Three projects to create a government-less Internet, and why it is needed. | Hacker News Reader

Three projects to create a government-less Internet, and why it is needed.

datelinezero.com

81 pointsnika23 comments

Here's what I think the citizens of countries in this situation need to do:

    - Coordinate activism without leaking information to authorities
    - Get the word out to the rest of the world about what is happening
    - Restore the ability to browse the rest of the Internet.  Make
      available the viewpoints from the rest of the world as an alternative
      to government propaganda.

In order to facilitate these three aims, I propose the establishment of new or adaptation of existing Open Source projects:

    - A DIY packet radio uplink kit.  
    - A one-click install for installing a Fidonet node on an older laptop 
      with a modem
    - Inexpensive WiFi-based "Text messaging" handsets.  
    - Open Source software to convert netbooks and smartphones into encrypted 
      communication devices
    - Weatherized WiFi communications cells which can be hidden, plugged in, and 
      abandoned (With some UPS capability to give immunity to power interruptions.)
    - A low cost "pirate cell tower" which can be used to re-enable SMS comms
      on conventional handsets when the government deactivates mobile service

In addition, I think it would be a good to found an international volunteer organization which designs and prototypes relevant hardware, with the aim of getting donations over the network through sites like Kickstarter and reddit and using the proceeds to rapidly construct and disseminate such hardware to affected citizens.

EDIT: Another hardware thing: A plan for acquiring 100's of cheap, outdated GSM handsets which can be reprogrammed to implement secure encrypted messaging over SMS.

EDIT: Yet another hardware thing: a drone suspended from a balloon, equipped with a long-duration battery, an altimeter, wireless, GPS, and aimable directional laser comms. Ideally, one would be able to give it the the altitude and GPS coordinates of another "laser link drone" and automatically aim the laser to establish line of sight laser comms to provide a long-distance network link.

The tower-devices will be broadcasting their presence and location, so hiding them isn't as important as making them expensive and time consuming to remove or highly mobile (on UAV's or back pockets) to run away from the government.

Who said anything about towers? Hiding them in attics and rooftops is precisely about making them expensive and time consuming to remove. Triangulating the precise location of such devices carries a certain cost to the authorities during a crisis. They will have limited equipment and manpower to conduct such searching as well as a need to provide security for the same.

So what if it takes them just two days for something to be removed? Those two days of comms for the activists have been highly valuable, and there should be many other units still operating and hopefully others in reserve to activate.

I think there is a place for mobile solutions, however. How about suspending equipment with enough power to operate for a week from weather balloons anchored by monofilament? Set up a sensor, such that an unauthorized drop in altitude will trigger releasing the anchor? This way, if the authorities are trying to reel the unit back in, it will at least have another 1/2 hour of transmitting time. I'd also add some switches to trigger erasure of the keys for the encrypted hard drive and overwriting memory in the even of unauthorized opening of the case.

> The tower-devices will be broadcasting their presence and location

To a certain extent, yes. You could use highly directional 802.11 antennas (yagi or shotgun) and disguise them. If they have minimal side lobe bleed, they may be indistinguishable from the normal AP background levels. You should be able to get 1-5 miles out of such a link. A few dozen such links could criss-cross a city such as Cairo.

With a coastal city like Tunis, there's the possibility of setting up comms with ships at anchor carrying hefty satellite equipment. Old fashioned satellite dishes could extend the range of such links to many dozens of miles.

I wonder about combining this idea with my idea of a balloon drone with equipment for aiming an antenna/laser link?

There was also a wireless networking technology developed at HP using the old Ethernet protocols, but instead of a wire, a patch of ceiling illuminated by infrared LED carried the signal instead. A network using such a technology and the side of a skyscraper could easily cover a large swath of a city and would be hard to detect.

Solar-powered medium-altitude flocking UAVs could provide a semi-temporary mesh WiFi (or sim.) network. We have flocking software. We have UAVs.

Is anyone researching solar-powered UAVs that can stay aloft for long periods of time, inexpensively?

Quick google only found this: http://www.gizmag.com/go/3891/

Solar powered flight is not easy/cheap. The first through the night solar flight was only pulled off this previous summer by Solar Impulse (it was manned). It's a -big- plane (width wise anyways). Its wings are as wide as a jumbo-jet's. You're not going to save much weight making it a UAV either, since I suspect most of the load is actually batteries.

On the side, it's -friggin- cool. Here's the Solar Impulse site: http://www.solarimpulse.com/index.php

A government less internet already exists, it's called Packet Radio AKA AX25. It's a network that spans the globe with gateways to the internet at various locations operated by radio amateurs using FSK modulation on regular HAM equipment. Bandwidth is quite restricted.

The problem with creating a government-free internet for mass use is one of resources and determination as well as some technical difficulties in operating networks like this with large numbers of nodes. In rural areas the problems will be the longer stretches without connectivity as well as all the issues that come with routing packets around the world without network congestion if the number of nodes should get large or if something exciting were to happen.

On such a network the communications should be limited to the essentials, rich media is pretty much out of the question.

On another note, the frequency spectrum is highly regulated, powerful base stations operating as hubs could be shut down by triangulation and force and operating transmitters outside of the alloted frequency ranges is forbidden by law.

The only part of the electromagnetic spectrum that does not currently need any regulatory before equipment can be sold or operated is as far as I know > 10GHz, and even then there are strict limits on power. Other options are UWB: http://en.wikipedia.org/wiki/Ultra-wideband and optical links (for instance lasers).

In the United States the current frequency allocation chart looks like this:

http://en.wikipedia.org/wiki/File:United_States_Frequency_Al...

Setting up a 'government free' internet for a large number of users and without using existing infrastructure is a serious technical challenge, especially if you want to operate it without breaking the law.

And you can be sure that in the areas of the world where interference free internet is most needed the laws will ban the sale or ownership of any such devices.

This is why DIY kits/plans/software are valuable.

In rural areas the problems will be the longer stretches without connectivity

The most pressing need for mobile comms and richer media will be in the cities, where this is less of an issue. For getting news out to less populated areas, something like Fidonet could work as a way of disseminating alternative news.

On such a network the communications should be limited to the essentials, rich media is pretty much out of the question.

What about USB flash drives carried by drone aircraft or carrier pigeon?

If one can rapidly disseminate many units, then infringement is not so much of an issue. Protesters are already engaging in civil disobedience, and the authorities already have a lot of other chaos to deal with during a crisis like we are seeing in Tunisia and Egypt.

Other options are UWB and optical links (for instance lasers).

Relevant to both Tunisia and Egypt would be hardware to establish Line of Sight communications with vessels offshore carrying satellite uplinks. Infrared lasers linking with boats 10 miles offshore would be hard to detect.

EDIT: WiFi has been transmitted a distance of 237 miles. http://www.wi-fiplanet.com/news/article.php/3695986/New-Wi-F...

Unamplified WiFi has gone 125 miles.

http://www.engadget.com/2005/07/31/unamplified-wifi-distance...

First, radio amateurs are regulated everywhere more than the current internet is.

Secondly, there is zero privacy in amateur radio communications. Encryption of any form is expressly forbidden.

Thirdly, commercial use of the radio amateur service is severely restricted as well.

Traditionally, in times of unrest and disaster radio amateurs were the last to stay in touch and the first to come back in contact.

Yes, they're severely regulated and there are governmental registries of the equipment they have. But I'd still put good money on HAMs being the last links to an area engulfed in unrest or disaster.

I was suggesting that A government less internet already exists, it's called Packet Radio AKA AX25 is not really true.

In Greg Egan's[1] latest novel, "Zendegi", the uprising people use cellular handsets reprogrammed into a peer-to-peer mesh network to continue email, messaging, and phone calls when the authorities shut down the cellular towers.

The hardware is already deployed in people's pockets and purses. It just takes a software download to make it happen.[2]

[1] Greg Egan is your man if you want to read a sci-fi book about, say, "given that much of physics was derived by viewing the sun, moon, and stars, how would a civilization without any knowledge of an external universe develop physics?"

[2] You do have to take full control of the phone's radio, which is non-trivial. They try to prevent that.

As much as we wax lyrical on the Internet being without borders and transcending governments, it's sobering--and important--to see just how quickly and easily a government can enforce its will.

Any networked system has a large number of vulnerabilities:

1. Physical hardware. You combat this in three possible ways:

a) Make the hardware as cheap as possible;

b) Make the hardware dual-purpose. For example, behind the Iron Curtain, they would modify radios beyond what the state allowed them to listen to; and

c) Use hardware that the state can not do without. For example, use of the telephone network by spies in occupied Europe during WW2.

2. Wired networks are inflexible and easily interrupted. It wouldn't take too many acts of sabotage to almost completely disable the Internet in the US. Wireless networks are subject to state tracking. You combat this in a number of ways:

a) Use a medium the state can't do without (eg power lines, although in extreme circumstances the state can and will cut the power);

b) Hide your communication channel (eg steganography);

c) Use a wireless medium where the transmitters are small, cheap and portable.

Note: in this context, carrying messages using people constitutes a wireless packet-switched network with high bandwidth and high latency, which solves the problems of cost and partially solves the problem of tracing as the state can still use armed forces to limit movement.

It's worth discussing encryption in particular. It is not a sufficient protection but it can help. For one thing, an intercepted encrypted message will be taken as a sign of guilt in extreme circumstances regardless of its content.

Also, the distribution of encryption keys is an age old problem.

Lastly, encryption has a human element. You don't really know who is receiving your message and what they're doing with it. They could be a government agent, a turncoat or simply surrender their secrets under duress.

So I'm not sure this is a problem that can be solved but it can be easier.

Has anyone been exposed to Wifi Direct? This is Wifi peer-to-peer sharing without access points. Access points as they function now would be perfect for connecting user handsets/devices but something like wifi-direct could be modded to run on access points and to serve as a cheap backhaul link protocol to interconnect APs.

The wifi spectrum is unlicensed, at least in this country (for now), so modding cellular handsets to "mesh" with each, assuming that someone were able to create a viable mesh protocol, would have handsets broadcasting in licensed spectrum without permission from the FCC. Probably something to avoid...

There are "regular" OpenWRT Wireless Battle Mesh meetings: http://battlemesh.org/BattleMeshV4

Open source router firmwares like DD-WRT or Android handsets would be good platforms for some kind of mesh network.

Having a wireless AP with a webserver and forum software would be a simple option for communicating with your immediate neighbors (who could communicate with theirs).

The global internet is not required to co-ordinate protests. Something country-wide (or nearly country-wide) is.

It wouldn't even need to be fast.

How about this: all wifi routers sold will have to also contain a low-bandwidth public mesh interface that cannot be configured or turned-off.

That means that as people buy wifi routers, they also set up a very-low bandwidth p2p net around their house. Any person with independent net capabilities -- say those using a sat connection or packet radio -- would naturally provide pass-through very low bandwidth to lots of other folks without having to set anything up (Without being able to help it, actually)

Not a perfect solution, but something that can be done easily from this end and it would cause a lot of consternation among those trying to totally control information flow.

or you can just walk those 50-100m (average WiFi coverage area) and ask your neighbors for news.

Scientific American had a piece on how to negotiate stable file transfer over risky connections with redundancy. The problem of file transfer over ptp cellphone is that you don't know which of your packets will make it through and which ones don't. The transmissions need to be Asynchronous (transmit and forget) because you don't know if the distant wireless device will hear you or not. Scientific American magazine, "Using Smart Redundancy to send messages"

http://i56.tinypic.com/332s5lv.jpg

Reply on news.ycombinator.com