Tuesday, October 12, 2004
Woah! Verizon's fiber service is six times faster than cable
This New York Times article, Phone Line Alchemy: Copper Into Fiber, by Ken Belson, October 11, 2004 says:
(If the telcos wanted to, they could do 600 times faster for about the same price.)
(Cable itself is poised to run 600 times faster than cable. Like dude? Moore's Law?)
(Oh, right, I forgot that without scarcity there'll be nothing to sell.)
(Senior policy people at the FCC believe that wireless technology (which can run hundreds of times faster than cable) will help "discipline" the telcos.)
(Of course, if the telcos are using PON (passive optical network) technology, they'll need a total network change-over to "impose discipline" -- that is, to make any one connection run at market speeds.)
(Guess what? They're using PON.)
(Woah . . . six times . . . wow . . . lemme have another hit of that doob . . .ah, good stuff . . . six whoooole times . . . ahhhhhh . . . . )
Thanks, Jeff Jarvis.
Verizon and the other regional Bell companies are losing customers by the millions as people drop their old phone lines in favor of cellphones, e-mail and ever cheaper phone services from cable companies.Check.
To battle back, the phone companies are trying to outdo their archrivals, the cable companies, by installing a network of fiber optic lines to reach tens of millions of American homes - lines able to carry not only phone calls but television programming and Internet connections at six times the speed of cable company lines.Six times, woah, six freakin' times!!! They should hire the Dell dude for commercials, to get across the idea of "smoking something."
(If the telcos wanted to, they could do 600 times faster for about the same price.)
(Cable itself is poised to run 600 times faster than cable. Like dude? Moore's Law?)
(Oh, right, I forgot that without scarcity there'll be nothing to sell.)
(Senior policy people at the FCC believe that wireless technology (which can run hundreds of times faster than cable) will help "discipline" the telcos.)
(Of course, if the telcos are using PON (passive optical network) technology, they'll need a total network change-over to "impose discipline" -- that is, to make any one connection run at market speeds.)
(Guess what? They're using PON.)
(Woah . . . six times . . . wow . . . lemme have another hit of that doob . . .ah, good stuff . . . six whoooole times . . . ahhhhhh . . . . )
Thanks, Jeff Jarvis.
Comments:
Cranky today, aren't we:) Nice rant...
Yeah, they could do an all-in-one Ethernet FTTH network instead of PON, but that has yet another set of problems and just doesn't fit into a 5 nines telco culture. I am not a big fan of PON solutions, because they do hardwire decisions that are better off delayed, but there are valid reasons to go that route.
You know that the internet speed doesn't include video bandwidth and it is a bit of a cheap shot to just look at internet BW instead of the combined TV, IP, and phone bandwidth. Would higher IP bandwidth really make that much difference in the possible applications?
Doc
Yeah, they could do an all-in-one Ethernet FTTH network instead of PON, but that has yet another set of problems and just doesn't fit into a 5 nines telco culture. I am not a big fan of PON solutions, because they do hardwire decisions that are better off delayed, but there are valid reasons to go that route.
You know that the internet speed doesn't include video bandwidth and it is a bit of a cheap shot to just look at internet BW instead of the combined TV, IP, and phone bandwidth. Would higher IP bandwidth really make that much difference in the possible applications?
Doc
Don McC,
Re: 5 nines, the Internet allows reliable apps over unreliable transmission. As a result, you can buy as many nines as you need, see http://www.isen.com/archives/020626.html
Re: Sharing the capacity of a fiber with TV, (a) the capacity of the fiber is, literally, DC to daylight and (b) I don't want TV as a separate, specialized service, just give me TVoIP.
Re: Feeling cranky today, Yer dern tootin I am. Proud of it too.
Re: 5 nines, the Internet allows reliable apps over unreliable transmission. As a result, you can buy as many nines as you need, see http://www.isen.com/archives/020626.html
Re: Sharing the capacity of a fiber with TV, (a) the capacity of the fiber is, literally, DC to daylight and (b) I don't want TV as a separate, specialized service, just give me TVoIP.
Re: Feeling cranky today, Yer dern tootin I am. Proud of it too.
you said:
"Re: 5 nines, the Internet allows reliable apps over unreliable transmission. As a result, you can buy as many nines as you need, see http://www.isen.com/archives/020626.html "
But in FTTH, you aren't going to have redundant paths near the home, or likely even to the neighborhood. Not enough bucks to be able to afford at consumer price points and no real need to be better than the MSOs.
Also note that for realtime video (aka TV :) delivery we can't have much jitter or packet loss. Can do FEC if needed, but this again eats up banwidth and CPU cycles. If we don't want to deliver "TV", but a typical PC streaming video or download and play scenario, then the problem is simpler.
you said:
"Re: Sharing the capacity of a fiber with TV, (a) the capacity of the fiber is, literally, DC to daylight and (b) I don't want TV as a separate, specialized service, just give me TVoIP. "
Oh? What can I go buy (at a price that makes a business work) that gives me daylight bandwidth :)? Or even something close? Say 4 Gbit around town and 100 mbit to the home? Sure, I can build this network if I make it distributed enough, but placing a lot of gear in the field has another set of problems - the problems the PON people are trying to avoid. Or should we just build an HFC network and be happy?
( Dropping my dripping sarcam tone for a moment, what IS the current affordable BW delivery mechanism for metro fiber? How much BW can we get over what technology? Ok, back to sarcasm >:. )
TVoIP also has a set of problems - multicast rarely works as advertised and deals with channel flipping painfully. VOD only scales well (storage and bw-wise)when it is distributed, but again that means more gear in the field. A STB for every TV adds more cost since we don't have any analog layer of channels.
The only competitive advantage the RBOCs have in the FTTH space is right-of-way and telephone poles. This is a big advantage, but somewhat outweighed by their high cost structure and tight regulation that increases the risk of them moving. Combine this with a risk adverse corporate culture and it is amazing that the move at all into this space.
Also, the longer they can delay, the cheaper the system gets - Last Mover advantage at the most brutal.
David, I think that you (and Gilder, even more so) tend to gloss over the dificulties in building this kind of network in your frustration that it hasn't been built yet. Yeah, give me enough money and I can build a profitable company around it, but in the current environemt there isn't anyway to raise that money except for maybe the RBOCs, and they are scared it will cripple them financially. The Community owned networks sound wonderfull, but cities are not known for their technical acumen.
thanks for the response!
Doc
"Re: 5 nines, the Internet allows reliable apps over unreliable transmission. As a result, you can buy as many nines as you need, see http://www.isen.com/archives/020626.html "
But in FTTH, you aren't going to have redundant paths near the home, or likely even to the neighborhood. Not enough bucks to be able to afford at consumer price points and no real need to be better than the MSOs.
Also note that for realtime video (aka TV :) delivery we can't have much jitter or packet loss. Can do FEC if needed, but this again eats up banwidth and CPU cycles. If we don't want to deliver "TV", but a typical PC streaming video or download and play scenario, then the problem is simpler.
you said:
"Re: Sharing the capacity of a fiber with TV, (a) the capacity of the fiber is, literally, DC to daylight and (b) I don't want TV as a separate, specialized service, just give me TVoIP. "
Oh? What can I go buy (at a price that makes a business work) that gives me daylight bandwidth :)? Or even something close? Say 4 Gbit around town and 100 mbit to the home? Sure, I can build this network if I make it distributed enough, but placing a lot of gear in the field has another set of problems - the problems the PON people are trying to avoid. Or should we just build an HFC network and be happy?
( Dropping my dripping sarcam tone for a moment, what IS the current affordable BW delivery mechanism for metro fiber? How much BW can we get over what technology? Ok, back to sarcasm >:. )
TVoIP also has a set of problems - multicast rarely works as advertised and deals with channel flipping painfully. VOD only scales well (storage and bw-wise)when it is distributed, but again that means more gear in the field. A STB for every TV adds more cost since we don't have any analog layer of channels.
The only competitive advantage the RBOCs have in the FTTH space is right-of-way and telephone poles. This is a big advantage, but somewhat outweighed by their high cost structure and tight regulation that increases the risk of them moving. Combine this with a risk adverse corporate culture and it is amazing that the move at all into this space.
Also, the longer they can delay, the cheaper the system gets - Last Mover advantage at the most brutal.
David, I think that you (and Gilder, even more so) tend to gloss over the dificulties in building this kind of network in your frustration that it hasn't been built yet. Yeah, give me enough money and I can build a profitable company around it, but in the current environemt there isn't anyway to raise that money except for maybe the RBOCs, and they are scared it will cripple them financially. The Community owned networks sound wonderfull, but cities are not known for their technical acumen.
thanks for the response!
Doc
About 2 hours ago a Verizon tech knocked on my door to tell me that he will be in the backyard for a few days pulling fiber to the houses. He offered that the top service would be 60megabits/s. That sounds like about 6x to me.
Post a Comment
