IEEE Standards Board Passes 802.16a
papason writes "Welcome the birth of the IEEE's first wireless MAN standard for broadband wireless access in bands ranging from 2GHz to 11GHz. Yes, the same group that brought you 802.11b has brought you a real
broadband wireless access standard. See wirelessman.org for more details."
I seriously doubt if this is going to use unlicensed spectrum like 802.11. You just can't move that amount of data over that much distance with those little 15 milliwatt(?) transmitters that 802.11 uses. And you can't have thousands of the things active in a city at the same time without clobbering each other.
So expect yet more monopolies given to whichever corporate greedheads have the best political connections, just like in radio and TV broadcasting. Sigh.
The mesh idea is a wonderful concept and something I hope for in the future, but has a number of problems yet to be solved:
* A way to guarantee high quality of service from one side of the globe to another, or even one US state to another.
Who is going to lay down the pipes across the ocean and across the desert? And even if some benevolent souls on the edges of civilization decide to donate long distance pipes, will they be large enough to handle all that multiplexed traffic?
* Manufacturers of Internet equipment and standards committees working together towards this goal.
Decentralization could potentially kill off markets, so what incentive do manufacturers have for designing protocols and building equipment for distributed routing?
* Technological barriers
How would this be implemented? I'm sure there has probably been research done on distributed routing and name lookup services, but will it work at a large scale? And would it be reliable?
* Adoption
With cable, DSL, and wireless, you give someone some money, and all of a sudden you have a connection. With a distributed system, you would need to coordinate with your neighbors, since you can't rely on a company to keep things running. A possible solution would be that companies could be highered to organize and setup a neighborhood network, and then be hands off after that, except for maintanence and upgrades.
Anyway, I hope that the work is done to make this feasible, and that people could be convinced to join a distributed network and get off the "feed".
LS
There is a fine line between being a cultivated citizen and being someone else's crop. - A. J. Patrick Liszkie
Why not offer to help them out, a small 200 seat ISP probably isn't making a whole lotta moola but if it improves YOUR speed that's all that matters :)
I'd start with a nice Linux box at the front end to handle gateway, firewall and transparent squid duties, you won't need to get overly fancy, especially if you skip the cache (although it would really be in their best interests).
For a couple G in hardware you could likely save them 25% in bandwidth at the proxy, plus you could use iptables etc. for traffic shaping, throttle using iptables or squid, etc.
I've done it, I've installed 2 headend gateways for small cable/wireless ISPs here in BC, works like a dream and it does save about 25% at the proxy. Mind you, they also have the luxury of being able to throttle users at the cable modem, so they only offer 384/128 service residentially but it beats dialup, although they also own the only dialup service in town too.