Sounds great, but isn't it still "pie in the sky?" We aren't there yet, are we? I have been following this technology for quite a while, I think you have too, if I am correct.
Since I am not an engineer, I follow it from the ISP angle, my favorite source is "Boardwatch," I read the magazine, but it's also on the net,
boardwatch.com
According to "Boardwatch" editor Jack Rickard, this last piece dated November, 1998, there is simply not enough bandwidth UPSTREAM for all the highspeed modems to connect to:
". . . [I]t is probably true that multiplexing is more efficient at higher delivery speeds. But it doesn't quite negate the laws of physics. US West has announced a goal of 100,000 DSL customers by the end of the year, a tiny fraction of the people on the Internet in their 14 state territory. But at that, the upstream bandwidth to support those connections is not there and I don't think will be there for some time. Assuming that the 4:1 multiplexing experienced in the dial-up world expands to 10:1 in this new better universe, and I think that's generous to an insupportable fault, they need 2.5 Gbps of upstream connectivity to handle 100K customers. And if you are one of those first 10,000 customers, it is indeed going to feel fantastic."
"And the technology is really already available to do switching at those speeds. The problem is, that the Internet works about as fast as its slowest link. And the interconnect between networks isn't operating at that speed, and currently there is no political will among service providers to fix it. The problem isn't in the servers at all doling out bits. It's in the core of the network. Currently, no matter what USWest, or your cable company, does, there is no 2.5 Gbps Internet to connect to, and, of course, no place to make that connection."
"I'm sure they'll worry about it after they have the customers, but I've been concerned for some time that someone was going to solve the "last mile problem" and find out that really wasn't the problem." |
