Intel wants to break the cable model, allow for pay per channel viewing

[sclitheroe]

Originally Posted by nfotiu

It gets absorbed into the content provider's cost, and sure some of that cost gets passed to the consumer in the subscription fees.

However, Netflix is getting a lot more efficient about how this is delivered anyway:

http://blog.netflix.com/2012/06/anno...t-network.html

But, it still costs pennies to deliver the content compared to what everyone is spending for carriage fees for channels they don't watch.

That doesn't address the single most constrained and expensive part of the network. Netflix peering with the ISP's at the head end helps upstream, but it's the downstream network to the consumers that has to do the brunt of the heavy lifting, and that is the most bandwidth constrained. That's where the real costs to the ISP's are - maintaining, upgrading, and rolling out the distributed network that exists between their data centers and the consumer.

Netflix essentially gets that part of the network for free, irrespective of the peering arrangements they are willing to make through OpenConnect.

And that's why we still need pay-per-byte internet - the ISP's need to be able to recoup the costs associated with delivering content from places like Netflix that Netflix doesn't pay for. Otherwise, you are right back to a situation where the ISP's are deciding which providers to allow on their network, based on how much the providers pay into the bandwidth costs for the load they create, which is the antithesis of net neutrality.

[arsenal]

Rather than a pure pay per channel model, you could a hybrid. Still pay for channel packs, but you choose which channels are included rather than the cable or internet company choosing.

Say standard local stations = $10/month
HD Package includes 5, 10, 15 channels of your choice = $10/$15/$20
and so on.

It is still very close to the current model, but it gives more power to the user to select which channels they want to subscribe to rather than the cable companies packaging channels together. Then it's up to the cable provider to distribute funds to channels from the subscribers funds. You would very quickly see unpopular channels go away, and the most popular ones will probably increase in revenue.

[nfotiu]

Originally Posted by sclitheroe

That doesn't address the single most constrained and expensive part of the network. Netflix peering with the ISP's at the head end helps upstream, but it's the downstream network to the consumers that has to do the brunt of the heavy lifting, and that is the most bandwidth constrained. That's where the real costs to the ISP's are - maintaining, upgrading, and rolling out the distributed network that exists between their data centers and the consumer.

Netflix essentially gets that part of the network for free, irrespective of the peering arrangements they are willing to make through OpenConnect.

And that's why we still need pay-per-byte internet - the ISP's need to be able to recoup the costs associated with delivering content from places like Netflix that Netflix doesn't pay for. Otherwise, you are right back to a situation where the ISP's are deciding which providers to allow on their network, based on how much the providers pay into the bandwidth costs for the load they create, which is the antithesis of net neutrality.

I guess it depends on the area. Most urban and suburban areas in my state have a choice between fiber to the house and docsis 3.0, so that infrastructure can easily handle things. Plus if it is over the top iptv, there are a lot of efficiencies that can be done by multi casting linear tv. Not to mention that everyone is already downloading and streaming boatloads of netflix, hulu, amazon and pirated material.

[nfotiu]

Originally Posted by arsenal

Rather than a pure pay per channel model, you could a hybrid. Still pay for channel packs, but you choose which channels are included rather than the cable or internet company choosing.

Say standard local stations = $10/month
HD Package includes 5, 10, 15 channels of your choice = $10/$15/$20
and so on.

It is still very close to the current model, but it gives more power to the user to select which channels they want to subscribe to rather than the cable companies packaging channels together. Then it's up to the cable provider to distribute funds to channels from the subscribers funds. You would very quickly see unpopular channels go away, and the most popular ones will probably increase in revenue.

Doesn't really work out that way though. I don't know the exact numbers, but I'd guess if you looked at a typical American television bill without equipment rentals or premium channels, it would be about 200 channels for $40-50. About $20 would go to a handful of sports networks, about $5 would go to the other 190 or so channels, and the rest would go to the carrier for infrastructure, operating costs and profit. The vast majority of channels, you are not really paying for anyway, they make their money by being available for free and are ad supported.