Will Internet TV Crash the Internet?

Stony Stevenson writes "Analyst groups and Cisco have come out saying that the internet is heading for a crash unless it increases its bandwidth capabilities which are being strangled by the increased use of Web TV. Stan Schatt, research director at ABI said: "Uploading bandwidth is going to have to increase, and the cable providers are going to get killed on bandwidth as HD programming becomes more commonplace." He added that the solution to the problem is to change to digital switching and move to IPTV. "They will be brought kicking and screaming into the 21st century," he said. Cisco weighed into the argument, adding that it had found American video websites currently transmit more data per month than the entire amount of traffic sent over the internet in 2000."

Comments

By digital switching, they mean IP Multicast

[mbone]

They certainly have had time to deploy it.

Re:By digital switching, they mean IP Multicast

[Znork]

"IP Multicast is great to stream shows on a scheduled basis."

Once you start using MythTV or other capable PVR application you change your view of 'tuning in'. You simply dont do that anymore, you just mark what you want and treat it as a delayed-scheduled download service. Heck, the next step in that evolution (as storage grows the next order of magnitude) is simply using multiple tuners and pre-recording everything, so you can, in effect, decide what you want to watch post-multicast.

If unicasts overload the network...

[JackHoffman]

then let's have multicasting. There you are, another good reason for IPv6. Get to it.

Imminent Demise of the Internet Predicted

[PhoenixHack]

Choice quotes from this article written at the close of 1995:

http://www.infoworld.com/cgi-bin/displayNew.pl?/me tcalfe/bm120495.htm

Dazzling product literature and advertising require at least ISDN speeds. But the major corporations upon which we are relying to upgrade Internet access past 28.8Kbps are the local telco monopolies, which like our postal service and public schools have become little more than jobs programs. The local telcos will escape demonopolization in 1995 and, while they pursue long distance voice business in 1996, their motivation to lower costs on high-speed Internet access will wither, fatally constipating the Web.

You've read that the Internet was designed to survive thermonuclear war, but it's repeatedly been brought to its knees, its circuits choked, for example, by the reaction to one measly jury verdict in Los Angles. The Internet is intermittently overloaded, and the TCP/IP architecture doesn't deal well with overloads. Furthermore, the Internet's naive flat-rate business model is incapable of financing the new capacity it would need to serve continued growth, if there were any, but there won't be, so no problem.

One of two bad things will happen with video over the Internet during 1996. Either the Internet's attached computers, operating systems, and applications software will fail to deliver video, or they will succeed. If they succeed, the packet-punctuated pre-Asynchronous Transfer Mode Internet will fail to carry it. In either case, without video the Internet will lack the energy needed to sustain its current expansion.

The Internet traffic carrying arguments about pornography on the Internet will during 1996 swamp the actual pornography, so even the most sophisticated Web search engines will too often fail to find any. What quicker road to collapse? More gems to be found via http://www.google.com/search?hl=en&q=imminent+demi se+of+the+internet+predicted...

Re:It's not rocket science

[Shishak]

I would gladly sell you what you are buying.

Lets see, 24mbps (ADSL2+) @ $30/mbps = $720/month. Are you willing to pay that much money or would you like me to overcommit 100:1 and get the price down to $7.20 ?

$30/meg is decent bandwidth, you can approach $10/meg for crap bandwidth but you do get what you pay for.

To put it another way, $30,000 for a GigE connection per month, but you need 2 of them because you have to be redundant, so $60,000 for 1 Gigabit of redundant bandwidth. A HDTV signal eats up 7mbps so you can support 142 of them on a GigE connection, $422 per channel. Using multicast you can send the same channel to multiple customers (IPTV) but that is broadcast, not pay-per-view. You wouldn't be able to watch on-demand or fast-forward the signal. You could pause/rewind it if you had a hard drive in your set top box. That isn't what consumers want. As a provider selling triple-play services you need to dedicate at least 7mbps per end user in your edge/aggregation network. You will also need massive hard drive caches in your POP to cache as much content as close to your subscribers as possible. Set top boxes with big drives so you can pre-load content using multicast/broadcast techniques (i.e. pre-load the new hit movie on all set boxes and make them available on the release date) The cable infrastructure isn't built to handle this type of content delivery. DSL is but DSL distance limitations make getting 7mbps to customers hard (10kft limit). FTTH is the obvious answer but that is insanley expensive

The days of broadcast television are dying. Things like AppleTV & YouTube are going to kill it. Soon independant television producers will be able to produce/distribute the show directly to the consumer, no need to sell the show to ABC/CBS/NBC/FOX. You'll be able to subscribe to the shows and download them. All of that is unicast traffic and it will destroy internet bandwidth ratios.

Apple & iTunes is the way to go, once they start distributing content around the Internet (ala Akamai) they will have the parts needed to replace the broadcasters.