Project Meshnet: Hardware for cjdns to enable new, censorship free Internet

[rikur]

With the newsrumors of China starting to censor Bitcoin traffic, I find this very important. And they accept Bitcoins for donations (I just donated 1.79 BTC).

Frequently Asked Questions
What is Project Meshnet, cjdns, Hyperboria and the Seattle Meshnet Project?
Project Meshnet is a colation of meshnet groups that works to promote the development of local meshnets. Cjdns is a networking suite that is commonly used to set up local meshnets, and Hyperboria is what we call the networkof cjdns nodes that weave in and out of both local meshnets and the internet. And on a closing note, the Seattle Meshnet Project is a local meshnet group based out of Seattle that has organized this crowdfunding project.

Why is wireless router support so important, and what does it mean for meshnets?
Once cjdns runs on routers (which means it will run atop the OpenWRT platform), it will be possible to easily and affordably bring Hyperboria in to the average home. Right now it requires somewhat advanced computer knowledge in order to join the meshnet and connect to Hyperboria, but once OpenWRT integration is complete, it will be as simple as plugging in your new router. Our network has grown quickly as it is, but to really bring the mesh to most homes setup needs to be simpler.

How do I know this will even work?
The Seattle Meshnet Project has ran previous group buys and sucessfully paid our current cjdns developer to fix bugs in cjdns. We intend to parlay this experience with paid bugfixing and group buys of hardware into this crowdfunded bugfix, and as with our previous group buy, we'll be distributing hardware and signed posters too.

https://fund.meshwith.me/
http://www.reddit.com/r/Bitcoin/comments/1ewxzw/rdarknetplan_accepting_bitcoin_for_crowdfunding/

Mods: feel free to move the thread to a better sub-forum. I think this is related to Bitcoin due to the China censorship rumors.

[benjamindees]

It seems there is at least one pool that supports mining over cjdns:

https://bitcointalk.org/index.php?topic=88305.0

Here's what I'm thinking needs to be created.  We need a meshnet like this one, which has Bitcoin built into it.  Bitcoin traffic is transmitted for free.  Everything else can be paid for with Bitcoins.  You connect to a node, and pay that node a fixed price to transmit your packets until the next block is found.  If the node starts receiving too many packets, it delays or drops them.  If you need more bandwidth, you connect to more nodes.  If you think the node is delaying too many packets, try a different node next block period.

So it's an auction for bandwidth every 10 minutes, on average.  The software on your computer keeps track of the nodes that offer the best service, and pays the best one(s) given your usage.  Free market competition leads to the correct price.

The network can run over the existing internet as well as purpose-built physical or wireless links.  There is a financial incentive to bypass the existing internet and build our own.

This is a legitimate use of Bitcoin.  It helps build a next-generation, censorship-resistant internet, while making Bitcoin ubiquitous at the same time.

[Etlase2]

I really, really like the idea of meshnets and have done a little reading, but I do not know the specifics. Is it possible for meshnet technology to discover the most advantageous way to get data from one point to another without saturating numerous unneeded hops? I really hope it can take off, but I also really hope that it isn't slow as molasses.

[rikur]

AFAIK cjdns doesn't compromise speed for "extra anonymity" ie. it uses the shortest/fastest path for the traffic.

There are even more hardcore projects like Fennec, although it's just a whacky idea inside some guys head for now: https://github.com/coyotama/fennec/blob/master/README.md

I will try to persuade a Project Meshnet contributor to join the forum to answer questions.. meanwhile, you can ask them in the reddit post linked above.

[Etlase2]

Alternatively, for data such as bitcoin's block chain where many people along the chain may actually want the data, is it possible to send it unencrypted, or make it easy to send unencrypted data so that all nodes wanting a copy of a distributed shared file do not have to transfer, without knowing, the information they are looking for without taking advantage of it?

I hope a clash of meshnet ideas does not stall its progress. It is something that must be accomplished for the sake of a free society.

[jdbtracker]

Agreed.  So how do we connect cities together? I'm asking how do we bypass the ISPs? I heard google is laying down fibre across north america, that's going to be hard to compete with.

it's a bit out there but you got to admit we will need mobile wifi systems to make this work, maybe wifi on quadrotor helicopters interoperable by users who build recharge pads for them?

someone will need to create laser communications as well for rural areas, I've worked with targeting lasers before that's the easy part... building them as home projects is harder alone. We can create terabyte internet connections using the quadrotor mobile wifi, they can swarm areas with high demand and local refueling stations.

http://www.extremetech.com/computing/84266-26tbps-transmitted-with-a-single-laser-could-supercharge-internet-backbone

http://www.theverge.com/2013/5/10/4318404/beamcaster-laser-internet-network-rit-technologies

I like the idea of using bitcoins to pay the system operation costs. so how are you guys planning on hyper implenting your idea? can the system go nation wide with the right know how? how fast?

[tvbcof]

I have two fundamental problems with mesh networks:

1) I suspect that they will be very difficult to engineer to a level which even comes close the the usability and performance of the Internet of 20 year ago.

2)  They would only be necessary to use in a situation where it was extremely subversive to do so.  In that case, the use of any bi-directional communications is going to be putting a giant neon 'arrest me' sign pointing straight to ones' door.  At the very least one will be losing their gear, and will probably be receiving a lot more unpleasantness as well.

[*Bob]

It seems to me that what most mesh development is lacking is incentive for those not invested by ideology, development, etc.

Would it be theoretically or practically possible to develop a new blockchain and *true* alt-coin that would take the same amount of hashing power from each account (node), regardless of CPU/GPU - using shared bandwidth and throughput via mesh node to calculate mining effectiveness?

Being that most IP's do not allow for resale of access, it seems this would be both a great way around that restriction, while providing plenty of incentive to adopt and share access via mesh AND providing a new, true alt-coin that has a creation value most novice could understand (ie.- access to internet)

My terminology may be completely distorted, but hopefully I explained the concept clear enough???

[benjamindees]

someone will need to create laser communications as well for rural areas, I've worked with targeting lasers before that's the easy part... building them as home projects is harder alone.

There is a conspicuous gap in the commercialization of the exact technology needed to enable such a network.  How can this NASA laser transmit 238,900 miles, while the only commercial terrestrial solution I can find will only go 2 km?  Where are the laser tranceivers that will go 500 km?  They can't be that difficult to build.

edit: found this

[tvbcof]

someone will need to create laser communications as well for rural areas, I've worked with targeting lasers before that's the easy part... building them as home projects is harder alone.

There is a conspicuous gap in the commercialization of the exact technology needed to enable such a network.  How can this NASA laser transmit 238,900 miles, while the only commercial terrestrial solution I can find will only go 2 km?  Where are the laser tranceivers that will go 500 km?  They can't be that difficult to build.

edit: found this

Couple of slight problems.  For one, the earth is spherical in case you have not heard.  It also has a gaseous atmosphere...for now at least.

It actually is fairly do-able to have long distance radio links on pretty much a DIY basis, so there is probably no need for sharks and lasers and all that.  But as I say, your link is likely to be visited by a drone within a few minutes...and quite possibly vaporized not long after.

edit: s/drown/drone/g

[ShadowOfHarbringer]

With the newsrumors of China starting to censor Bitcoin traffic

Source ? Last time i checked China happily encouraged people to use Bitcoin on their national propagandist TV.

[rikur]

Couple of slight problems.  For one, the earth is spherical in case you have not heard.  It also has a gaseous atmosphere...for now at least.

It actually is fairly do-able to have long distance radio links on pretty much a DIY basis, so there is probably no need for sharks and lasers and all that.  But as I say, your link is likely to be visited by a drown within a few minutes...and quite possibly vaporized not long after.

I highly doubt that you would get into trouble if you're using the standard WIFI channels for your mesh network. If you do, there are much bigger problems to worry about that Bitcoins / Internet.

[rikur]

Source ? Last time i checked China happily encouraged people to use Bitcoin on their national propagandist TV.

https://bitcointalk.org/index.php?topic=214296.0
http://www.reddit.com/r/Bitcoin/comments/1ewswo/china_now_blocking_bitcoin/

It seems that so far text messages with "bitcoin" in them have been requested to be blocked, but Bitcoin Internet traffic is still running fine.

[mmeijeri]

Meshnet and Bitcoin have great potential for synergy. Meshnet can help provide a decentralised communication infrastructure while Bitcoin can provide a payment mechanism for supporting the nodes and the communication links between them.

[jdbtracker]

someone will need to create laser communications as well for rural areas, I've worked with targeting lasers before that's the easy part... building them as home projects is harder alone.

There is a conspicuous gap in the commercialization of the exact technology needed to enable such a network.  How can this NASA laser transmit 238,900 miles, while the only commercial terrestrial solution I can find will only go 2 km?  Where are the laser tranceivers that will go 500 km?  They can't be that difficult to build.

edit: found this

It's the communication hubs, using what we know from terrestrial observations of the Earths Geoid(earth gravity map) along with gps UTM co-ordinates of base stations,  the laser transceiver is able to find an exact point to target, taking care of the rotation movement, the curvature of the earth etc using a least squares method to account for the movement.

The laser would be targeting a moving target, this target has gps the Receiver that is communicating it's coordinates to the base station through the mesh and being auto corrected by the base station for millimetre accuracy targeting at several km, using this technique I could hit a target 50km away with an error margin of only 5cm, 2 inches; A rural target I can hit from land up to 100km away with 10cm accuracy .

It involves calculating, atmospheric refraction, pressure, temperature, distance to horizon etc to calculate an effective point to further take that 5cm and drop it down to mm.

for those Nasa space shots it requires the entire GPS reference station grid to calculate the stratospheric delay and distortion from all the GPS and Glonass satellites, very complex and very time consuming... but this data is available for free.

I could hit the moon if a I wanted to with a laser so distance is not a factor, beam diffraction is, but that is awesome for communications! It'll get absorbed by the atmosphere if not calibrated properly, and to bypass any danger, just use an infrared laser, trust me the communications part, how to get a message through is the hard part

Radios are a solution as well but we are limited to 5 watts within the city, unless you are on a tower that will only propagate 10km at best, very weak intermittent signal. there are radio spectrums that allow you to use 30+ watts within the city but... who cares... that would give us a 100km range maximum.

[benjamindees]

So it seems like it might be possible to build a 1 GBPS laser link between cities for less than $30,000.

[jdbtracker]

So it seems like it might be possible to build a 1 GBPS laser link between cities for less than $30,000.

lol! much less than that if you take the risk of making the laser grid yourself. I'm thinking of building multiple lasers into one array, they'll be sending info independently, just have to get access to a fibre optics link or a swarm of smaller nodes to feed it city to city.

I figure a smaller laser grid to feed the large array connecting cities.

[benjamindees]

I've been looking into Hyperboria a bit, and ran across this service that plans to support internet access through cjdns:

http://roaming-initiative.com/blog/about.html

"World's first CJDNS VPN ISP"

They accept Bitcoin

Apparently the person to talk to is Rainfly_x on hypeirc.

[BTCLuke]

Why no love for Satellites, bros?

You can get a small payload into space for $10k now, right?

All we gotta do to bridge all gaps worldwide is to have a hub in each metropolitan area that is connected with the satellite... With enough of them up there we could even connect each household directly to a satellite!

[benjamindees]

You can get a small payload into space for $10k now, right?

My understanding is that "into space" at that price means "in space for a few days before it burns up in the atmosphere."