Slashdot Mirror
Hong Kong Protesters Use Mesh Networks To Organize
wabrandsma sends this article from New Scientist:
Hong Kong's mass protest is networked. Activists are relying on a free app that can send messages without any cellphone connection. Since the pro-democracy protests turned ugly over the weekend, many worry that the Chinese government would block local phone networks. In response, activists have turned to the FireChat app to send supportive messages and share the latest news. On Sunday alone, the app was downloaded more than 100,000 times in Hong Kong, its developers said. FireChat relies on "mesh networking," a technique that allows data to zip directly from one phone to another via Wi-Fi or Bluetooth. Ordinarily, if two people want to communicate this way, they need to be fairly close together. But as more people join in, the network grows and messages can travel further. Mesh networks can be useful for people who are caught in natural disasters or, like those in Hong Kong, protesting under tricky conditions. FireChat came in handy for protesters in Taiwan and Iraq this year."
Interesting! I first heard that idea from David Brin, who was proposing it as something to be used for disasters.
http://davidbrin.wordpress.com...
Maybe the governent of Hong Kong qualifies as a disaster.
http://www.geoffreylandis.com
FireChat requires that users create an account online (with an email address) before they can use the app. This and the lack of encryption limits its usefulness.
Mesh networking, peer-to-peer, power to the decentralized people -- it all sounds great. But some of those people will still be on the side of the government. I wonder how much information one mesh node could accumulate to incriminate other participants? How many of "the people" will be willing to participate in an uprising like this if they know that a government stooge is likely no more than two or three hops away?
The wireless networking research community has been working on mesh/ad-hoc networks for over a decade, citing communication in disaster areas as (one of the) main applications. At some point some people started to sort of laugh at it ("oh look, another mesh networking paper!"), because despite all the research it didn't seem to get any closer to reality. My guess would be that the reason why we're seeing it finally being used is because in order to be feasible, you need the density of devices to be above a certain threshold, which means a) it was never going to work in the pre-smartphone era -- with smartphones, you can just download an app to do it, but otherwise you'd pretty much need to spend major $$ to get the necessary number of dedicated devices out there, or else there needs to be wide-scale agreement to implement a specific protocol on all new devices, which was never going to happen because it's not a selling point, b) it won't really work in major natural disasters, because, well in order to maintain the density of devices, a large number of people need to have continuous access to power, which is unlikely if a disaster is so severe that communication infrastructure is offline (I imagine celltowers are less fragile than power lines).
weinersmith
Well, speaking from experience in the Japan 2011 earthquake, you are kind of on the mark kind of not.
b) it won't really work in major natural disasters, because, well in order to maintain the density of devices, a large number of people need to have continuous access to power, which is unlikely if a disaster is so severe that communication infrastructure is offline (I imagine celltowers are less fragile than power lines).
After power was turned back on, I, and a lot of other people, went out and bought a hand-cranked USB charger(also doubles as a flashlight and radio, a handy device to be sure). It doesn't take that much energy to power a cell phone.
As for the tower issue, the towers where I was at(Tsukuba, which is about halfway between Tokyo and Fukushima) all kept power even after the quake but since so many people were using their phones to either call people or check the news it was almost impossible to get through(the bandwidth of the tower may have very well been degraded as well). A mesh network *might* have been useful there, but it would have had to have enough density to work. Really the biggest problem with using a mesh network for disaster is that anywhere you have enough people to support a mesh network, you could probably just as easily use a bullhorn to communicate.
Monstar L
Firechat uses Apple's Multipeer connectivity for IOS, and a similar protocol for Android, to achieve a mesh network. I do not believe that any of this is MANET (the IETF's favorite mesh networking protocol).
I am no expert in mesh networking, but I was under the impression that addressing in them does not scale well. The best technique seems to be BATMAN [1]. AFAIU it requires everynode to perform a full broadcast regularly and that each device stores a complete routing table to each other device. That will not scale to build a city wide network.
Somebody knows more?
[1] http://en.wikipedia.org/wiki/B....
This is easy to defeat with a simple 2.4ghz jammer in the protest area. Both Bluetooth and most WiFi would be disabled. So the devices cannot mesh. Turn off the cell networks and ability of protesters to coordinate is gone.
So it could be useful when Government is not the adversary such as in a disaster, but is easily disabled by Government if that is it's intention during protests.
Facebook is billions of individual "Skinner Boxes." And if you use it you are the pigeon!