To: John Trader who wrote (50880 ) 8/21/2001 6:15:26 PM From: Jacob Snyder Read Replies (2) | Respond to of 70976 re: is the internet growing: First....... this guy is speaking his position. "Dr. Lawrence Roberts led the team that designed and developed ARPANET....which evolved into the modern Internet....... Today, Roberts is the founder, chair and CTO of Caspian Networks, a high-profile networking startup." So, he has a big financial and intellectual interest in seeing the internet "grow". And a lot of scientific results are determined before the data is gathered, by the assumptions and biases of the observer. Next, his results cannot be verified. "Service providers normally keep network traffic statistics confidential for competitive reasons.", and then, "...Roberts has special access to top scientists at data carriers." So, he is a "special" observer, with "special" access. A basic requirement for Good Results (using the Scientific Method), is for data to be gathered by disinterested observers, and for results to be reproducible. This isn't possible in the real world, but it's a big Red Flag when these principles are violated so egregiously. I learned, when reading medical journals, to always ask, "who funded this research? What Money is speaking here?" OK, let's suspend disbelief, give the guy the benefit of the (mountain of) doubt. Let's assume that data traffic in core networks is still growing rapidly. What does this imply? AMAT sits at the end of a chain of buyers and sellers, a chain of cause and effect. The end-users of chips, driven by a need to do something that requires chips, are at the other end of this chain. It's a long chain. Some of the links in the chain are elastic. That is, a tug on one end of a link may not translate into a tug on the other end, or it might not translate into a tug for months....or years. So, the connection between traffic in data networks and AMAT stock price is tenuous, longterm only, and is dependent on a lot of other variables. Once a chain of cause-and-effect gets longer than about 3 steps, small discrepancies in each step can add up, to make the end result a near-random number. Further, I don't know how or even whether growth in data traffic translates into profits for any company. And, in investing, the only thing that matters is profits. Not page hits. Not data moved. Not hours spent online. Not number of web sites. Not even sales. All numbers are irrelevant if you can't define their impact on profits. So far, the history of investing and the internet, has been a history of a series of failed proxies for future profits. What is said in the article doesn't really contradict what I've read elsewhere, because (as the article says), everyone is talking about something different, when they talk about "growth" in the internet. It's the old "5 blind men describing an elephant" problem. Let's say those networks have, over the last 6 months, increased capacity utilization from 2% to 4%. That would translate (other things staying the same, which they haven't) into a doubling of traffic. That would be Exuberant Growth, right?