Becoming My Own ISP

Now, frankly, I give zero shits about either side of "MUH NET NEUTRALITY WAR", but it got me thinking. What's keeping me from becoming my own ISP?
How much money in hardware would it cost?
Is it the ultimate way to disconnect from the botnet?

Other urls found in this thread:

blog.thelifeofkenneth.com/2017/11/creating-autonomous-system-for-fun-and.html?m=1
en.wikipedia.org/wiki/Freifunk
en.wikipedia.org/wiki/High-speed_multimedia_radio#Security
ubnt.com
diyisp.org/dokuwiki/doku.php
lists.ffdn.org/wws/info/diy-isp
ffdn.org/en/projects
youtu.be/FFPjJM6yYS8
youtu.be/cNbsiZcwGSY
youtu.be/1B0u6nvcTsI
twitter.com/SFWRedditImages

connect to a meshnet

Because the parasite telecoms own all the infrastructure.

This. Everyone is going to have to learn about wifi meshes soon.

this is something i've considered for a while
can we(referring to the general technical community) just create our own net in light of this shit?

and actually promote it and tell people about it. A big thing that kills stuff like this is people fall in love with the idea of exclusivity and promote secret club ideology and never tell anyone else about it

Some guy actually did this once, and eventually got arrested for it. We're entering the age of literal ShadowRun. Corps are going to have to be waged war with directly within the decade.

Arrested for violating what law?

Literally billions of dollars.
Contrary popular belief, the internet is not simply transmitted in waves like radio, you would need to create infastructure and hire a great many number of people.
Alternitively you could use a ham radio and transmit at a silky smooth sub 1kb/s

>ham radio
encrypted transmissions aren't even allowed in ham radio, this would be no way to live

Using the telecom's private property without their permission. As I said, the telecoms own all the lines and satellites. You're not allowed to use them and thus the internet without their permission (and paying them money). Most people don't know this is how it's set up. SOME parts of the net are publicly owned, but it's practically insignificant at this point.

I mean, we might as well start with 1kB/s transmission.
The internet wasn't built in a day.
If every faggot on here contributes let's say 1-3kB/s that's already better than most connections of the 70s.
Considering that we have modern technology on our side, once a basic network is created for transmitting just text people would see that it's possible and invest more into it.
Imagine how good this shit would look on a resume.

Very soon "what is allowed" and "what is not allowed" will be nearly irrelevant. The capitalists are simply becoming too despotic for people to care about their masturbatory laws anymore.

And what would actually happen is that telecoms would lobby all tiers of government to shut you down for "pirate" or "disruptive" transmissions.

Look up neighborhood owned fiber projects, that's basically what it takes.

What about neighborhood wirelss intranet?

Honestly we may have to do this soon. But like I said, don't be surprised if corporations lobby local governments to fine you.

So theres basically no way to do this and not face action unless we start digging our own lines?

There's so much encrypted wifi traffic everywhere how tould they even find out?

One day, you're going to have to draw that fucking line in the sand, because the predators won't stop nipping until they get to feast on your corpse. Everyone is going to have to fight the greed of the rich eventually. And with the way shit is accelerating to hell at warp speed, sooner, rather than later.

You would become a wisp. For starters I would buy a 60ft tower in a small town that I can get a fiber connection to. Using some gear made by a company called "ubiquity" I would hang an Access Point or 4 of them, N S E W, then you start hooking up customers and charging them. 50mb is the current high speed. Look up "RiseBroadband", I work for them. More than a few people are looking into wireless internet. 5g is what's gonna really change rural internet access.

They did. I wish I could remember the name of the guy. That was the most high-profile wifi mesh that I'm aware of. And to answer your question, because the fucking telecoms have the government in their pocket.

Wifi is shit man, if you're going to provide neighborhood fiber you need a 10gb uplink and provide 1gb to each home.

The same FCC that would control the net with NN makes it illegal to have a large wireless network.

All it would take to set up a community internet is wireless routers every 400 yards.

>implying anybody will be willing to give up convenience
they'll just change the definition of what's "right"

I'm the Rise Broadband guy from above, we actually have interested a few companies like Google and Mitsubishi. The problem is not overloading your towers and handling peak hours. People are cable cutting and bandwidth intense programs are becoming more and more normal. It's extremely hard to keep up with current wants and needs while being able to churn a profit for the investors. We constantly have to upgrade gear, and plan for the future.

You can't become your own ISP. There is no backbone of the Internet that you can just connect to; it would be more correct to say that the backbone is everything else. If you want access to the sites you know, at some point you have to connect to everything else through an ISP. You are either paying them, in which case you'd be just a customer (not your own ISP); or you're in an agreement where you let their users' connections use your network and infrastructure to access the sites hosted by you and vice versa. You don't have millions of users nor a lot of hosted sites, so you can't bargain that kind of deal.

You won't have a choice soon. The capitalists are already TAKING convenience AWAY from customers. It's leaving whether you like it or not. Remember when you could just hop onto a website or limewire or vuze and just download whatever you wanted without a VPN and basically nobody gave a shit? The internet is ALREADY infinitely shittier than it was 10 years ago. And it's going to keep getting worse.

You would pay for a fiber pop from another company, then build your subscriber base.

I know this is extremely inefficient, but this could be a place to start.
>use a ham radio network as previously suggested to transmit miniscule amounts of information
>over time upgrade to local wireless connections built upon ham radio networks
I know without satellites or communications cables we won't be able to rebuild the web but let's suppose: I'd say there're about 1 - 2 anons for every 3-4 km^2, let's say they establish communications with one another.
Can't we build open this mesh of interconnected relays to transmit information?
For example let's suppose user #45 is located in Toronto. He wants to transmit to New York, however can't due to the limited range he has with non-conventional connections. He uses relays situated every couple of km by other anons (raspberry pis or ham radios) to strengthen his transmission and keep sending it along until it reaches user in New York.
Would this be possible?
A relay strengthening network?

Why would anyone want to ever work with corporations ever again, you faggot? Capitalism is the problem that caused this. Literally the government invented the internet (more or less) and private interests destroyed every facet of it.

Or you might just let your neighbors use your WiFi for $3 a month. You'll make it big some day.

>confusing capitalism with oligopoly of one flavor or another

Man now I have to figure out how to get my hands on a HAM radio that'd work for this and how HAM even works

nothing is stopping you from renting a trunk from a larger service provider, and pointing a large dish to a remote island and sell the people who live there better internet

good luck finding an area that one of the big companies havent already claimed

Thought about doing this from mainland wa to whidbey island or some of the other puget sound islands. Too much work, im lazy

Capitalism always takes the same path to despotism. Always. The lie has always been that "this time will be different, we can control it by letting it get totally out of control (to serve the rich. Again.)"

>only
Please, this isn't South Korea.

Also, enjoy 2 hour latency and probably shitty total throughput never mind trying to actually route that monstrosity.

>OP asks about setting up his own ISP
>Idiots respond by telling him it's too difficult to set up a telecom
Do you retards know the difference between an ISP and a telecom? If not, you shouldn't be attempting to answer OP's question, you fucknuggets.

You're running up against the limits of bandwidth vs cost of implimentation. Microwave backhauls run at 30-40 GHZ and have to be line-of-sight at that frequency range.

That's the story of nearly every communist experiment. Keep drinking that marx-aid. The US hasn't been capitalist in a while.

It's not really costs, it'd just be a pretty neat project to do. The same reason that people still communicate with Morse code. Not because it's efficient but because it's interesting and amusing.

Are there any alternative methods of data transmission?
Could ANY type of device be connected to another one over an adequate distance (at least 2 miles) for effective relaying?
It literally doesn't matter what the device is as long as it doesn't need to be wired.
Once we have a plan we can work from there.

Closed shell networks are illegal now?

Sure, sure. Gilded ages aren't capitalism, corporatism isn't capitalism and China is communist!

If local networks over wifi ever get traction enough to compete with ISPs, you better bet your asshole those bastards will have governments passing laws to fine you for it.

And by the way, that's not what he was doing. He was selling a device that allowed people to connect to the greater web anonymously.

CAN'T. HAVE. THAT.

This blog.thelifeofkenneth.com/2017/11/creating-autonomous-system-for-fun-and.html?m=1 explains how difficult is to be your own ISP.

You have to buy fucking ipv4 addresses, but there is none available in the market, only ipv6, so, if you still want to continue with the project, you will have to rent a range of ipv4 addresses. After that, you will have to spend a couple of grands in a server capable of handling the traffic, and after that, in the last mile infrastructure.

Good luck

Lads, lads.
We can create a WANET internet.
"The Wireless Ad Hoc Network is a form of wireless connection in which devices transfer data as relay nodes to one another in a chain."
Anons can set up Raspberry Pi or other equivalents maybe even old computer nodes all over the world to connect to one another.
This IS possible and you need nothing more than a Wi-Fi router, a hot spot, or a multitude of other devices.
It CAN absolutely be done.
If you fags are interested I'll make a thread once this one dies.
"Project Sup Forumsternet" or some shit like that.

How could that possibly be constitutional? I wonder if it jas ever made it to the supreme court. Probably not, because the feds don't just sit around monitoring ham radio.

Cuba has massive city wide networks

www.polygon.com/platform/amp/features/2017/5/15/15625950/cuba-secret-gaming-network

You might have better luck with something more like cellular service. The protocols are pretty complex. Also, governments reallllly like controlling what you can do with radio waves. Bandwidth itself is worth billion$$$ and the tiny slices left for the public are just table scraps.

I worked in a fab that built Japanese-contract cell tower router boards for a short stint.

That's cool, but cuba is tiny.
how about
>sneakernet over IP

Yes, and you're forbidden to dig lines in many states/cities.

Even the mighty google can't do shit.

>mfw cubans have a literal cyberpunk internet
>illegal nodes and connections all come together to make one mega-mesh network full of pirated shit
I want to go to Cuba now.

Time for the nukes people just let it end.

this. for far out areas, one could set up a microwave relay using a string of AirFiber (by same manufacturer) transceivers at each point, until it gets muxed to fibre in some significant town where facilities exist

Interesting idea... There is so much computer waste out there just because of upgrades. Most units work fine. Maybe combine it with some sort of PC recycling program? I could see this working easily on a local level if people were organized enough.

>Sup Forums

please fucking do this. I don't want to get all faggy and politcial but I really wish the internet was run by the people. The infrastructure should be public desu so this would be super interesting.

already exists, since 2002

en.wikipedia.org/wiki/Freifunk

how about montana?

I'll frame this post.

i like where this is going.

Genuinely interested in this. Please make said thread.

wifi meshes and meshes in general don't scale well, at least if you plan to have each person be a node in the mesh. available bandwidth goes down logarithmically afaik.

what you could do tho is have a hub-and-spoke where you connect your neighbours to other hub-and-spokes in a mesh. that's the only way to make a normie-accessible mesh anyhow.

sauce on this?

1KB/s is good enough for shitposting on IRC desu. you can get away with a lot as long as you exclude normies

this is basically how shit like fidonet worked back in the day to move messages between bbses and get around long distance charges.

just keep in mind, if you're a burger, that US regulations prevent fully encrypted traffic over amateur radio frequencies. see en.wikipedia.org/wiki/High-speed_multimedia_radio#Security

i can make the logo

Ubiquity.com

ubnt.com

Becoming a WISP is pretty easy. This network was built by an over 60 year old guy because a lot of his neighbors where just able to get < 1 Mbit/s DSL connections. You still need some uplink but you can get a decent line without all the consumer bullshit.

With Freifunk's software running on DD-WRT, and an app built to interface mobile devices to the network, and (assuming that most devices aren't using 100% available bandwidth themselves) wouldn't it be possible to split bandwidth across whatever devices are currently connected to the internet, allowing everyone on the network speeds in the Terabyte or even Petabyte range? Not to mention the fact that if packets are being split and pulled from multiple IPs and even ISPs, that the data would be untraceable? Plus, and this might be a bit of a stretch, but shouldn't it even be possible for users to set aside a fraction of their disk space for local storage, so that downloads of static data could be requested locally instead of across the internet?

Say I take a picture and upload it to something like tinder, most of the people who are going to view that image will be in my local area. One user views that pic, which is uploaded from my device via intermediate devices within the network, and while it's sitting in her temp folder, other users in the area, all sharing bandwidth, download that image from all available sources, which would be my device and the first other user's device, as well as the intermediate devices, all without anything being uploaded to the 'internet'.

All devices are both clients and servers, but only serve packets currently being held in a temp folder. Users could designate any amount of space as a temp folder, with new data over-writing oldest data first. As long as packets are all being split across multiple nodes, no server that wasn't also a client of an entire package download would have all of the data, so no user would have readable information that they didn't request. And since packets that can be requested within the local network wouldn't require external internet connection, that would further increase the bandwidth available to users that did need the external connection.

If three people are sitting in a room, each with a 6mb connection, one person can start streaming a movie at 18mb/s, then if one of the other people wants to watch it also, they would be able to stream it directly off the first device with no additional bandwidth load.

>Unlic
>Terabyte or even Petabyte range?
Hell no.

>Not to mention the fact that if packets are being split and pulled from multiple IPs and even ISPs
Thats what multipath TCP is for right? Whats the status of that? Haven't followed it recently.

>shouldn't it even be possible for users to set aside a fraction of their disk space for local storage, so that downloads of static data could be requested locally instead of across the internet?
i think this is kind of how freenet works, or things like content-addressable networking. everything is a big cache.

>Hell no.
Why not? Most gamers probably only use 80% of your bandwidth at the high end, practically nothing while they're afk. You take all the bandwidth across an entire city and serve it all to the devices currently requesting packets, the speeds would be incredible.

>multipath TCP
I just saw a video on YT, I think it was sentdex, had two cable connections on the same network, getting 2x speed he was getting before. It would be taking that packet spreading concept and splitting it among all devices on a city-wide network.

>content-addressable networking
Exactly! So the only data actually being passed through the internet would be data that has to travel from one major city to the next, once that data is there, it would be redistributable locally, essentially meaning that a YT video uploaded in Seattle and viewed by everyone in Portland would only utilize internet bandwidth once, and local nets thereafter.

I don't know where this >Unlic came from. This should have been.

>Wireless

It's just not possible. Most likely even physically to transfer Terabyte/s with only a few hundred MHz spectrum.

>not possible.
But only because of current hardware limitations. The bandwidth would still be available for it though, which means that it's entirely possible to maximize the thoroughput to the devices' limitations.

Companies like Google (mostly YT, Blogger) and Netflix would save a fortune on bandwidth and server space, especially if NN fails and they have to start forking out millions to keep their users' connections from lagging.

Google could put this in their next Android release and all android phones with 802.11ac WiFi antennas would suddenly have 162.5 MB/s (not Mb), and it would be free, because all that bandwidth would be sourced from devices on WiFi connections, not mobile data.

Okay, so maybe if each device only has one antenna it would only be 80 MB/s, alternating package capture and redistribution... It'd still make Apple go bankrupt overnight.

>But only because of current hardware limitations
No. You have about 700 MHz of spectrum on the license less 2.4 and 5GHz band you can use. The most you can get with 802.11 ac is 6933 Mbit/s per 160MHz. Thats about 30Gbit/s theoretically in a Lab environment. In real life you won't ever get close to that. Sure you may increase the bandwidth with better modulation and more mimo but there are also limits for that.

If you could encrypt HAM Radio it becomes completely unregulatable. The reason why the FCC regulates radio transmissions is because if they were unregulated, radio would be a complete clusterfuck, and people wouldn't be able to control their RC cars because their Wi-Fi is on. We really don't want people to start building RC cars that operate on HAM Radio frequencies using encrypted transmissions if we want HAM Radio to continue to exist.

You need government licenses and a few million dollars of hardware + permits to establish that hardware.

10-20 million dollars total cost I would guess, depending on region etc..
And a few thousand ours of time of course.

No, it does not stop you from becoming part of the bother.

Which is how much faster than the typical residential internet speed?

I'm getting about 1Mb right now, wouldn't that be 7,000 times faster? I think the literal cap is a moot point compared to the end-user's relative experience.

Unless the hardware was already out there, being paid for by the users. If the network is built off of mobile devices, hard-coded to the OS, it really comes down to who writes the code first, Google or Apple; Once one side of the markets' users are getting free internet thousands of times faster than the competition, they control the internet.

I'm the user from i know that WISP are a nice thing and easy and cheap to setup. Excluding regulations ofc. I just told you that Terabyte/s is not possible over a wireless connection.

best way? become a wisp as twenty other people in the thread suggested

i work at a wisp rn, and one of the older tech guys hopped ship and literally does his own ~2k subscriber network by himself now

granted that took ~10yrs of work but yeah

like he said you arent going to see crazy high speeds (our highest offering is only 10mbps) but a decent connection is possible assuming line of sight & decent equipment

>The most you can get with 802.11 ac is 6933 Mbit/s
That's 7000 times 1Mb/s
>You have about 700 MHz....6933 Mbit/s per 160MHz
So, even faster than I said before... I'm using your math here, so what am I not understanding? The top Google result for 802.11ac speed is 166 megabytes per second (MBps). But if that's wrong, why don't you just tell me what the theoretical limit on 802.11ac is?

I'm not arguing that there isn't a cap, I'm just saying that that cap is going to be significantly faster than anything anyone could ever hope to get from their ISP.

>meshnet
>this
----
> mesh topology
high level guys, keep working.

You mean those cancerous wireless signals?
I'm afraid we'll have to shut you down so you can keep paying for our services :^)

ISPs only have local monopolies, even Comcast/TimeWarner wouldn't be able to compete with Google if every Android and ugh... ChromeOS device was redistributing local decentralized networks.

>why don't you just tell me what the theoretical limit on 802.11ac is?
I told you. 6933 Mbit/s in a Lab environment. You will never reach this in the wild.

>I'm just saying that that cap is going to be significantly faster than anything anyone could ever hope to get from their ISP.
In theory. You are using a shared medium and everyone uses wifi at home for their laptops and smartphones too. They also need a part of the spectrum. It really depends on the distance and radio pollution in your area what you can get. As told you they are offering max 10mbps. We are not enforcing a specific limit. But out customers can't get more than about 100Mbit/s. And that high speed is only because we are controlling about 90% of all wifis in that area and can do proper channel management.

Doesn't scale.

It's possible. I think there's a future for a more decentralised connection system, something that provides more freedom and prevents surveillance. The projects to create wireless networks are very interesting and worth pursuing.

It's never going to be a viable alternative to the massive amount of existing infrastructure though.

Regardless of all the nonsense surrounding NN, the true way to "save the internet" is to enable more competition. I'm not sure how that's possible but if someone figures out that then they would massively improve the rate of improvements and decrease costs.

>Use infrastructure belonging to someone else
>Charged with a crime
>We must stand up against such injustice! Viva le revolution!

>6933 Mbit/s
That's still insanely fast.
>they are offering max 10mbps
>Doesn't scale.
That's not a distributed network though, that's basically just a large-scale home WiFi network. Totally different things.

>It's never going to be a viable alternative to the massive amount of existing infrastructure
But it would take enough of a load off of the primary infrastructure to increase available bandwidth several fold, while increase speeds for anything that can be locally served to within the capacity of the hardware, rather than a bandwidth cap. And IF the hardware was consumer-provided, the service COULD be free.

And I'm not saying that it would completely replace the current infrastructure (it could if Google gets on board and builds something new), just that it would possible to share bandwidth across multiple IPs and ISPs to serve greater speeds to a larger number of people.

In a realistic scenario, there would probably need to be micropayments, and those who request more 'new' data from the internet would pay more than people requesting 'locally available' data. But it may even turn out that people could make money by providing server space for frequently requested data on home servers.

Hardware would cost nothing compared to the fees you have to pay for the data and subscription.

>That's still insanely fast.
THEORETICALLY. There is a huge difference between theoretically and practically possible speeds.

>That's not a distributed network though
Which makes it even worse. Multiple devices on the same channel reduce the bandwidth drastically. Mesh networks don't scale with hundreds of nodes. Having to pass data over multiple nodes occupies frequency on the whole path. Smartphones send in all directions which makes it even worse.

>But it would take enough of a load off of the primary infrastructure to increase available bandwidth several fold
The primary infrastructure consists of 100Gbit/s ethernet and google and netflix etc. already have caching servers in the data center of most isps to serve the content locally.

>THEORETICALLY
But in theory, it's thousands of times faster; even if the practical speed was only 1% of the theoretical speed, that's still tens of times faster than what we have now.

Your argument is like if someone offered you a job that paid 20 - 30 times what you're making now, and you were like, "Not a million times more? Then fuck you."

>Multiple devices on the same channel reduce the bandwidth drastically
With the current hardware, that's a limitation. But if this were to become the 'new internet', the tech would catch up quick.

>The primary infrastructure [already does that]
Which is why we know that it'll work, we'd just be expanding that infrastructure to bring those data centers closer to the end-user.

If my neighbor and I both watch the same movie on Netflix, and the video is cached on the first device that downloads it, it doesn't need to be downloaded again, because it already exists on the same network.

Minor changes to how we handle data would have a drastic impact on the amount of data flowing over the ISP. That translates to lower costs and faster speeds.

Encrypted audio broadcasts arent allowed. It's still legal to broadcast digital data transmission.

You need to get a block of IPs from your RIR and colo space for a router at an IXP. Configure the router, then purchase and run optical to your house. You can see about somehow getting an audience to apply to use existing cable infrastructure from an ISP at a contract cost, but you'll likely pay over $200,000 monthly for that kind of contract, peering, colo, and addressing.

If you ran your own cable, you're looking at at least a million one time with massive recurring fees.

You need to have a lot of money to do this.

Alternatively, you can create a wireless ISP with nested access under an existing provider. You just need to install the radios and interconnect them with all your routing protocols sorted out (this is not easy)

Any of these may be relevant to your interest in relation to creating your own networks and sustaining some of the equipment around it.

diyisp.org/dokuwiki/doku.php
lists.ffdn.org/wws/info/diy-isp
ffdn.org/en/projects

youtu.be/FFPjJM6yYS8
youtu.be/cNbsiZcwGSY
youtu.be/1B0u6nvcTsI


I'm more excited at the prospect of people maintaining their own connections to other peers than a centralized network. With projects like cjdns, IPFS, I2P, et al. I'm pretty hopeful that it may one day become the norm, but we'll see.

>than what WE have now
Who is we?

>practical speed was only 1% of the theoretical speed
Thats about correct in a heavily polluted area.

>With the current hardware, that's a limitation. But if this were to become the 'new internet', the tech would catch up quick.
Wireless technology is already heavily researched.

>bring those data centers closer to the end-user.
It's already close to the user.

>That translates to lower costs and faster speeds.
No it translates to a LOT more complexity and therefore problems and thats way more expensive than a new 10 or 100Gbit/s link to your area.

sure im my own ISP in my own room, have like 4 devices connected to a switch

>This. Everyone is going to have to learn about wifi meshes soon.

Wouldn't that result in an insane number of hops? Wonder what the latency would be like.

>tfw idiots try to 'stick it to the man' by creating meshnets
>the quickly work out that a full mesh is terrible and some form of permanent infrastructure would be helpful
>then some people make even better permanent nodes, that are liked together by some hardline - but they charge for this access, so some meshnets and connected to others via fast hardline and others by routing through a slow meshnet.
>eventually the permanent infrstructure takes over the routing between all of the major meshnets
The Simpsons?