Mike, we need to address a "key reality" which is implied (but not stated) by Gilder...fiber optics may be popping up all over the place--however, the reality is we live in an electronic world--NOT a photonic world. Let me explain, and I think you'll "see the light" (pun).
Whereas today the "biggest WAN backbones" could possibly be all-photonic, this has huge ramifications "down the pipe." Why? Almost everything in the world is electronic rather than photonic. Examples: PCs, Intel processors...human brains. So, we would expect more and more optical-to-electronic interfaces as an "expression of the evolution of the telecosm." ...Wow
It is this very electronic nature that makes Vitesse semiconductors and high speed LANs so important--they provide the bridge from high-speed fiber optic to the "real" electronic world of our PCs and our human brains.
...so you can think of Vitesse semiconductors and high-speed LAN switching as the beneficiaries on the "telecosm" in the real world.
Sure, someday human beings may be connected by fiber optic cables, hopefully not in my lifetime (that might hurt, and look funny besides--would it be mounted in the nose?)
Dan
You said:
"My read on Gilder's article is that ALL transistors (including GaAs) are slow relative to the speed of a all photonics network. His point being that the creation of an all photonic net, including photonic switches, will in the next few decades, decend in stages from the major national backbones all the way to the "desktop". The most significant opportunity being the residential entertainment/computing center "desktop". This will take time (as in more than you think) so I wouldn't overhaul the portfolio tomorrow based on the article." |
