Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : George Gilder - Forbes ASAP -- Ignore unavailable to you. Want to Upgrade?

To: George Gilder who wrote (4765)	9/5/2000 6:50:42 PM
From: Tom Clarke	Respond to of 5853

Terrific! When will we start seeing changes at the American Spectator?

To: George Gilder who wrote (4765)	9/6/2000 1:26:47 AM
From: Xenogenetic	Respond to of 5853

George, I have been hearing that you will soon be doing a newsletter on biotech. Is that true? I always had the impression that a successful biotech investor would need a degree in a molecular biology or an M.D? If you are indeed tackling the biotech sector, how do you anticipate approaching that sector? Surely, biotech cannot be approached the same way that technology can, can it? -Xeno

To: George Gilder who wrote (4765)	9/7/2000 12:21:58 PM
From: The Phoenix	Read Replies (2) \| Respond to of 5853

RE your recent report on the ultimate demise of CSCO - the migration of routing to edge consumer electronics devices connected to all optical networks... although this all could happen you're talking 20 years - at least - from now. State of the art in optical switching: lightreading.com Speculation about the technology being used in Corvis Corp.'s all-optical switch reached fever pitch last week after Light Reading published a story suggesting that some new developments from Iolon Inc., a startup, might provide a clue (see Corvis's Secret Sauce? ) A much more likely candidate has now emerged. Corvis, it turns out, filed a patent some time ago -- International Publication Number WO 00/05832 -- which spells out a switch architecture based on research originally carried out at Imperial College in London. Although Corvis itself declines to say whether its switch is based on this patent, other sources close to Corvis and Imperial College say that it definitely is. Corvis is reckoned to have developed the only all-optical switch that's anywhere close to deployment. It's known to be a 6-by-6 port switch, and Corvis has made a big thing out of saying that it's built from standard components -- the reliability of which has been proven in practice. That description exactly fits the switch architecture cited in the patent. Essentially, it's built around 36 fiber amplifiers that allow light to pass when they're powered up and block light at other times. Each input port is connected to a splitter that divides the light among six of these amplifiers. At the other end of the amplifiers, the reverse happens. The light is collected by six further splitters and steered to the output ports. The upshot is a relatively small, clunky, all-optical switch made from standard components -- rather a lot of components, which raises question marks regarding cost and reliability, according to some experts. It's also worth noting that Corvis's all-optical switch is unlikely to deliver all-optical networks in practice. In order to use bandwidth efficiently, the wavelength of some streams of light pulses would need to be changed. Right now, the only way of accomplishing this is to shunt them out of the switch, convert them into electrical signals, retransmit them at the new wavelength, and then shunt them back into the switch for the next leg of their journey. Electronic cross connects make this state of the art look like a joke. Yes, optics will get there but deployable in a competitive, cost effective way to electronic switches...... what 3-4-5 years from now? And that's just the core switching piece. As for Network processors. They can not accomplish content management, profile based routing, traffic management and whole host of other higher layer functions... and will likely never be able to. Products like the IXP have had no - zero -traction on the high routing platforms and their penetration has been at the edge where processing requirements are low and simplified. The recent piece you wrote/co-authored takes a very simplistic view - assuming point to point - non service based (get bits from here to there) perspective. I'll do a more detailed piece on this later but you should be more clear about your position given the size of your following. BTW: It has yet to be proven that all optical switches are any better than OEO's. Yes, OOO is the rage and the hype but OOO does not have the repariablity, cost, or management functionality as an OEO and it has yet to be proven whether they're any faster. They will certainly be more expensive to maintain and technical experts are simply not available. The point - we are many many MANY years away from where your article is pointing and many many MANY things can happen between then and now and it's simply a disservice to say to your readers that one day we'll live on Mars and we'll all be wearing white suits that allow us to move and breath freely while walking on the surface of the planet. Yeah....... maybe...... but what about the next 50-100 years... what about that? OG