When the Internet dies, meet the meshnet that survives • Hacker News Board

Re: When the Internet dies, meet the meshnet that survives

#11

Post by krapp » Sun, Apr 20, 2014, 6:59 PM UTC

pessimizer wrote: post #5

krapp wrote: post #4

Earlier quoted context omitted.

>Perhaps it's a good idea to have random take downs of the internet in large areas for extended periods of time so that there is more motivation for people to create a second, non-centralized single-point-of-failure single-point-of-control internet. That's not a good idea. That's a very bad idea, considering how much critical infrastructure, globally, depends on the internet working -- what you're suggesting, to even…

>considering how much critical infrastructure, globally, depends on the internet working This is the very bad idea.

Perhaps. But solving one evil with a greater evil can't be a valid solution.

Re: When the Internet dies, meet the meshnet that survives

#12

Post by Paul_S » Sun, Apr 20, 2014, 8:51 PM UTC

wcummings wrote: post #6

What on Earth makes people think a finnicky meshnet that requires a high density of nodes will survive a Sandy-like disaster? Doesn't seem like the most obvious use-case for meshnets, wouldn't people be better served getting ham licenses and learning how to improvise antennas in an emergency?

That depends on the types of devices you use for creating it. You can get a km out of an ordinary router with a directional can antennae. What distance do you think is required for it to be viable?

Re: When the Internet dies, meet the meshnet that survives

#13

Post by wcummings » Sun, Apr 20, 2014, 9:23 PM UTC

Paul_S wrote: post #12

wcummings wrote: post #6

What on Earth makes people think a finnicky meshnet that requires a high density of nodes will survive a Sandy-like disaster? Doesn't seem like the most obvious use-case for meshnets, wouldn't people be better served getting ham licenses and learning how to improvise antennas in an emergency?

That depends on the types of devices you use for creating it. You can get a km out of an ordinary router with a directional can antennae. What distance do you think is required for it to be viable?

>You can get a km out of an ordinary router with a directional can antennae

With a line-of-sight and some careful positioning, maybe.

There are numerous examples of people using traditional radios to coordinate disaster response [1] (actually why ham licenses are issued IIRC), this seems like a more complicated, error prone, cool-kid-hackerspace solution. What advantages does a mesh network have over traditional radios (which have bands allocated for this) for disaster response? How well will a mesh network hold up if half (or more) of the nodes fail due to power loss, flooding etc?

Mesh networks are cool, and I'm all for decentralizing communication, but their usefulness for disaster preparedness seems very dubious to me.

[1] http://en.wikipedia.org/wiki/Amateur_radio_emergency_communi...

Re: When the Internet dies, meet the meshnet that survives

#14

Post by kwhitefoot » Mon, Apr 21, 2014, 9:00 AM UTC

This is what Internet was supposed to be from the beginning; a reliable network with multiple paths between nodes so that the removal of even a large number of edges would not prevent packets getting through.

Re: When the Internet dies, meet the meshnet that survives

#15

Post by kwhitefoot » Mon, Apr 21, 2014, 9:02 AM UTC

Actually for low rate communication in a disaster zone store and forward text messaging via mobile telephones would surely work more easily and could be combined with the mesh to enable greater coverage.

Re: When the Internet dies, meet the meshnet that survives

#16

Post by simplicissimus » Mon, Apr 21, 2014, 7:44 PM UTC

wcummings wrote: post #13

Paul_S wrote: post #12

Earlier quoted context omitted.

That depends on the types of devices you use for creating it. You can get a km out of an ordinary router with a directional can antennae. What distance do you think is required for it to be viable?

>You can get a km out of an ordinary router with a directional can antennae With a line-of-sight and some careful positioning, maybe. There are numerous examples of people using traditional radios to coordinate disaster response [1] (actually why ham licenses are issued IIRC), this seems like a more complicated, error prone, cool-kid-hackerspace solution. What advantages does a mesh network have over traditional radi…

Ongoing innovations in open firmware like OpenWRT allow deployment of reasonably robust routing schemes, such that the mesh nodes themselves compute the routes most efficient for transport (and also repair those routes when nodes go down). Groups like Commotion Wireless, Freifunk, Funkfeuer, and LirbeMesh all deploy firmwares derived from OpenWRT to this purpose. The actual capacity, range, and sensitivity of the radio hardware used, whether unlicensed or not, will still be gating factors. Along with ambient noise levels on whatever bands are used.