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 : Silkroad -- Ignore unavailable to you. Want to Upgrade?

To: Sector Investor who wrote (85)	11/16/1998 10:05:00 PM
From: Frank A. Coluccio	Read Replies (2) \| Respond to of 626

Hi Sector, >> let me say that this thread's technical level is way beyond my skills.<< [[ Don't sell yourself short, SI, I've followed your posts around the boards. Anyway, how do you think I feel having to keep pace with Dr. Ahhaha's seemingly improvised rants?!? Actually they are not, but man... what a departure from the normal chatter around here, wouldn't you agree? I may actually be intimidated into learning something new, if this keeps up! ]] The point you are questioning was in my reply # 60. Since then I've been trying to throw some stones at SR, figuratively speaking, trying to illustrate where I think that there will be some barriers. And your points in that regard are well taken. But getting back to my original point, it's the mental lock, the "mind set," that I was referring to. If the industry proceeds as though DWDM is the only means of getting beyond the silicon-friendly OC-48 rate (or one marginally higher), then this will be stifling and counter-productive from an innovation standpoint. I say OC-48 is silicon friendly because producing system speeds above this rate at this time is considered economically unattractive according to many, although an increasing number of OC-192 systems are now being deployed. -- Here, I'm taking a lot of factors into account in defining an end-to-end "system," and not just the boxes themselves .....[route-distance vs amplification intervals vs hardware and net management, SONET element costs, WDM costs, etc. -- it's a lenghty, and often subjectively influenced series of trade-off analyses to determine if you want to do 4 OC-48s or a single 192... I think that that is the gist of this discussion, as well, only at a higher order, come to think of it! ]. When multiple OC-48s are created and each is mapped to its own lambda, effectively, at these 'relatively low' speeds, it has a tendency of creating an unnecessary and burdensome level of administrative 'make work.' This is because it creates an unnecessary number of physical (optical) tributaries on the "span" side, and a correspondingly higher number of network management problems, when all along, a single stream could have done the trick very well. It's the difference between dealing with one port at each end of a pipe versus possibly hundreds of them, between two points in a mesh that do not necessarily require fan outs, such as those between major peering NAPs. Granted, there will be times when grooming and splitting will be required. But I think that this is one of the areas of photonics that needs to be further cultivated in this space, instead of being obviated due to a DWDM workaround. {{...oh man, I'd better set my shields up now, 'cause I'm really pushing my luck here...}} We've seen this many times with lower order transmission systems. Take ordinary copper circuits, for example. Why would a provider want to deliver 24 separate copper lines into a user location when a single T1 would do just fine, and save money at the same time? Why deliver 28 individual T-1s, when a T-3 does the trick? Why deliver 12 individual T-3s to a location, each over their own fiber strands or coaxial feeds, when an OC-12 pipe does it all within a 7 to 9 micron silica core? There are reasons why some users would prefer the counter-intuitive selection, i.e., to elect to have 24 lines, or 12 individual T-3s, due to their belief (with some justification at times) that there is robustness in numbers, but that's an exception, and not the issue here. OK, so much for the confidence level and rationalization of this scheme. And keep in mind that after all is said and done, these assumptions are all predicated on the hypothesis that this stuff actually works. [grin]

To: Sector Investor who wrote (85)	11/17/1998 12:29:00 AM
From: ahhaha	Read Replies (1) \| Respond to of 626

To add a little to Frank's comment, think of a multi-lane freeway. As the car number density rises the flow slows, so more lanes must be added. You can't get people to refrain from lane changing, cops tagging speeders slows the flow, collisions do the same. When the freeway has more than 12 lanes in one direction, the flow realizes an unexpected turbulence. In heavy traffic drivers can't see where the edge of the road is, so they make crazy moves to catch their exits. And you get paranoid because you're trapped in a sea of careening cars. Each lane is like a WDM laser color. You can increase the car number density of a lane, you can try to get drivers not to change lanes with signs, and you try to educate drivers on freeway etiquette including single car strategies to promote laminar flow, but there is a limit to these strategies. QoS tries to implement diamond lanes and outside lanes for trucks to organize traffic by certain priorities. ATM tries to put cars on long flatbed trucks and haul them along so that they can't change lanes or get in accidents. Other strategies use office daily opening staggered times to alleviate bunching. At about OC-768 you have the 12 laner. You have a daily mess. You simply can't add any more lanes because people go crazy. SR builds busses. SR builds multi-layered car delivery trucks. Either all the lanes go into one lane with n-decker car delivery trucks or the people ride busses in one lane. We haven't figured out exactly which of these techniques is being used by SR, but we are hot on the trail. Suffice it to say both these are four dimensional phenomena or need four degrees of freedom to describe. The people are taking a seat in the bus while the bus is zooming along. Or the cars are being stacked while the truck flies along. When the vehicle reaches a destination, they get off or are unloaded the way they were loaded. Trucks and busses can be n-deckered which implies something like OC 192000, and you can have traffic going both ways simultaneously in one lane without collision. That might need another degree of freedom! The End.