Homebrew Wireless Servers...a grassroots approach?

[271608]

Would it be possible to -- using todays technology -- set the internet up to run from homebrewed Wireless Servers thus eliminating the necessity for Internet Providers?

Has anyone ever heard of anything like this?..sort of like what ham radio is to AM/FM...hey, it could be called 'hamnet'

I've never messed with ham, but I would definitely tinker with something like what I'm talking about.

[lupin]

If you actually wanted it connected to the Internet you would have to make that connection through some sort of Internet service provider, but I have heard of community wireless networks being setup in a similar manner to what you describe.

[Sniffing4Prison]

Would it be possible to -- using todays technology -- set the internet up to run from homebrewed Wireless Servers thus eliminating the necessity for Internet Providers?

Has anyone ever heard of anything like this?..sort of like what ham radio is to AM/FM...hey, it could be called 'hamnet'

I've never messed with ham, but I would definitely tinker with something like what I'm talking about.

There was a paper where this was discussed as "The day of decentralization" (which also refers to wireless everywhere). The issue is mostly legality and ease of use.
The paper is called "Interconnecting the Network of Networks". It discusses ISP monopolies, game theory, and All of that great stuff =)

[271608]

@ lupin

Ahh, ya I guess you would have to host your own sites...It would be fun to learn more about networking from -- literally -- the ground up.

@ Sniffing4Prison

It seems interesting, but I don't know about all that Econ talk...still worth a looksy though.

[Jac01]

Look into mesh networks... then from there, have everyone host their own web servers.
<sarcasm>
That would work out great... have my gmail/vpn/my bank account info flowing through my local black-hats router/comp... not like that would be a MITM waiting to happen or anything.
</sarcasm>

[Thorn]

Would it be possible to -- using todays technology -- set the internet up to run from homebrewed Wireless Servers thus eliminating the necessity for Internet Providers?

Has anyone ever heard of anything like this?..sort of like what ham radio is to AM/FM...hey, it could be called 'hamnet'

I've never messed with ham, but I would definitely tinker with something like what I'm talking about.

Hams have had packet radio running TCP/IP since at least the mid-nineteen seventies.

In addition, the idea of a 'shadow', RF-based Internet, dates back to at least the mid-nineties, when famed RF hacker Brian Oblivion (of the L0pht) was concerned about control of the Internet wrote the original "Guerrilla Net" papers. Guerrilla Net detailed the requirements for an independent, RF-based TCP/IP network. By searching you can probably find most of Oblivion's papers.

There's one big problem with this idea though, and that is with high speed backbones as currently connect the Internet cannot be replicated with RF. Long distance with RF requires certain frequencies that will travel the required distances. However, due to the physics of RF, the frequencies that will travel long distances cannot have high bandwidth. Frequencies that will conduct high bandwidth cannot travel long distances. So unlike with wire and fiber, you cannot have long distance, high bandwidth backbones. You can have short, interlinked backbones, but it introduces a huge amount of hops and latency, especially when compared to wire and fiber backbones.

[mahdimc]

Might I suggest (maybe OP suggests it too) that instead of using this "Ham-net" to be a long path to the known internet, just have a bunch of systems networked together?... sort of, say, a wireless ARPAnet? Just you and the neighborhood buds running your own services on your own machines. Those machines can then be interlinked to other friend's machines, then eventually... you can have a *very* small scale model of a "wireless internet" (not just a wlan), where packets from client 1 would hop through client 2 since client 3 is too far away... etc.

Would be a nice project... I support this idea.

[kidFromBigD]

Hams have had packet radio running TCP/IP since at least the mid-nineteen seventies.

In addition, the idea of a 'shadow', RF-based Internet, dates back to at least the mid-nineties, when famed RF hacker Brian Oblivion (of the L0pht) was concerned about control of the Internet wrote the original "Guerrilla Net" papers. Guerrilla Net detailed the requirements for an independent, RF-based TCP/IP network. By searching you can probably find most of Oblivion's papers.

There's one big problem with this idea though, and that is with high speed backbones as currently connect the Internet cannot be replicated with RF. Long distance with RF requires certain frequencies that will travel the required distances. However, due to the physics of RF, the frequencies that will travel long distances cannot have high bandwidth. Frequencies that will conduct high bandwidth cannot travel long distances. So unlike with wire and fiber, you cannot have long distance, high bandwidth backbones. You can have short, interlinked backbones, but it introduces a huge amount of hops and latency, especially when compared to wire and fiber backbones.

It should be a rule here at the RE forums: When Thorn speaks, the thread can be considered closed since his ideas on the matter are correct/insightful/seasoned-with-wisdom 100% of the time, even if he tells someone to FOAD.

Thorn, thank you for sharing, and I hope to add just one more thought: This is precisely why the recently cleared 700MHz specturm in the USA drew so much attention (and dollars) when the FCC auctioned it off. The 700MHz band is well suited to penetrate into buildings and structures where current 1900MHz and Wifi simply cannot. Yes, the Verizon LTE bandwidth it will carry is lower by virtue of the base frequency, however build-out investment (CAPEX) should be less since towers and cellsite can be spaced further apart.

Apologies to all if I'm rambling or off topic.

[Snayler]

Might I suggest (maybe OP suggests it too) that instead of using this "Ham-net" to be a long path to the known internet, just have a bunch of systems networked together?... sort of, say, a wireless ARPAnet? Just you and the neighborhood buds running your own services on your own machines. Those machines can then be interlinked to other friend's machines, then eventually... you can have a *very* small scale model of a "wireless internet" (not just a wlan), where packets from client 1 would hop through client 2 since client 3 is too far away... etc.

Would be a nice project... I support this idea.

Uh... Ad-Hoc Networks, maybe?

[Thorn]

Might I suggest (maybe OP suggests it too) that instead of using this "Ham-net" to be a long path to the known internet, just have a bunch of systems networked together?... sort of, say, a wireless ARPAnet? Just you and the neighborhood buds running your own services on your own machines. Those machines can then be interlinked to other friend's machines, then eventually... you can have a *very* small scale model of a "wireless internet" (not just a wlan), where packets from client 1 would hop through client 2 since client 3 is too far away... etc.

Would be a nice project... I support this idea.

What you're describing is a mesh network. Mesh networks do work to some extent, but they do suffer from the problems that Jac01 described.

Look into mesh networks... then from there, have everyone host their own web servers.
<sarcasm>
That would work out great... have my gmail/vpn/my bank account info flowing through my local black-hats router/comp... not like that would be a MITM waiting to happen or anything.
</sarcasm>

It should be a rule here at the RE forums: When Thorn speaks, the thread can be considered closed since his ideas on the matter are correct/insightful/seasoned-with-wisdom 100% of the time, even if he tells someone to FOAD.

Thanks!

Thorn, thank you for sharing, and I hope to add just one more thought: This is precisely why the recently cleared 700MHz specturm in the USA drew so much attention (and dollars) when the FCC auctioned it off. The 700MHz band is well suited to penetrate into buildings and structures where current 1900MHz and Wifi simply cannot. Yes, the Verizon LTE bandwidth it will carry is lower by virtue of the base frequency, however build-out investment (CAPEX) should be less since towers and cellsite can be spaced further apart.

Apologies to all if I'm rambling or off topic.

There are a lot of emerging RF technologies, and we'll have to see which actually take off. The market and usability will be the determining factors.