth from different providers. I've been evaluating these technologies over the past few months and the results are patchy. Again, the technology doesn't work as well with live, scheduled or dynamically delivered content. Another problem is security (despite claims made by the vendors, you're opening a port on your computer to unknown data and offering it as a service to third parties); there is also the fact that a user may be paying for another user's bandwidth use if they're on a metered plan.
Another problem is asynchronicity - most broadband customers have the capability to receive far more data than they can send out (hence Asynchronous Digital Subscriber Line - ADSL). This also hits products like the Slingbox, where you make use of the data coming out of your home.
In the US, where competition is intense and major corporations such as Verizon and AT&T are rolling out fibre in the major urban areas it's easy to pick up good synchronous providers at very high data rates.
With this rollout comes a hybrid solution to the bandwidth problem - that service providers manage bandwidth within their own networks and redistribute it from one customer to another - this overcomes some of the security and performance issues of existing peer-to-peer services. With Net Neutrality teetering, this would be one genuine and defensible way of offering added value within a specific service provider, and possibly, dare I say, charging for it.
Comments