Slashdot Mirror

The attentuation you're talking about is the wavelength/(2*pi) "skin depth" which is why we need more shortwave and more of a spectrum spread to counter the baud limitation.

If we're talking about plain old analog bandwidth with no algorithms then you're right, there's probably not enough to go around. But if we take significant chunks of the entire spectrum using short and long waves, combine multiplexing, and data caching/distribution models not unlike bittorent, it just may be possible to develop a pretty good working system.

Technology has an amazing way of evolving ingenuity. Take interfence for example. Collision dectection/aviodance was unthinkable even to the best engineers just a few years ago. The idea that you could run 100 access points within range of each other and get megabit throughput was thought impossible. Yet from my rooftop in downtown San Francisco I can see 150-200 Wi-Fi access points, many of them open, to which I can connect with no noticble interference problems, getting up to 1.5Mbs. CSMA/CA has solved the problem.

At the very least I think we amateurs should have the same rights in terms of bandwidth and power that any of the private providers have. Perhaps we could do a better job.

someone will have to design it, someone will have to upgrade it, someone will have to maintain it and someone will have to run it.

Ivan Seidenberg (who's an MBA not an enigineer) makes this sound way more complicated than it needs to be.

In my experience most bandwidth (business, residential, government) is mostly unused. San Francisco has a lot of surplus bandwidth that taxpayers have already paid for. What we'd like is for the city to fund wireless nodes via sflan that give some of this bandwidth back to the people.

I think Seidenberg's real concern is that Verizon won't be able to charge $29.95/month for something that should be practically free.

I have 3 words for this guy. Get a job.

$15-$20 million AND you have to pay monthly?

Let's suppose you buy 400 top of the line $500 access points from some wf-fi company. That's only $200,0000.

They should hold back on the fiber and use the access points to relay to a few key wired points like sflan does. Cuts down on the cost.

We're not crying "fix it!." It's already fixed.

I'm involved with sflan and all we're saying is buy some nodes and we'll give the city free wi-fi.

Of course if you want to give your money to SBC I'm sure they could have their MBAs find a lot of creative ways to spend it on credit card schemes, free airline miles, etc.

I build access points and donate bandwidth for the sflan project with the hope of bringing low to zero cost Internet access for everyone in San Francisco. One of our problems is the ridiculously low FCC imposed power restrictions on our trancievers while phone companies who paid millions to buy their part of the spectrum are allowed to use thousands of times the power we are.

I don't want to come off as a pessimist but my concern is that the furture of wireless be look more like the control-and-toll method of owning the spectrum and charging what you like for, spending nothing on R&D yet billions on marketing to create a lockdown system of over priced mediocre service.

Don't know how much of a Linux user you are but I built a couple of wireless access points/relays for my neighbors that hook up the to the sflan network.

I used a couple of old junker laptops running Linux. This should be completely scalable to your situation. Here's a link to how I did it. For a relay/access point the same principle applies, just use two cards, one with a directional antenna the links to you, the other with an omnidirectional that serves DHCP. I found that cheapo cards on old hardware works great.

There are plenty of sites that have intructions for building directional antennas. From my experience you don't need to spend a lot of money for this kind of thing.

I'm glad your sharing your bandwidth. More people need to do this. Hopefully, one day the Internet will be entirely wireless and free. Free meaning you don't have to pay an ISP, you just buy your equipment and you're online. Just like CB radio.

Feel free to use my email address at the above site if you have any questions.

Yes, but with 802.11g and 802.16 the specs allow for up to 50+Mbps, several T1s. For surfing the net and reading email this is enough to supprort more users than may be commonly found in the limited area covered by the current power restrictions.

The 802.X specs cleverly implement CSMA/CA, a collision avoidance system that seems to work pretty well. From my rooftop (downtown San Francisco), I can see 150+ networks yet never experience any symptoms of interference.

Also, in setting the 802.11 limits on power the FCC seems to have overlooked the "skin depth" wavelength/2*pi attenuation factor that limits microwave ranges by a factor exponetially greater than for the longer wavelengths of many cell phones.

I operate a couple of nodes as part of the sflan project and it would be nice if we could reduce the "line of sight" requirement just a tad by boosting the power, say from 200mW to 800mW.

It just ain't fair than big bucks phone comapanies can "buy" an FCC license and blast away at 1500W while us folks trying to provide a free community service have to operate with our hands tied.

Still, what's the good of a home generator, Mr Anderson, if you're unable to find an ISP that works?

Well, something like this might work, as long as enough people have home generators.