The Sup Forumsternet project

### Sup Forumsternet - Seriously get on the IRC and start building edition ###

Mark yourself on the map with your contact info (can be a cock.li throwaway if you don't want to dox yourself)
zeemaps.com/map?group=2787145

IRC (SSL): irc.jollo.org:9999 #gternet
Wiki: mesh.gentoo.today
Git: git dot jollo dot org/gternet
Last thread: What is Sup Forumsternet?
Sup Forumsternet would be our own decentralized p2p network running on our own infrastructure, totally disconnected from the clearnet.
It will be a combination of DIY meshnets on a city wide basis, and a platform for interconnecting these city wide meshnets over p2p openvpn bridges and high powered wireless dishes for City to City relays.
The Inter-Meshnet part of the project will allow any existing DIY meshnets to join our network by adding a bridge node or a city to city relay.

-The short term goal is to have a few anons set up their own local meshnet and link them together using the bridges as a proof of concept.
-A long term goal is gain a large enough user density that we can disable the brige nodes and have our own decentralized network idependent of the internet.

Currently for the local and city wide meshnets we're going to use batman-adv. For the p2p internet bridges we are going to use a trust based system of key exchanges, mostly done by direct contact between node operators
The network will use IPv6 throughout, we have an issue on our git that you can vote on (after registering) to determine the address space we will use. On git you will find scripts to help automatically deploy openvpn intermesh bridges, and batman meshnets.

This project is almost two weeks old, so we need all the expertise and help we can get.
Every contribution is valuable!
Gather a group of friends and join the IRC and wiki.

Other urls found in this thread:

git.jollo.org/gternet/software/gternet-cli.git
airlink.ubnt.com/#/ptp
youtube.com/watch?v=zXTEWFb_6w4
turnpoint.net/wireless/has.html
twitter.com/NSFWRedditImage

Currently under development:
- A tool called geternet-cli that will tie all the deployment scripts together into one place. (knots)
- A hash based system to assign addresses across a decentralized network (cat is looking for more input on this; if you have any comments/questions direct them towards him in irc).
- A simple GUI for those inexperienced with the linux command line.

To Do:
- decide on hardware standards for City to City relays
- create resources for newbies on the wiki
- gather more people
- start building internet bridges for the intermesh connections (Linux Sysadmin tier)
- start designing diy kits for meshnet end users (Should be newbie-midlevel friendly tier)
- look into croundfunding options for purchasing city to city relay backbone equipment (Richfag tier)

Things NOT To Do:
- Talk about logos
- Talk about names

Just a reminder- once you have access to the gitlab project you don't need access on a repo basis

Network architecture diagram

output of:

sudo batctl o

after using mesh setup script on 3 devices

whoops wrong one, i dont seem to have the one from the previous thread saved
all well you get the idea

Anyone wanna give a newbie the rundown of what we got going so far?
We using VPN? How secure is this? Etc?

what about laying wires or hijacking some existing infrastructure?

Basically you set up your distribution point using a BATMAN script that was put up on the gitlab site, run the network distribution software, and throw on an AP to the network.

The said server uses a OpenVPN to join the current mesh network infrastructure.

Then throw yourself on the map that is in the wiki, hangout in the IRC with us and give your input on how internet v2 should be made.

Things being decided:

1. DNS servers - centralized or distributed hashtables?
1. How do we make it easy for mainstream to get involved?
1. how do we keep a single admin from poluting the network? An aged version of this is comcast or Verizon in the US (gone from a single businessman to a large single ownership).

Hey everyone!

We are making a lot of progress very quickly, and its getting more exciting every day. The deeper we get, the more we seem to be onto something really cool. Here are some updates for tonight from my end so far:

the progress on gternet-cli is going much faster than I expected. I am going to open it for testing by all of you now. Spin up your box and switch to the root user and make sure you are in the /root directory
mkdir git
apt update && apt upgrade -y
apt install -y git
git clone git.jollo.org/gternet/software/gternet-cli.git
cd gternet-cli
**use this command to deploy an Openvpn p2p Intermesh node**
./gternet-cli deploy netbridge
**use this command to deploy batman on any mesh device**
./gternet-cli deploy mesh


All of this will be up on the wiki at some point soon
I will also begin reviewing and accepting merge requests.

Yeah man, that's great and all...but I don't want you going to jail over wrecking some corporate lawfirms fiber lines.

Can someone tell me how these inter city links work? 5ghz line of sight things?
What's the speed on this thing?
So total speed = speed of this los thing / no of people interconnecting between the cities? Won't that get slow real soon?

Latency would improve as more nodes are available to connect to, there wouldn't be just a single node between cities, but as many as possible.

New to programming
When do you use C vs shell files?

Don't run this on a computer with sensitive info

I know i'm not in the IRC yet but i feel compelled to answer questions 1 nd 3 with one word: Decentralized. Throughout history, centraization has led to lots of strife. By decentralising we make it hard for any 1 person to take control of the entire system.

As for making it mainstream, that jsut requires offering an alternative wehn whit hits the fan with unfair network operations and charges. If we decentralize the system, we can go door o door and offer soon-to-be customers the alternative and (possibly) free Sup Forumsternet, which they will take when it comes cheaper and provides more then paying 100$ for 10mbps. The problem with selling it is we could end up ike verison down the road (and fuck that) so the real question now is how we should model our small global startups bussiness wise to ensure profit so we can continue to spread it, while avoiding the inevitable corruption that comes with that.
Part of this corperate spread scheme involves advertising up the wazoo and bribing local politicians (politics, amirite?). As far as i am aware it is the only tried and true method to get this to the masses.

As for distribution of the product, we will need to figure how to connect this to the net while it's fragmented?

More questions:
- How fast is this net compared to the oldnet?
- Is it scalable (as in, with increasing users, how will it handle the load)?
- How does it distribute data?
- How do we plan on expanding?
- When the expansion inevitably leads to trouble with corporate behemoths (like verizon or telus) gettign involved, how will we be able to combat their mafia-like aggressive corporate tactics?
- Is the network centralized?
- How would we go about profiting from a decentralized network? Normies are easy to profit off of regardless so i'm sure we can keep decentralized and still make some sweet, sweet ca$h.
- etc?

all the code is available in the repository for you to review. It is open source af.

the map is a bit worrying, won't that oust our location and make us easy targets? Would it be fine if a gave the general postal code (like V5B 2A2 for example)?

It looks like most people are just putting in cities, which is OK

oldnet shill

uhhh opening a vpn in my SOHO network would open up my LAN to you, no? That doesnt seem smart

Yeah that seems pretty good.

So open scripts/p2p_deploy_node.sh
Go down to the main function near the bottom.
Comment out the call to set_iptables.
There. You can run the scrip without forcing the placeholder iptabes rules.
I will modofy gternet-cli to allow being more configurable tomorrow.
Thanks for the input!

I also should have more explicitly stated in my post that this is DEV only and is intended to be run on a fresh install, new vm instance, or new raspberri pi

just checked the map again.
Why and how in the fuck is there a NoKo (North Korean) in our network? WTF?

Is it jsut a fun-poster or is this real?

I created over 20 entries
Go figure

What a masterful troll

b-b-bumpaaaaaaRUUUUUUUUUUUUUUUU~

Byampuruuuu~

This seems vulnerable to attack. How exactly would you boot some wiseguy connecting to multiple points in the mesh and conducting a DDOS?

So how do I host a service on this? Are there domains I can get?

Piratebox can mesh, what about that?

IRC appears to be trying to address this from what i can understand.
That dude cat on IRC is trying tobdo some shit woth hashes to get domain names.

From what i can tell as an IRC lurker, seems that they are still trying to proof of concept the netowk. At least thats what knots keeps saying. They have madeba shitload of progress for the two week mark, but there is still a ways to go before we can send memes to eachother via gternet.
Ill keep purking and once its noob friendly ill try and use the gui thing or something.

right now OpenNIC seems like best bet for dns

bump

Namecoin

Unlike netrunner this general existed for 2 weeks
Rip, you will be missed next week. NOT!

i guess you would rather shill jew products than try to create your own

>people with almost zero knowledge about networking, data transmission, or even simple programming discuss logo and name of their future replacement of one of the most intricate and hard network (internet) the tread.

>people with almost zero knowledge about networking, data transmission, or even simple programming skills whatsoever discuss logo and name of their future replacement of one of the most intricate and hard network (internet) the tread.

can you start making sense and contributing, or just gtfo

Is this the appropriate topology for a network segment?

What do you think about the logo I made? I took the liberty to give it a name also

I can't even fucking........
I'm just glad you didn't name it localhost you fuckhead

Is there some kind of new software being developed? Orrrr is this just a theoretical thing?

>keknet
It's happening!

See

Bumpo

The oldnet shills are here

Morning Sup Forums

Im tearing down gternet-cli right now to have some improvments i described last night
>63712548

Everyone should hold off on using >63711988 until i can commit gternet-cli alpha v0.2 later today.

This might be a retarded idea, but would a block chain IP distribution system work for decentralized IPs? I'm not super familiar with block chains past the absolute basics so I'm not sure if this would work.

I've mostly been using it since this started so I figure I should update everyone on my progress.

Met with a local maker space and got them onboard. I couldn't get them onboard with the p2p option, they only want physical links

Looked into 900mhz equipment. It look promising, better range. However it's roughly 2x the price of 2.4ghz equipment.

Started work on a website last night. The goal being for users to add themselves to the map and have it automatically find the closest node, calculate line of sight, and make hardware suggestions.

>Inb4 Reddit spacing
It's call formating things so that they are easy to read.

It=irc
Phone posting from work.

this new Tamagotchi version even have infraorange

>The goal being for users to add themselves to the map and have it automatically find the closest node
This is cool, but I guess most anons wouldn't want to pinpoint their precise location, see
but probably it would work if you shill it in real life, placing it on your city free announcement boards, school halls, mc donald's bathroom, car windshields etc..

Exactly. We should all start trying to spread it locally as much as we can
I went into NYC last saturday and met with one of the dudes involved in a meshnet project in brooklyn. They have a functioning mesh that covers around 10 city blocks right now. They are using it for secure messaging and emergency alerts. He said that once we have a network setup then he will push them to add a p2p bridge or two.
I also spoke with my boss at the Uni i nix admin at, he thought it was a super cool project and will try to get the university to donate any older wireless equipment they were going to send for recycling anyway. Im looking at potentially getting up to 50 or 60 high powered cisco APs to fuck with and potentially use to prototype a more solid city wide network here in my home town

I just found this airlink.ubnt.com/#/ptp
Not as user friendly and only for ubiquiti products, but a good stand in for now

Also I was learned from the maker people that Batman sucks for directional links. So what they suggested doing was setting the it up as a normal bridge and then have that feed into a mesh node. It's not a huge deal because you would have to coordinate to set up a directional link anyway.

LOL you people just now found this?

Everyone should use this tool and find out just how retarded they are.
Protip: the literal earth itself will be in your way for most cases past 40 miles.
And I'd recommend that you don't exceed 15 miles with 5ghz.

5ghz sucks in all cases anyway

>Protip: the literal earth itself will be in your way for most cases past 40 miles.

this has been discussed at length in irc

Tbh I don't even know how long this project has been going for, but I've been using ubiquiti gear for a couple of years, and use airlink all the time

what kind of equipment are you using ? How much distance does it cover ? Are you in good los or are there obstacles, if so, what kind of (vegetation, houses, buildings..) ?

We've only been going for about 2 weeks so there's plenty you could contribute. Hop into irc and say hi!

Reminder that this could exist.
youtube.com/watch?v=zXTEWFb_6w4

Most of my experience comes from working for a wisp. They used rocket M5 international radios for most links under 5 miles. They also used some nanobeam radios for short sub 500m links or for mile links with no interference.
For long links or for high interference, we used rocket m365 radios, but those now have to be registered.
The longest link is believed was 14 miles, and it wasn't very reliable.
You absolutely must have LOS. Not only that, you need to have clearance for your fresnel zone radius.
You can maybe have some trees covering up to half of each fresnel zone, but the line of sight must be completely free of obstruction at all times.

I'm personally in the process of building a link between my house on a ranch and a town. It's going to have a 32Km link at the longest, but it's going up 1200ft in elevation. It looks good on paper, but it's damn long.

I may, but I think my IRC client was on a server that died a while ago.

You're exactly what this project needs. We have no hardware guys atm

Decentralized systems will always be toys.

Do you have any ideas for a mascot? Or a logo? I've got a friend who draws real good.

I want to call it "the people's internet of Sup Forums" just to put the Sup Forumss in a fluff

Read

thank you for your invaluable input

You'd be better off turning cellphones into P2P nodes by using the bands that are unused in America meant for overseas to communicate with each other. Trying to setup a "mesh network" across the country using cheap wireless 2.4 and 5GHz is going to be futile and slower than the slowest satellite internet.

Mesh is local my friend, there's a wired backbone

>turning cellphones into P2P nodes
That's what the MEGA guy is planning to do. I think he said in ~5 years smartphones will be powerful enough to make it happen.

Oh really, you're going to run a wired backbone across the country? With your neetbux?

hey knots, you asked about the Pringles antenna in IRC.
turnpoint.net/wireless/has.html

That's illegal

I wasnt being sarcastic. She draws really good

Phones are powerful enough to do it now. The issue is with many phones being locked down, deployment is nearly impossible

Right but the problem I'm seeing is that you may be able to cover most of a city with enough nodes, but that's about it. I've never done a mesh net, but I've got a few ideas for a personal project using one. I just can't see city to city connections at minimum though unless you have a geographic advantage. There's just no way to stay off the www.
Yall need a satellite.

There are existing enthusiast fiber systems out there. Not to mention that we're using the internet as a backbone until we can get peered with other alternets

Not if you're using 1850-1990 Mhz band

mon ami , to expand on my previous comment,
We will lease the existing backbone
Access to the backbone will be completely decentralized with an ICO
The ICO proceeds will pay the lease
We had an user three threads ago who said his cousin knew someone in OneWeb. So we could launch a satellite if this takes off.

lmao launching a satellite costs millions and months to plan. Knowing a guy who works there doesn't equate to free satellite

So OneWeb already has launch contracts for network satellites with ULA and Blue Origin. We just need to buy a couple. The ICO will fund the costs.
We have to start planning big league. Humans will be on Mars in a decade, so we can get another satellite to connect to Mars.
My cousin works for AT&T so I'll ask him more about that
The best part? Even if it spans planets it will still be net neutral. That's the power of decentralization!

>alt
Don't use that word please, makes us associate with alt right

How many satellites? I could fund one. I have lots of bitcoins from 2012

so basically you are just making a hyperboria ripoff except it has a cool logo and is affiliated with Sup Forums so the risk of law intervention grows
just why?

One should do all of north America.

Alternatively, we can do this for way less money
That is a balloon

Nobody said it would be easy. Obviously we are not going to get to another city in one hop. It's going to take major coordination and a good bit of luck if it happens at all. There is a reason there has never been a city to city mesh connection in the US. I think this is the greatest thing we could hope to achieve.

Like this project needs optimists and all, but god damn.

You can have her draw a crescent moon or a moon man look alike.

I approve the balloon.
This also looks cool.

Oops wrong thread

So what's wrong with a point to point microwave link? Line of sight i guess.
might work in flat cities. or if someone lives on high elevation.
The microwave link is also affected by rain.

Just a quick survey -
How much are willing to spend on this personally?

depends, I won't be spending much right now because I couldn't contribute that much because
>talking to other people

There is nothing wrong with it. It's just expensive.

I'm not taking crowd sourcing. I'm asking how much will you spend on yourself