Slashdot Mirror
Lanlink Linking The Coasts
Dan Bricker writes "A guy in Parma Heights, Ohio has a website to promote an idea of linking the east coast to the west coast using standard off-the-shelf 802.11 equipment. He is aiming for a July 4th, 2006 first coast-to-coast ping. This project appears to be totally volunteer based, With no other stated reason than fun with pringle cans and bad weather, and do it just to do it. Can this be done? What real world applications does this have?"
Ok, as we all know there are some pretty desolate regions of the US. Now it would be possible to throw a bunch of routers in the middle of the desert, but they would have to be battery powered or something. The most significant problem would be getting everything to work correctly without even a single down router. Assuming each router covers a tenth of a mile, you'd need about 30,000 routers to make it across the US. Dozens will break or have problems every day, so you'll need at least two per site. That means a total of 60,000 roters. At $100 each that brings the total to $6 million. The battery powered routers for the desert will obviously be more expensive though. Also you'd have to stop people from stealing these somehow, which would be a serious problem.
In conclusion, it would be really hard and really expensive to do this, but it is possible.
Creating ex-temp webs like this might assist insurance adjusters and other computer-needing personnel to work better in emergency hot zones... it would be nice if a company out there started manufacturing the "cans" for emergency use and the FCC made some modifications to the rules for emergency usage ... so every little town could have a few "wi-fi" kits in storage to chain up when a hurricane has leveled everything.. you could also throw some authentication mechanisms on the idea and build a quick "emergency VoIP network" the same way.
Just a thought from the thoughtbrew:
www.bigattichouse.com
meh
Something like 70% of internet backbone is owned by half a dozen companies. The RIAA & co are putting increasing pressure on businesses and universities, and backbone providers may be next.
The Government is, frankly, outright hostile to many forms of free expression, and some basic civil rights we've come to take for granted (abortion rights, for starters, never mind the Bill of Rights).
This project may teach valuable lessons about using open standards to form a non-owned, alternative internet backbone.
Hexayurt - open source refugee shelter,
I have never heard about Hands Across America (probably because I was 7).
Anyway, here's a link for those that were drinking out of juice boxes in 1986.
http://eightiesclub.tripod.com/id248.htm
It's hard to believe that such a thing was possible.
A speech...
The real world application is, perhaps, psychological: getting people to realize that with a bit of effort each, we can all be networked to each other at high speed WITHOUT paying some company OR government for the privelege of just moving data around using equipment we own and airwaves that belong to everyone.
It depends on what the gain and such is of the antenna. With an omni, maybe, but with a wave-guide cantenna you are probably safe.
See here for details
Besides, I think this is definately more doable that hands across america. With the possible exception of the rockies/cascades etc, just set up some cantenna's, and aim it off into the horizon. With GPS and such, it should be easy to coordinate. A handful of people at each horizon, should do it... How far away is the horizon anyways? I know I can see the buildings in downtown from here, and its like 20 miles from here.
"Second Internet" - Man, how cool would that be... a completely underground (reverse pun intended... gettit... it's overgrou... oh never mind) Internet, detatched from the 'real' one...
In case of an all out war, the 'real' internet may be shut down, but this air based one could keep on keeping on... although without electricity after the war, only as long as all the laptop batteries lasted... so really only about 1 hour after the strike... just long enough for the users to start a thread:
"Woh! What was that?"
"Dunno... kinda bright though"
"Dude... I think this is bad"
"Yup"
"BBFN"
So...um...this would be called LANs Across America?
/me ducks and runs for cover.
Ron
No more pringles cans.
I was recently involved in a fairly casual discussion of how to create a WAN link between computer labs at two different campuses of a university in Ghana. The main campus, in the capital city of Akra (sp?) has a limited satellite connection to the Internet costing something around a few thousand a month, supposedly. None of the other three campuses have or can afford a similar connection. This isn't a big enough gateway to share WWW access, but a WAN could allow Intranet and Internet-based email, as well as Intranet sites, file sharing, and perhaps even VoIP to augment the poor phone systems.
So the big problem was how to set up this connection. The telco system apparently isn't too good; only around 400 new lines are added per year, so getting ahold of a large number of leased lines would be virtual impossible. Obviously, setting up an independent wired backbone is financially out of the question. So we started toying with the idea of a WiFi link, which seemed like the only possibility.
The only problem is that if we are trying to set up a 200km link (between the main campus and one in the north; I don't recall the name of the city) we would need repeaters in some remote areas without consistent power, not to mention having to plot good line-of-site and build fairly secure base stations. What we realised was that we could attempt to piggyback the existing private cell-phone infrastructure. There is a cell system spanning the north and south, which means a stable backbone, on which we can either rent data bandwidth (probably expensive) or, better yet, on who's repeater stations (probably microwave antennas) we could rent physical space.
Our informal conclusion was that the University should consider renting space on repeater stations for their own WiFi hubs and create a WAN using long-distance line-of-site connections with off-the-shelf, inexpensive WiFi components. Projects like this pave the way to practical, inexpensive applications of WiFi technology.
Are we talking about a giant, shared, 10mbs pipe across the US that we could all use together? Wow, that would really last for at least 10 seconds. Talk about /. effect....
that was one example of a network whose structure could handle host disconnects. also freenet, which has redundancy built into its design. and gnutella, as you point out.
all of these essentially use P2P as their structure, but fidonet and freenet remind us that P2P-the-structure has a far wider range of uses than just downloading mp3's. right now the internet dominates "cause it's there" but even its structure was historically envisioned (by some, anyway) as much more decentralized than it is now. as it moves toward centralization it becomes increasingly unsatisfactory for many purposes, and momentum grows to build and use alternative, decentralized structures.
Standard 802.11b ethernet won't get this accomplished. There's no way. Can a packet ping from the east to west and back again in 255ms? with Wifi?
I don't think so. I think we should observe the way
Amature Radio Operators have ran packet radio stations. We'd need to write drivers that would
emulate a packet radio connection. There's will be
too many hops to implement a 802.11 WiFi solution.
We would have to go with packet switching.
We'd be able to use WiFi hardware, but all the drivers would need to be written to emulate packet switching.