Timothy:
While HFC and many other technologies may work well, you have to get the connections to millions of individual sites, and that is terribly expensive. It is not a technical problem; there are many technologies which can perform very well IF you are allowed to build the system any way you want it, and IF you can pay for it. It would cost HUNDREDS of BILLIONS of dollars to replace the POTS systems we now have with any other kind of high speed system. The best near-term solutions will be those which can make use of POTS, somehow, and get installed into the POTS system relatively quickly.
Although there are a lot more Internet users now than ever before, there aren't nearly enough to cover the cost to upgrade to a true high speed network, whatever technology is used. When it is time to upgrade our whole telecommunications infrastructure anyway, it will be upgraded with high bandwidth systems, and we will have tremendous bandwidth. Until then, though, most people will not have access to true high bandwidth networks; only those who live in areas where the population is dense enough to justify the cost will have them, or, areas chosen for trials which don't need to pay for themselves or make profits.
Cable MODEMS are not the problem; there have been useable cable modems around for several years now. It is the SYSTEM that is the problem. For example, my cable TV company has over 1,000 people on the cable loop that I'm on. That means that if, say, half of them were on the Internet, and the bandwidth of the cable system was 10 Mbps (which is higher than it is now but it could be made that high), each user could be limited to only 20 kbps if everyone happened to be receiving something, and since Internet access is mostly downloads, that would happen sometimes. Even if you start assuming this and that, you still have to realize that each user is only going to see bandwidth ar rates about equal to dial up, ISDN or other modems much of the time; they are going to see brief "bursts" of high bandwidth but we found in our tests that these only tended to frustrate customers; they'd get used to fast downloads and would become angry and complain when the lines were slowed down by heavy useage.
In addition, for the time being, those 500 cable customers would also tie up 500 POTS lines for the upstream connection. If you modify the entire cable system, to make it two-way (which is immensely more complex than might be imagined), the cost is simply too high to be borne by the numbers of people who would pay for it for Internet access; aside from the "core" customers who crave high bandwidth, many others simply aren't that worried about it, and won't pay high costs for it. If they can have it at or near today's rates, they'll take it, but not at expensive rates.
Some time in the future, all this will change. But it will happen much more slowly than the pioneers who are now pushing the technologies say it will. It is their job to push technology; that is what will make it happen sooner rather than later. But, unless you live in Santa Clara or someplace like that, you've got a long wait for high bandwidth of ANY form that really performs at the levels being talked about. SHucks, we just got ISDN here recently!
Things will get steadily better, and if you're lucky enough to live where you can get on a particularly good system, enjoy it. But across the board, my belief is that whatever can be done with POTS systems will be what the majority of us will be using for the forseeable future. It isn't a technology issue, it is a price and logistic issue. It would take years just to put in HFC everywhere even if we started tomorrow morning, and if we had the $$$$ to do it.
I'm putting my near term bets on companies that can do something with what is there: POTS. I think xDSL and other POTS technologies have a better chance of widespread adoption, but I worry that the RBOCs aren't going to do anything with them, so even POTS based solutions worry me; (in act, I'm not impressed with ANYBODY's solutions right now). Perhaps some of the companies, such as Worldcom/UUnet, which are starting to bypass the RBOCs, will force something good to happen, who knows.
Far out in the mists, too far to see clearly, are the fiber technologies of all sorts, which will be the best choice when new systems are installed or replaced. But I believe that will happen gradually and we may not realize it is happening, then one day, we'll realize that it creeped up on us and is gaining ground. So those would be a long term play, in my opinion, based on what I believe.
Whether or not any of these are good investments depends on what your goals are. Fiber based systems are long term, POTS systems are more immediate, but, also hard to call. I would ALWAYS look past the technologies, though, and examine the cost, logistic, political/regulatory issues; they will control what happens much more than technology, which always leads market acceptance by years.
That's only my opinion, but such as it is, that's what I believe.
Larry |
