Broadcast vs. point-to-point

Dan Clemmensen (
Wed, 25 Nov 1998 20:53:55 -0500

This is mostly for Ken, but othres may be interested.

I sympathize with the irritiation Ken and others have shown RE the structural monopoly the phone company has on the "last mile", and I'd really love to find a way to break this monopoly. Let's agree on the goals:

--bottom-up networking to replace the Phome Company.
--fully-distributed switches under control of individuals
--Massive bandwidth available at nearly zero cost
--full support for mobile

The problem is that radio is expensive by comparison to wireor fiber, and that the radio spectrum is limited. Therefore, I feel that we need to reserve the radio spectrum for mobile applications and use point-to-point connections where possible. Your concept of neighbor-to-neighbor connections is worthwhile, but don't use radio: use either fiber or point-to-point through-the-air lasers. This is much cheaper and provides much higher bandwidth. so, where you can do it cheaply, you run a fiber to your neighbor's house. Where this is infeasible, you try for line-of-sight laser. on your own property you certainly want to run fiber or coax to your own security cams, etc. This preserves radio bandwidth, decreases your and costs, and increases your own security. With fiber, you basically get as much bandwidth as you want to pay for on each fiber, with 100Mbps being incredibly cheap even today, and 40Gbps costing a bunch. Now all you need to do is work co-operatively with your neighbors to buy a line to the internet since your neighborhood is now fully interconnected, you have enough bandwidth and purchasing power to attract the nodice of the bypass folks.

Now what do we do for mobile? I reccomend that we use radio for this. Unfortunately, a sophisticated cellular system costs a bunch. However, we may be able to beat this by adding smart point-to-point laser systems to the call system. When your mobile unit can "see" a fixed laser installation, it can hop onto the laser link just as it hops between cells. This preserves the radio cells for the units that are not in laser line-of-sight, which means that you need fewer expensive radio cells to serve the same mobile population.

Now, let's think about the evolution from today's system to the system outlined above. Frankly, I believe that the evolution will be driven initially by hackers, for fun. This is what we are seeing in the Open Source software community. Will this be enough to have a real effect on the cisco-Lucent-RBOC oligopoly? Only time will tell.

Why is this relevant to extropians? well, some of us believe that human progress is now driven by computation and connectivity.