Slashdot Mirror

wouldn't it be cheaper to setup multiple 802.11 b/g/n channels running 20-30km hops between nodes using dishes?

Like these guys

Optic has some great benefits of wide bandwidth and less prone to environmental factors such as storms but it is very expensive and time consuming to dig and setup. Wifi nodes can literally take a few minutes to setup and test and are configured for multiple points of failure redundancy. And solar is very lucrative seeing as these areas are the farmlands where sun is abundant. Running this equipment away from the grid would be trivial at best.

I guess something along the lines of http://www.wafreenet.org/ is what is really being suggested (sorry if I've missed someone else mentioning it). I know a few people part of this network, and aside from the occasional expected segregation, it's rather well run.... Albeit small -Keith

I wish it were so...

Therre have been and still are guerilla wireless networks, and as you say, we are the value in the Internet, but we are not valuable when we question and create the content ourselves.

In the early days of the internet, the freenet (yes, it has been around that long) was almost as valuable as the non-free one. Now though, expectations are different, and freenets can't grow fast enough to be anything more than a pallid and dated reflection of the main act.

There was a big effort to do something similar in Western Australia with wafreenet. They've put together a massive wireless resource collection at E3. It'd be worth having a good browse around there.

We had this problem on the WaFreenet, so we set about creating some software to fix it.

The result was frottle. It's a bit of a kludge, but essentially provides a virtual token bus over ethernet. It runs at the wrong layer (UDP), but is suprisingly effective. Before, with 14 clients to the HillsHub AP (many clients in the 10's of kilometers), we'd get crippled throughput rates below 10kB/sec. Now multiple users can sustain data rates above 80kB/sec (or better depending upon load).

you mean....like frottle ?

I'm involved in a wireless community freenet in Western Australia, and we've been suffering from the hidden node problem.

We investigated Karlnet, but found the software expensive, and only supported on old kernel versions.

As a result, we started experimenting, and have come up with our own solution: frottle.

We've been running it for a while now, and are still tweaking some parts of it, but it's been working extremely well.

Check out the frottle page for more info.

I'm involved in a wireless community freenet in Western Australia, and we've been suffering from the hidden node problem.

We investigated Karlnet, but found the software expensive, and only supported on old kernel versions.

As a result, we started experimenting, and have come up with our own solution: frottle.

We've been running it for a while now, and are still tweaking some parts of it, but it's been working extremely well.

Check out the frottle page for more info.

I'm involved in a wireless community freenet in Western Australia, and we've been suffering from the hidden node problem.

We investigated Karlnet, but found the software expensive, and only supported on old kernel versions.

As a result, we started experimenting, and have come up with our own solution: frottle.

We've been running it for a while now, and are still tweaking some parts of it, but it's been working extremely well.

Check out the frottle page for more info.

...it provides heaps of motivation for
C
O
M
M
U
N
I
T
Y
Wireless Networks. And the government finds it pretty difficult to argue against them.