Alternative Internet

Alternarive Internet thread! /g/ternet revival edition. >What is /g/ternet? Quite some time ago, when Net Neutrality dissapeared in the USA, /g/ wanted to make its own Internet dubbed /g/ternet. The project has made quite some progress and is at a pretty nice stage at the moment. More info can be found in the /g/ternet Wiki: https://mesh.gentoo.today >Where is the project? The project can be found here: https://git.gternet.me/gternet/gternet-cli >This project is dead OP, just let it die Yes, I know it's dead. Hence why I called these threads Revival Threads. I really think this is a neat idea that has quite some potential. So yeah, please do contribute as much as you can anons! Read up on the Read Me and the TODO in the repository and study the code if you want to and give any suggestions you might have! Also, if you're unfamiliar with this project, I recommend reading the previous threads that have been posted on the Wiki. I saw this thread on 4chan/g/ and thought it was an interesting project. What do you think, lainons?

I'm pretty sure there's an internet out there that functions quite well without whatever this is. Could you explain some of the technical aspects of the /g/ternet? Is it literally just 4chan assholes memeing "install Gentoo" on us?

>>14587 No it isn't. It's a p2p network that relies on anons building their own DIY meshnet node so we can have a private and anonymous alternative internet. It is powered over p2p vpn bridges allowing secure connections! Anyone with the money and the knowhow can build their own meshnet node and participate in the network. If you live too far away from others you are connected over a bridge node which makes it rely on the internet still... Anyway, by first creating small city local meshes, we wish to expand it to larger decentralized groups eventually. You can see the code in the repository in the OP post. Pic related.

imho, creating alternative internet that are not easly accessible is like creating a new island that soon will sink in the void. I was wondering around and I found anoNet, but rn can't join as their vpn certificate is expired. As you can see, they became just like metanet (mentioned on anoNet.org site). They use simple tools like OpenVPN to create this virtual private network (I think you know that internet is literally inteconnected networks), but to join is not like connect a copper cable and pay a fee, instead you have to join their trust network and you can't pay for trust. I see that many dark net or alternative network born from a community and as they grow the can either become an island or not. When it become an island, nowone if not invited, can join and this ruins the community as people could loose interest in it. As I see more and more p2p net I can only hope for some kind of tool that will connect all of them in a truly p2p internet. imho (as maybe I'm wrong), for now it's just an experiment that looks like is working at half: you can create your net and noone can find/join it if they don't know about. Maybe the problem I'm pointing out is how to encourage more people join an anonymous p2p net, and that if in this net there is nothing to do, it will loose interest. So what would you do in this anon net? How would you convice your friends/collegues to use it?

>>14593 That's a very good question lainon. And to be frank, we don't know the answer (just yet). This new network could be used as an anonymous and private file sharing network. Perhaps it doesn't even need to become popular. Maybe this program can be used more as a framework for other people to create their own sort of local intranets to share/upload data to. That seems like the most plausible outcome for this projects. It can be used as a basic system that others can use as a basis for their own networks that, ultimately, all can be connected together at a certain point to have one big network. But let's leave tomorrow's questions to tomorrow's us!

Here are some images that clarify the project's goal and intended structure.

A freind and I were talking about doing something similar a while back, a neat idea we had was to have each node in the network host a part of a distributed cache so that you could keep traffic at the end node a bit lower and optimise routing a bit better (moving items in the cache across the network to recache them in high use areas etc) we ended up being a bit too far apart for directional 2.4ghz to work, and we considered a AMPRNet type system, but that didn't work out. AMPRNet might be worth looking at if no one is in your area.

>>14602 Thanks! I'll definitely take a look at it.

What hardware can be used as a node? Can it be something simple like a Raspberry Pi? Sorry, I don't understand much...

>>14600 >all wireless how would this be possible without massive global cooperation? it's idealism

>>14586 Am in minneapolis, this seems cool. Any way I can help?

>>14615 https://en.wikipedia.org/wiki/Freifunk

I think there would be people that can be nodes in major cities. I want this to happen.

>>14616 There's a TODO list in the Git repository. Take a look there and see if there's anything you think you can do! Also, you can help improve the Wiki if you want to.

>>14586 >>14600 NYC Mesh is an existing project that's doing something similar, but more locally centered, with pretty good success. https://www.nycmesh.net/

>>14621 Wow, seems quite cool and similar to what this project is trying to do. Thanks lainon! Yeah, I think that this project is similar but less reliant on the internet. Only to connect the more local nets to other faraway ones.

>>14622 More reliant* I meant to say

This is Kali! The mascot of /g/ternet!

>alternative internet >p2p, distributed, serverless, encrypted, private I ran across a project the other day that claims to be all this and more. It has been in development since the mid 2000s and it seems that they are finally nearing the end stages of development. It has been very under-the-radar. Tim Berners-Lee even tweeted about it. And it has had mentions in TV shows and Movies (whether those are good things or not, I leave for you to decide). It is called the 'SAFE network', formerly known as Maidsafe. If you look this up online you'll probably get results having to do with a cryptocurrency, but that is just part of it. They put a coin into the network so that people could make money from having valuable content, but it's not something that you have to buy like Bitcoin in order to participate. It truly is trying to be a new form of internet. The project is massive in scope and scale, and I am still not quite sure what to make of it, but it's worth checking out as something realistic that may actually come to pass in the near future. There are some very exciting implications with the way they've structured things that puts tons of power back into the hands of the people who actually create. I suggest putting some time aside to do a deep dive on this project and wrapping your mind around it. https://www.youtube.com/watch?v=R6TE6Cmivlg https://safenetwork.org/ https://safenetwork.tech/faq/ https://en.wikipedia.org/wiki/SAFE_Network https://safenetforum.org/

>>14631 Wow, didn't know about this project. But just like you said, I don't know whether much media coverage is a good thing or not. Sadly it more often the latter than the former. But it does seem like a cool project that has a lot of potential. Thanks Lainon!

>>14586 I'd been hoping that you'd post this on Lainchan sometime, anon. I lurk /g/ occasionally but generally don't post due to recaptcha. I'm planning on introducing this to my university's local hacking group. I figure it would be an easy way to get started, given that everyone is close together. How far along is the project? If we put aside say $200 USD for hardware could we create a decent network with this? >>14631 Seems really interesting. Active development as well: https://github.com/maidsafe/parsec Seems almost suspiciously ambitious. They plan to have people volunteer their computing resources in return for safecoin. Infrastructure wise, it's still dependent on the Internet, right? So it doesn't achieve the primary goal of a meshnet, at least on its own.

>>14638 The project is at quite a decent state, it just needs some improvement and it's more like a beta than a fully fledged system. But it's nice to see that people are intrested! The problem is, I think that the original creator has gone and abandoned the project, so perhaps it might be a good idea to start a new repository. Anyway, the estimated cost is about $50 per node, so 200 would get you about 4 nodes. But I'm sure you can make it cheaper and you probably already have some materials. Success lainon!

>>14636 No problem anon. To be fair, they didn't get much coverage at all for the first 10 years that they were trying to build this. But I am not an expert on them since I just discovered the project recently, and it's always good to be skeptical. >>14638 >Infrastructure wise, it's still dependent on the internet, right? Yeah, it doesn't solve the infrastructure problem. But it seems to have the spirit of decentralization built in to the way it works, so it is a step in the right direction, and might be well suited to running on those alternative network types. >suspiciously ambitious I got the exact same feeling. Tried digging into it a bit... I'm sure it goes deeper than just being the brainchild of a single company. Like the original internet itself, there are probably ties to other entities.

>>14638 I'm not someone who likes to directly call stuff that I'm suspicious about a botnet, but come on; >They plan to have people volunteer their computing resources in return for safecoin That does seem a tad bit suspicious, doesn't it?

>>14663 what's suspicious about that? Homomorphic encryption is a thing

So this uses batman-adv. I encourage anyone thinking of doing this to just install LibreMesh on their router. I'm already running it on a spare and it works a charm. Also installed cjdns on it, easily.

>>14663 >what's suspicious about that? You're asking what is suspicious about running unspecified code on your devices in exchange for memecoins?

>>14673 I'm not asking anything. You probably meant to reply to this lainon: >>14664

>>14673 you do that every time you run code you haven't read yourself. If it's sandboxed there's no risk.

>>14683 That may be true, but something as ambitious as this; a private network built on top of the Internet that also produces crypto coins, just sounds too good to be true. It sadly has to have an ulterior motive behind it.

>>14624 Kali a cute! Cute!

>>14683 What was the last code you didn't read and "sandboxed". How did you do it?

>>14624 Okay now I'm interested

>>14758 As you should rightfully be. Kali is the best mascot ever!

>>14624 It's cute and fitting because loading screens is what the /g/ternet mostly consists of. Doesn't it lainon? I would like to believe that a p2p approach can work but it's dependent on a lot of people playing their role and thus just not efficient.

>>14828 Yeah that is very true and does pose a problem. However, if we have a few big nodes that are up 24/7 especially in major cities; smaller nodes of regular visitors can connect to the meshnet via these nodes and help as well. This project is more of a template, a framework so to speak, for local meshnets if people are interested in having one.

>>14755 I navigated to lainchan.org and allowed javascript to run.

Didn't there used to be a map?

>>14850 There was, but it went down and the original creator has to reactivate it. And I guess that won't happen, as it seems as though the original creator has abondoned the project.

>>14863 Is there any of the point data still available? Making a map is easy, getting the data is harder.

>>14590 sounds like the biggest hurdle to widespread success is physical infrastructure.

I wonder if we can do something similar to /g/ternet... How about us Lains make our own p2p internet like service? We could make a team or something.

>>15303 on such a slow board? how would you organize it?

>>15303 if you dont name it wired imma find u and hurt u

>>15303 We already have the lainnet (hyperboria)

NEVA BEN DUN BEFOAR! >Created in 2005, SIPRNet was instituted to replace the former Defense Secure Network 1 (DSNET1) as the “SECRET” portion of the DISN (Defense information Systems Network.).There are four USDOD agencies jointly responsible for the management of SIPRNet; Joint Staff J6 [the directorate for Command, Control, Communications, and Computers], the National Security Agency (NSA), The Defense Intelligence Agency (DIA), and The Defense Information Systems Agency (DISA). Any information passed via SIPRNet is automatically designated with a “SECRET” classification, and becomes the property of the DISA. SIPRNet is maintained, and security to SIPRNet is managed, by the Defense Information Systems Agency (DISA). The DISA is a division of the USDOD functioning as a combat and security support agency. The DISA provides real-time information technology and communications to the President, Vice-President, Secretary of Defense, Military, and Combatant Commands via the Defense Information Systems Network (DISN). While there is no such entity as a SIPRNet computer or terminal, any computer that is to be utilized in connecting to SIPRNet is marked and tracked by the DISA. To accomplish this task, the user connects via computer to a secure line, which connects into a SIPRNet server. This secure line negotiates an encrypted connection which meets pre-designated SIPRNet standards. If these standards are not met when connected, the communique is then dropped. This same process is utilized in not only land-line and wireless connections, but is also effective when Voice over Internet Protocol (VoIP) is the transmission delivery system. The DISA uses this encryption method to maintain an audit of all SIPRNet users, as well as all transmissions.

Wireless User Groups are a thing in my country, they seem to function more or less as an alternative internet among enthusiasts in the country http://www.wug.za.net/ (The forums are dead on this site, but you can find more information in the forums of each city)

Wasn't there a map? I saw an old map with a person from Portland, Maine on it.

Just a bump - as we should loose the request post - not this thread.

Bump. This project seems quite interesting.

>>14593 >imho, creating alternative internet that are not easly accessible is like creating a new island that soon will sink in the void. This! Agree with creating more alternatives to the internet, even if they are just techdemos that die quietly in the end. After all, using bitcoin as example, there were many seemingly useless technologies and failed starts that finally were combined to make bitcoin possible. Early attempts at cryptocurrencies date back to the 80s. And yet, here I disagree with the topic, ACTIVELY cutting the cord to the rest of the internet is stupid. It will not lead more users to it and will not do anything for the current users.

bump

Bump. I want to see /g/ternet take off.

Could they have named it anything worse? Even worse than "gopher"

>>19105 Idk man, the name sticks.

what if there were a way to send electrical signals over existing telephone lines (without having telephone service) and have a smol p2p text-only bulletin board software that can operate within the limitations of such a network?

>>19112 I don't think that's how the POTS works. Also: who still has a landline? Regardless of whether you pay for service: do you even have the wire coming in to your building?

>>19120 Modern "landlines" often actually don't even run on landlines, they tend to send digital signals over your usual internet. The old telephone network mostly still stands because it'd be expensive to tear down. Running signals over that is doomed to fail, I fear. Maybe something could be done with cellphone networks. Think modem for a cellphone.

I'd like to see this continue, bump

This sounds really interesting

>>14586 >Quite some time ago, when Net Neutrality dissapeared in the USA, /g/ wanted to make its own Internet dubbed /g/ternet. Actual cancer. Noone cares about net neutrality aside from clickbait readers. Where are the good warez sites? Where are the good scientific sites? Where are the good technical journalism sites? They're gone. The government got rid of them. It's not illegal to do anything remotely interesting on the internet. And some retard fuarrrk ITT will defend the fact that it's illegal to play a game on a LAN without paying for 10 copies. It's actually de-facto illegal right now to have a website that allows you to browse two articles without having to load 1MB of JS and pressing next 17 times to get to the 18th paragraph.

s/It's not illegal/It's now illegal

Hmm

>>19525 >They're gone. The government got rid of them. You gotta lurk more. Present Day, Present Time! AHAHAHAHAHA!

>>15365 So, it's a VPN with rules for what hosts may connect. Gotcha.

>>21870 Domain expired, duh. I friggin hate these mozilla's error messages for mentally disabled, you have to click 2 times in order to proceed, and first page gives you nothing coherent describing the issue, only passive-aggressive buzzwords to scare normals and make them click blue button quicker

Bump because it need revival

bump

>>21890 ...Not the domain... the SSL certificate...

>>22475 I don't think its only the cert, the page: https://git.gternet.me/gternet/gternet-cli returns a 404 for me

Hams have been working on an outernet using off the shelf routers and custom firmware for decades. They've also organized BBSes using shortwave radio. Why don't YOU have your ham license, Anon?

>>14586 >Maximum visits allowed for maps without a paid plan is 100 visits. This map is currently frozen. To unfreeze, please Upgrade the account that owns this map to one of our paid plans for more visits. >In case the map is not owned by any account, Sign-Up for the service and claim the map with a paid account. Halp pls

>>14631 This is really interesting.

Do you lay new cables? >yes Hahaha you don't you don't have the money idiot. You can try something radio/bluetooth local distributed peer to peer but you'll be completely alone forever >no It's not a new internet, you still use the Internet Protocol to address stuff. You've simply created a new protocol as part of the old internet.

I can't access any of the links you posted OP but I feel that I got the gist of it. It would be nice to revive to renew this project but we would need to restart with a new repo and map since the creator dropped the project. I think it would be a good idea right now to see what resources we currently have and document similar implementations.

>I'm pointing out is how to encourage more people join an anonymous p2p net, How would you convice your friends/collegues to use it? This is the million dollar question. 8ch is dead, (RIP /cyber/), and convincing anons at large to use a hidden service seems nogo. Asking them to join a meshnet seems impossible - and they are "internet" people. Lainchan has a twitter instead of federated decentralized social media. This is disheartening. The future seems to be collection of "Deepnet", unindexed Apps (Discord, Yak, etc.) where anonymity is impossible, data corporate owned, and privacy is a joke. The situation seems objectively worse than status quo - at least you can filter JS and Requests in a non-google browser. >>14597 >Perhaps it doesn't even need to become popular. Maybe this program can be used more as a framework for other people to create their own sort of local intranets to share/upload data to. IMO best shot. Dominate a niche. Its impossible to share files now, because people simply would - You need Googles permission, and powers can always Kim Dotcom you. I think that free speech hasn't proven to be a popular platform with the masses, and so perhaps the ability to easily share files would be attractive. >15303 I wonder if we can do something similar to /g/ternet... How about us Lains make our own p2p internet like service? We could make a team or something. Or how about lainchan leaves twitter and endorses Matrix or IRC over discord? How about a lainchan hidden service? How about anons run there own self-hosted services before they rebuild the internet infrastructure ? Surely, this is an easier and more impactful course? Run a PirateBox drop box.

It seems as though, sadly, the original project is down. However; the premise remains. Perhaps any Lainons would be interested in working together to work out such a system from scratch? I'm wondering if anyone is still interested in this project. Thanks in advance! :)

>>23406 >Asking them to join a meshnet seems impossible - and they are "internet" people. If by meshnet you mean a physical meshnet, and not a virtual one running over the regular internet, then you're going to run into significant issues involving the laws of physics themselves. Transmitter power levels, signal attenuation due to foliage/buildings/etc, and line-of-sight issues are basically insurmountable problems for meshnets in all but the most densely-populated areas. I don't mean to be a downer on this. I love the concept. But let's be realistic.

>>24036 It is because of this reason I firmly believe the meshnet should be both physical for local connectivity and virtual for international connection. It's a shame this project seems to be dead because it held quite some potential as an alternative sub-network to safely transfer data on and its implications were quite interesting. Hopefully some other lainons ever see this thread and decide to pursue their own version of /g/ternet further building upon the ideas presented in this thread.

>>24036 Maybe not. Look up mmWave point to point. Not cost feasible, but nonetheless.

>>25390 is there any point to having a large virtual meshnet when darknets exist? they're both functionally just overlays but darknets give you anonymity too.

>>25531 This. Meshnets are a last resort project. At present, completely unnecessary with the technology we have.

>>25533 Ya. meshnets should be developed on the small scale as an eventuality, but otherwise they're pretty worthless.

>>25539 >>25531 Meshnets can be digital mailman systems. There are projects with the goal of connecting remote people instead of trying to beat the net in quality. There was a semi-popular project for sailors using gps instead of dns to give direction to proliferation. You could build a dense meshnet based on smartphones. People walking past each other would be sufficient to exchange short messages. I think you shouldn't even try to rival the net but gamify the experience. Make it fun and people will come.

>>25559 >smartphones huge battery drain, people won't be pleased. >gamify I don't think this would actually work. The thing we need is to significantly lower the barrier to entry.

>>25560 pokemon go. People already burn battery willingly for similar games.

>>25560 You say that 'gamifying' the process doesn't work, but as >>25561 already said, people willingly have their phone batteries completely drained for other similar purposes. It would also serve as an interesting experiment to see how fast data can travel when it can only do so in close proximity from other 'hosts'. I feel like this idea could be worked upon to make something interesting out of it.

How do I prevent cheese pizza from entering my node when I am using this?

>>26532 You do not. Cheese pizza is inevitable.

>>26532 The packages should be encrypted during transport anyway, to ensure you don't have the feds reading everything. All you need to do then is go with the Shaggy Gambit.

>>26534 The Shaggy Gambit...?

>>26535 Enjoy.

Bump. Is there anyone still interested in this project?

websites don't work for me, any guide on how to access this?

>>25559 This was how 5g was designed to be but the problem with it being a complete radio layer blackbox. Now if there was some source code leak for 5g routers and a 5g chipset, it'd be the best possible scenario. Though, wi-fi already works and is just as good. The only problem being regulations but there should be as signal amp boosting would prove to be harmful even for plant life. Maybe something like this. https://yggdrasil-network.github.io/ I think it's a better hobby than being a hamradio autist.

>>26543 >>26557 Although I still think the idea should be kept alive and further refined, the original project has sadly been taken down. I'll try to find the files again and put them back online again. The lainons in this thread have offered many new and interesting ideas, I feel like we could branch off from /g/ternet (which is dead anyway) and try work out our own form of this project.

>>14586 Good news, Lainons! A clone of the project is still online after all! The repository can be found here: https://gitgud.io/AMM/gternet-cli I suggest we clone it ourselves as well and work from there, slowly implementing the featurers we discussed in this thread.

There are so many of these. The issue is always that there is no content because of the small userbase. After the novelty dies down, there is no reason to use it anymore. A few IBs with 3 users and a dead forum isn't enough these days.

>>25559 Mobile mesh networks already exist. The GoTenna mesh is pretty large in itself. Probably only would take the right program to set up a mesh to do websites.

What about gnunet, I know it has been in development for a long time. >>26558 This one seems pretty cool'

IPFS+Eth token based name resolution exists now and is supported by brave and opera. Firefox is supported through metamask. Check out almonit.eth.link

>>26616 I am really liking the idea of a mobile meshnet, although I am afraid it'd never work with only the Lainchan userbase. And what'd we do if we want it to be internationally connected? Will we just have info spread by pure chance, or will we have dedicated links between large places?

Yeah, mesh nets look great but wouldn't that only work in cities? What about us poor peasants?

>>26698 That's what I am unsure about as well with the Mesh Net idea. It'd indeed make it hard to connect to isolated lainons. Does anyone have any ideas? I am personally afraid namely that we have no choice but to connect isolated lainons with encrypted connections over the internet, but that does defeat the purpose a little...

>>26698 > but wouldn't that only work in cities? It depends. If you're talking about cjdns/Yggdrasil networks, those are overlay networks. It doesn't require to work on top direct wireless connectivity. Users can even connect nodes over the telephone line, or any other preexisting medium. I think focusing strictly on direct connections is not a productive strategy here. It'd be much more effective to first build a large meshnet community over the Internet, eventually adding direct physical connections where possible. It needs a good kickstart and develop exponentially.

>>26689 Why would it just be Lainons? Knowledge of it would naturally spread. Just jump on an already well established mesh network and find a way to make it let you send and receive websites. Several already support audio calls, text messages and sending photos. A website isn't a stretch. It would have to be encrypted for security purposes of course. Though yes it's not easy to get it to spread enough it's functional but one way to do this is by in person meetup spots such as festivals, conventions, and hackerspaces where once there it will at least work in these small areas and once people disperse if they are in range of eachother enough it should maintain some functionality. Though I do have another idea. Instead of making it function like normal websites it could function like a subscription system that once subscribed to a site you want to follow while in range and it will automatically update that feed to your device whenever you're in range or at least should notify you if you don't want a copy locally saved to your device since too many subscriptions may take up too much room.

Zeronet. I2P. All blockchain based soykafs. Wireguard. Quantum computing. Quantum encryption. The history just began lainons.

>>26732 I tried using zeronet and it didn't even work to let me post on milhouse-chan or whatever it was called, it hosting CP on people's computers when they even see it aside. I'm sure that other stuff isn't problematic at all though.

>>14586 >decide want to click OP links >security risk ahead etc >go anyway >first link: "No hostname specified Sorry, but this resource is not available by this address. Please ensure: that you are accessing this website using a hostname and not an IPv4 address that your browser is sending the Host header, if you are using IPv4 that, if you support IPv6, you are using the IPv6 address. This server has virtual hosts set up on unique IPv6 addresses, since there are enough of them to use for this purpose. If you have any questions, please E-mail <opal at wowana.me>" Nice thread op. >second link security warning about certificate again then: "Not Found The requested URL was not found on this server. Apache/2.4.29 (Ubuntu) Server at git.gternet.me Port 443" Real nice.

>>15860 >>wug.za.net was last modified in 2010 and still uses php5 and mysql4 routines, we're busy updating as we've moved to a new server, be back soon..hopefully! Oh wow this is bullshit. Leaving the thread now.

>>26735 I posted this thread back when during the entire Net Neutrality thing. The original project has since been taken down, but a clone of it can be found here: https://gitgud.io/AMM/gternet-cli Besides, many lainons in this thread have offered new ideas that seem worthwhile as well. I'm thinking about making a start on it and perhaps making a new thread eventually. Hopefully something'll come out of this.

>>26715 >It'd be much more effective to first build a large meshnet community over the Internet, eventually adding direct physical connections where possible. This. This is how the internet became seperate from Telephony

>>26611 Exactly. Step one of mesh networks is not "post about it on imageboard with users scattered very thinly throughout the world". It is "set up a network in your local area with people you already know". Every successful mesh network has started this way, e.g. Freifunk, NYC Mesh, the Catalonia one, the ham radio ones, the ones in South Africa ones.

>>23385 >>24018 >>26572 One of the first contributors to this project here. Sadly this project fell apart since we were only a few active members and most of us had very little time. It would be very nice to get it going again, and the fact that there are still people talking about it seems to suggest that there is some kind of need for it. >>27053 Yep. We were working on making an easy-to-use install script for openvpn to build a little community, and then eventually replace those connections with physical links.

>>14609 Some of the suggested alternate internets would be fine with just a raspi. Others that need really heavy encryption, maybe not so much. A pi would work, but it'd have a hard time. The pi I use (an old Raspi 1 B) handles yggdrasil fine, except every once in a while the load spikes and everything crawls for a bit. my raspi can run i2pd and tor, but running both at once is not good. i2pd also required adding a gig of swap otherwise it ran out of memory during compilation. I use wireguard and quicktun for most of the point-to-point links I have and they barely use any resources.

>>27076 Well, let's discuss how to do that, then? What do you think is a good way to present the idea of a local mesh to people who aren't computer nerds? The people I know aren't nerds except for the ones I know over the internet.

>>27076 >>27108 I outlined a strategy here >>26731

I think its obvious that we will have a problem attracting new people, especially ones who aren't very comfortable with computers, if we don't have something tangible to show off and at least a few active users. We have no chance of making an actual physical mesh without a good number of dedicated contributors. I propose that we build a quick POC mesh with VPN connections and try to iron out the details while recruiting more users. Making a physical mesh would also require quite a lot of money. I don't think any one of us are willing to spend on a project before seeing that other people also are somewhat invested. Of course we could just join an existing mesh, but that isn't really the point of the project or the thread.

>>27115 individuals can also take pirateboxes around with books and papers on cybersecurity and anti-jewgle stuff to local hackerspaces, makerspaces or anywhere else with people who might be interested in a mesh, and include a text file about the mesh we want to build and how to join.

>>27127 I described here >>26731 a way of using a system like pirateboxes as a mesh net. You just set up websites or RSS feeds that update automatically to subscribers of those sites and feeds. Does that make sense how I word it?

If this project is going to go anywhere I think we need a place to discuss this more in dept. There is an existing dead irc channel with around 10 users and I have some emails addresses saved, but the important thing is that we organize ourselves in some way. Do we go with the original channel or are there better ideas?

>>27159 How about a matrix room?

>>27159 Xmpp

>>27159 It would probably be a good idea to find a new medium of communication. Matrix would be a good choice. I think it would be wise to rebrand from gternet to a new name. Gternet isn't a very good name and will always just be an abandoned 4chan project.

>>27173 Easy enough to bootstrap a mesh overlay network with OpenSSH tun interface forwarding. This obviously wont scale.

>>27251 Wireguard would be a better choice. It's more robust than SSH and more intrinsically point to point than openvpn. You could run BGP between nodes just like a hardware link. The hard part is assigning names and numbers in a decentralized way. IPv6 only might help with this, if only for the fact that the private use address block is bigger than v4's 10/8, but you still have the fundamental issue of consensus.

>>27266 anonet has just been using a git repo to keep track of what numbers are claimed by who. The repo just gets pulled directly from whoever you know that has a copy. You add your own claim, commit, and then git daemon so someone can pull from you. It isn't "secure", but if someone is causing a stink it isn't like we wouldn't know who is trying to cause the stink.

>>27266 if it is going to be another BGP-based network, then it doesn't really matter what each link uses as long as both ends want to set it up. Just get something set up first and then do upgrade later. There are going to be bugs you can't plan-ahead for.

>>27286 If you're going to use BGP and IPv4, you might want to consider whether you want to be compatible with other networks. Like, if you wanted to be compatible with dn42 you might try to register a range of ASNs and IPs with dn42 that can be used however you want for lainnet distribution (instead of getting each lainnet node to register their own dn42 resources) If you don't want to do NAT for dn42, you'll need an RFC1918 range, but most of those are claimed by various other networks. I usually check this list: https://wiki.hamburg.ccc.de/ChaosVPN:IPRanges You'd have to have an active dn42 peering to see exactly which ranges are currently in use right now.

Interesting

>>14586 I found an interesting article about how computing is structured, and how the structure of computing impacts privacy and freedom. It argues that desktops will fade in importance and hobbyists must fight to preserve them. >http://www.winestockwebdesign.com/Essays/Eternal_Mainframe.html >In the computer industry, the Wheel of Reincarnation is a pattern whereby specialized hardware gets spun out from the “main” system, becomes more powerful, then gets folded back into the main system. >The mainframe outlasted its replacements. >So much personal data in the hands of a small number of corporations presents a tempting target for governments.

>>27286 >if it is going to be another BGP-based network, then it doesn't really matter what each link uses as long as both ends want to set it up very true. >>27287 iirc there is a standardized testing/local/whatever asn that's functionally like the v4 10.0.0.0/8. also asns aren't that important to the functionality of bgp. they exist basically for convenience of defining routing rules because in a production setting different asns will correspond to different metering, so you might reject routes that come from an asn corresponding to a transit provider if you don't want to pay for transit. if this were one big network there's really no reason to not just have everyone use the same (dummy) asn just to make the bgp software happy. if you wanted to peer with real networks you'd of course need a real asn and real ip prefixes but doing this would totally defeat the point of having an overlay network in the first place. although i have to say i don't understand how a mesh of point to point tunnels is preferably to something like i2p. restricting access to resources on the network to only verified peers is the one thing it can do that i2p doesn't but it doesn't scale at all. i guess it could in theory be faster than i2p since packets will try to take the shortest path between nodes instead of being bounced around by onion routing, but i2p also has more nodes so in practice i doubt this would be the case. every other feature i can think of re: security/privacy/ease of management and use is either matched or surpassed by i2p. i think a non-onion point to point overlay mesh would be better suited to a small community of people instead of trying to replace the internet. if we lainons have enough hosted resources to populate such a network it might even be fun and productive to try it (i'm sure many of us have a cheap vps we could hook up without too much hassle, even if it isn't exclusive to the mesh net). i like the idea of a small lain-net with a couple dozen personal blogs, a git server, maybe some file hosting and chat services, etc. >>27285 is that basically what anoNet is?

What is your opinion on GNUnet? Seems to be a million years from a usable release. And it's hard to discern if there is ongoing progress on the project.

net neutrality is ass

What makes this any better than Tor? Can't some glow-in-the-dark just come along and set up their own node, then monitor the traffic?

>>27436 Nodes can encrypt E2E and transmissions can be p2p. Its not better than tor, it has a different threat model: Really easy to do traffic analysis on p2p networks. But, a meshnet could be bridged to it physically if it were big enough.

>>27406 >GNUnet i was impressed by the technical documentation on their website although i also can't really tell how usable it is at this point without rolling up a node myself. at least based on that it seems purpose built to enable a gradual progression from an overlay network to a hardware-supported microwave mesh. this is all due to its flexibility wrt which network layer its operating on (and wrt which protocols you can run on top of it).

>>14593 I think if things get to the point an internet alternative is necessary than popularity wouldn't be a concern, but rather having the network set up to allow people to trade precious information. If people talk about it with each other in real life it should be enough to keep it up, as of now it would die because there's no incentive to use it instead of the regular internet. And let's not pretend that it's the average user who makes the internet run, they're only important for websites looking for big traffic, like Youtube.

So, a total outsider looking in here, to me the concept of a wireless mesh network seems pretty flawed, I would have thought that a system where everyone that wants to host/access /g/ternet should just connect to something like a VPN based lan, i.e, all traffic traverses the openweb in encrypted VPN tunnels and all clients must host a basic form of router that is hosting and connecting to the next node in the chain via a VPN. I know very little about networking but what I'm proposing is sort of a large scale campus LAN architecture that traverses the open web via VPN connections, possibly with hosted "internet" nodes that allow users to connect to the openweb too.

>>27661 what you're suggesting is kind of orthogonal to a wireless mesh. the whole point of the latter is to eliminate ISP hardware infrastructure

>The project can be found here: >https://git.gternet.me/gternet/gternet-cli DAY OF THE SEAL SOON

>>27159 >>27160 >>27173 Seeing as Lainons actually want to work on the project, we do indeed need a better means of communication. I can make a Matrix room. What should I name it for now, and should I make it public or private? Your call, Lainons :)

>>27701 Why matrix? It is bloatware and XMPP does everything way better.

>>27702 xmpp is impotent and has clearly failed in mass adoption. Matrix is the key to bring security and anonymity to the masses.

>>27701 What services will we run on it ?

>>27702 >>27718 >>27701 practically speaking whoever actually hosts the thing first gets to pick the software

>>26650 >set up website and host it on IPFS >try to register a domain using ENS >over $150usd a year to do so no thanks

We should all consider running LibreMesh on old routers with Yggdrasil. Even if we don't pay much attention to them after the fact, every little bit helps. Essentially what this would do is give us an alternative to corporate and expensive internet we have today. https://jnblog.neocities.org/running-libremesh-wo-internet.html

>>27701 Make it public first to see how it goes

>>27163 >it is bloatware That's Riot, not all Matrix clients.

>>27960 You forget Matrix servers.

Sorry, lainons. But what is the problem about I2P? Why there are so many these similar solutions, what (new) problems did they designed to solve?

>>27968 >Why there are so many these similar solutions, what (new) problems did they designed to solve? They rewrote it in Go (but forgot to fix the part where the entire system depends on webshit).

>>27964 Do you mean the servers are bloated?

>>27857 What's the deal with yggdrasil ? I mean, what does it enable you to do ?

>>28000 Synapse is rather heavy and runs on the JVM. Dendrite is a lighter reimplementation in Go (iirc) and supposed to replace it, but isn't "production ready" yet, according to the matrix foundation. Also, anything is going to look bloated compared to IRC or even XMPP, so there's that.

>>27980 > They rewrote it in Go (but forgot to fix the part where the entire system depends on webshit). Well, I don't think it is "webshit". I can't see too many disadvantages.

All of this is fine and dandy, but what about the existing projects? Are ipfs, yggdrasil any good?

>>28174 Also, why not just I2p or freenet? What's wrong with those? Genuinely asking, I'm pretty ignorant on this subject but I want it to worj

If this happens, the way of the BBS will make a comeback. What are your thoughts on this, Lain?

>>28182 If we pretend it's actionable, get ready for the end of any kind of community. Also a great time to be a lawyer I guess

There is also Zero net and Beaker Browser, just so people know.

>>28192 >Beaker Browser I wasn't aware of this project, how's it like? I'm wary of it because it uses chromium for its engine.

>>28000 100 MB vs ~1 MB RAM per user allocated on Synapse and ejabberd respectively, go figure the conclusions.

>>28014 It is pretty much cjdns (an IPv6 overlay network) except that yggdrasil also has a range that can be used on your LAN so you don't need to install software on every single machine you want to be able to communicate with the network.

>>27744 Alright then, how about... point an IRC client at zpm7mpgxat7tyck3edpvkv4eiyahixusahxqvf7367rhrujkguvefyid.onion port 6667 and channel... #altnet with channel key of lainchan. the way the server behind that port works is that it won't show join or parts, so make sure to say something or nobody will know you're there.

You're doing this all backwards. "Build it and they will come" is bullshit and that's already been acknowledged several times in this thread. You should focus on community first, platforms second. No, you don't need a specific platform in order to have an online community. You can build a manual network of individuals connected only via email and chat, and who use whatever platforms they find convenient for the specific activities of that community.

>>28639 I know that might sound a bit old-school but what you loose in convenience and automation you gain in flexibility and freedom. It has to be based around something other than giving each other head all day so you're most likely talking about a file sharing community for starters. Such a group could mass-adopt whatever platform, protocol or alternative network they wanted and make it popular overnight. It would need its own economic engine to draw more people in and expand in scope. I've thought of a novel way to use cryptocurrency in such a community which could attract businesses and incentivize long-term investment. I'll go into more detail but first I want to make sure this thread isn't dead and I'm just talking to the wall.

>>28640 Please, go on, im not very friendly with criptocurrency things but the general idea sounds interesting.

>>28662 Well let's say we hypothetically form a rag tag community, connected across chat clients, webmail, retroshare, etc for the purpose of file sharing. Something manually built like this, using multiple forms of contact, can't be taken down just by taking down a specific website. And by "file-sharing" I certainly don't mean piracy, of course. That would be like downloading a car and I would never do that. However the same idea could be applied to a piracy community just as well as it could to a legitimate file-sharing community. So anyway.... These kinds of informal networks already exist but I want to build one on purpose and with larger intentions in mind which go beyond its initial purpose of just file sharing. That's just the seed activity. Instead of just sharing files freely or even trading them I want to introduce crypto, but in a way that's easy to get into, gamify's the experience, incentivizes even more file sharing, and includes long-term financial incentives as a bonus. For this to work we would have to agree to adopt a crypto that has either zero transaction fees or else transaction fees that are so small they are negligible. This is so we can trade them in minuscule amounts like in the thousandths or millionths. Something like NANO or IOTA. We adopt a crypto like this, and agree to treat its value as being insanely higher than what it actually is. Maybe 0.0001 NANO = $1.00 USD. So now when someone requests a file, you can price it according to what you think it's "real world" market value is, and convert that into what you would charge in NANO. This makes it really easy for newcomers to get their hands on enough currency to participate in the community. Price the currency at just the right value that they can gather up a reasonable amount in a few hours of mining or collecting from faucets. They don't have to fuarrrk around with crypto exchanges and all the pain-in-the-ass verification requirements that go with it. And since the amounts we're dealing with are worthless anywhere else besides within our network, the money will stay within the community and circulate. Since users will be able to buy things for dirt cheap that they would normally have to pay a lot more for, this will create an upward pressure on the "real world" value of the crypto we choose to adopt. This is the long-term incentive I was referring to, since before that happens the network would have to grow substantially until it became a financial force in its own right. But once it does, those few hundredths of a NANO that you earn from file sharing will begin to pay major dividends as their real world market value climbs. It gamifys file sharing, which is another key point. MMORPGs have their own currencies too, and they took on a value all their own which ended up spilling outside the scope of what they were intended for. A black market formed to facilitate the trade of game money for real money. People even started using it to launder their cash or pay their bills. Later games like Second Life even adopted this into their business model and actively allowed their user base to profit from the in game currency. I want to translate the business model of Second Life into a file sharing community. This is what will attract other types of entrepreneurs, such as content creators, coders, various merchants, and expand the scope of our file sharing community into other areas. And finally, since this community would exist outside of any particular platform, it could adopt whichever platform, mesh network, darknet protocol, etc which best suited its purposes. It would transform the development of an alternative internet into a profitable venture instead of just an activist venture since a self-funded community can pay for the cost of its own network infrastructure.

>>28640 It already exists https://althea.net/

>>28777 Althea is a decentralized system of routers, but I propose we build a decentralized userbase. If you have friends on an onion site and other friends on Retroshare, they can't talk to each other. But they can still both talk to you. I think we've neglected the potential of a human to human network. We implicitly assume that we need some platform or automated protocol to streamline our interactions but it also limits our interactions. For instance you need to be close enough to another node in order to use Althea. I'm proposing we take something that millions of people already know how to do, just do it a little differently, and use it as the basis for building a cross-platform, cross-protocol, cross network network.

>>14590 How does one get access to this? the git link doesn't seem to work for me.

>>14586 >>10088 >http://archive.is/SJ9vl >This is Fine: Optimism & Emergency in the P2P Network <The last fifteen years has seen a surge of interest in decentralised technology. < Decentralisation advocates roleplay as antagonists for change, but they have yet to truly threaten incumbent power. Instead, the de-prioritisation of privacy by design – regardless of its justification – enables its behaviour and offers it new scope for surveillance and control. An Essay on how use of decentralized networks has before via Bit-torrent enabled surveillance that benefits dominant centralized authorities and how current non-E2E p2p products are enabling surveillance on under-informed dissidents. What do you think Alice?

>>29760 I think they severely under-estimate how difficult it is to provide anonymity on a network as massive as BitTorrent.

>>29763 Well, as factually stated in the article is that adoption of p2p protocols by unwitting users resulted in massive lawsuits that harmed them. However eady it was to provide, it wasn't provided.

>>29764 that's the thing. It is not easy at all to provide. Look at i2p and all the complexity they have had to add to make their network both anonymous and decentralized. Look at Tor and how they have had to make their entire network hierarchical to achieve acceptable latency and throughput. Look at Freenet (which by the way, they cite as an example of how reader privacy is "easy" to implement) and how it basically becomes useless the instant you turn on DHT mode. I've studied a lot of these systems, and from what I can tell, providing decent reader anonymity while also operating at significant scale is an extremely difficult technical challenge.

>>29765 Where do you study these systems? Where can I find resources?

>>29767 read the documentation of the different systems. Tor, i2p, and GNUnet in particular have very good documentation on their websites.

>>14590 So it's anonet.

>>27744 Why not just discuss this in the lainchan irc? It's already there.

>>29952 And it's not like too many people connect to it in the first place.

>>28639 the platform matters a lot, the people who would only join a Discord are not the same people who would join on IRC

>>26732 I like the idea of P2P networks as an alternative to the internet we currently have, but it has multiple issues that make it undesirable to most people right now; >Speed Compared to what we are used to, P2P networks are very slow and fluctuate all the time, if a website isn't popular at a certain time of day and fewer people are actively browsing and uploading files for it, you will have a noticeable decrease in speed for every aspect of the website. >Downloading The website you're using may have to be downloaded onto your PC which can take up precious storage space, this might be insignificant for people who have a lot of storage, but if you want to run a thin-client and use P2P networks, this could be a very real and serious issue. Though in all fairness, it may be possible just to offload that issue onto a server. >Security P2P networks have multiple security issues, because you are required to download the website you are visiting, if there is any malicious code or potentially illegal data in the website you are visiting, your PC may become compromised or you might face legal consequences for having any illegal data on your PC. I am not sure if these issues have been fixed or mitigated since I last P2P networks, feel free to correct me if I am wrong, but assuming that they are true, I really don't see how you could convince an average person to use P2P networks any time soon.

>>30752 there's so much soykaf in this post I don't really know where to start. - there already are fast p2p networks, best one so far being lokinet. - you download (and execute) a website, or as they like to call it "webapp" when connecting to a regular webserver too. hint: it can be buffered in ram, no need for "precious storage space". - the security aspect already is true, visiting a website unprotected is dangerous because js and css (malicious by design), and other weak implemented soykaf.

>>14586 The best odds we have at producing an alternet is by coordinating with existing projects and joining up with local meshnets and hackerspaces. There is a hackerspace run alternet internet used mostly to learn bgp. Perhaps this would be interesting. An alternet internet is possibly very important with the Trump Administration's "Clean Network" , and Russia and China's "Internet Sovereignty" ideas. We are about to see the balkanization of the internet.

>>30788 CSS isn't bad though? It's just a style sheet. It doesn't do anything.

>>31013 it does. check out css-keyloggers as the most simple example. also css is turing complete. a bit too much for a "style sheet". it's not as bad as js but still worth looking at its flaws.

>>31027 >check out css-keyloggers is this blocked by ublock hosts?

>>31013 >A never ending extending pile of features centered around the vague goal of transforming the document in "some way". what could ever go wrong

Are there actually people left who want to build an alternative infrastructure to the Internet? Where do they meet and coordinate? I have some things to share, but not sure if this is the right place.

>>31277 Everywhere is the right place. How could you not benefit by sharing your ideas about network infrastructure. For now, it might be interesting to see a lain mesh VPN logical overlay perhaps on 100.64.0.0/10 but honestly, a /28 would probably suffice for all the activity here. Mesh point-to-point vpn via sshd with BGP should be called The Wired. One day, lains bridge net over analog radio.

>>31038 >A never ending extending pile of features centered around the vague goal of transforming the document in "some way". This is the best description of CSS and generally the web I've ever seen. In my opinion HTML, CSS and JS should be replaced by one, general purpose programming language. Scheme would be perfect for the task, because in Scheme program and data are the same.

>>31307 >In my opinion HTML, CSS and JS should be replaced by one, general purpose programming language. >Scheme would be perfect for the task, because in Scheme program and data are the same. fuarrrk no. JS and documents as programs is already a huge mess for privacy and security. Documents should be declarative. CSS is a laudable attempt at abstracting visual theming from content.

>>31308 >fuarrrk no. JS and documents as programs is already a huge mess for privacy and security. This is mainly because of proprietary software. Web pages already became proprietary applications with DRM and obfuscated JS. Does it really matter at this point? Making one language for everything would shrink web browsers and allow more competition. >Documents should be declarative. Can't it be declarative with one language?

>>30789 >The best odds we have at producing an alternet is by coordinating with existing projects and joining up with local meshnets and hackerspaces. Damn straight. But that means the underlying tech can't be our main focus. Smarter people than us are already focused on the tech. Our focus should be on uniting the fragmented user bases of these various disparate projects. We do that by building communities that span across them that are based on shared interests BESIDES the underlying technology. Realistically that probably means "file sharing" communities of one kind or other. And definitely involve money changing hands. A community that has an economy, no matter how basic, has the potential to grow and evolve under its own momentum. If anyone wants to start an "informal network", let me know how you'd like to get in contact. I need to see actual interest before I'll post any sort of contact info.

>>14586 Ever heard of Reticulum? It's a fully encrypted mesh network designed to use LoRa, WiFi, ethernet, tunnelling over IP, etc... It even has an associated open source hardware LoRa transmitter you can buy for $70, range is up to 15km at 2kbps, with up to ~40kbps at closer range. And someone even already designed a low-overhead BBS/USENET/Messager library that runs on it, seems pretty simple to use it as an imageboard, though filesharing would be pretty slow depending on what kind of connection you're using. The mesh network program - https://github.com/markqvist/Reticulum The LoRa transmitter - https://unsigned.io/projects/rnode/ The BBS - https://github.com/markqvist/lxmf Seems like an excellent foundation for a network since it can bridge between all the available network interfaces you can start out with connecting over regular internet and then branch out into WiFi, LoRa, etc...

>>31434 What services could lainchan operate on an informal network? There are posts here about Wiki's, and there already exists IRC, there is also a radio stream. We have a zine and imgboard. It would be cool to mirror some of this stuff on a overlay network, perhaps an SSH VPN. I would host a copy of my blog on "the wired". You could just take the 100.64.0.0/10 space, use BGP for routing, and use a hosts file. Lains could then take piratebox kind of APs and bridge them to the wired. Then your neighbors discover SEL. ??? Profit? SSH MESH VPN would be interesting. There are a couple of Prior Art Overlay networks with BGP routing that are popular at hackerspaces whose names escape me, that we could join.

>>31454 I wasn't thinking about lainchan specifically. I was thinking more about forming individual connections: Building a manual, human powered overlay network. One of the main reasons the web has turned into such a botnet is because we've forgotten the art of doing that. We've outsourced it to platforms. I think people under-estimate the potential behind going old-school: connecting via email and private chat and talking one-on-one. They think platforms and social media are the only places where anything interesting or significant can happen. But I think a manual network, if done right, can be far better. And it can indirectly influence the established tech giants to be better as well. >>What services Whatever type of network you're trying to build it needs to serve some kind of purpose other than soykaf-posting. That's why I proposed a file-sharing community since this implicitly includes a wide range of people with diverse interests. >>Lains could then take piratebox kind of APs and bridge them to the wired. Plain old encrypted email/chat is a hell of a good start. You'll find it hard enough just to get people to contact you privately and directly, let alone get them to mass adopt some kind of obscure technology. Start with offsite individual connections first.

>>31316 >Making one language for everything would shrink web browsers and allow more competition. No. >>31308 is correct in saying >Documents should be declarative. One language would result in a _programming_ language being used over a _markup_ language. The current problems exist _because_ of _programming_ languages being introduced.

>>31441 This looks sick.

>>31441 The 15 km range of this thing is incredible, this is easily the best meshnet option. You could use this to access a node from anywhere in the same city. This will probably be used for SMS, IRC, email and HTML as soon as people get LoRa transceivers.

>>31548 15 km is great too, it makes it actually usable for rural areas.

WRT. webshit, what are Lain's major gripes and proposed solutions? Please be specific. >tracking The web is about documents, and documents shouldn't be able to phone home. Probably the most straightforward way of dealing with this is not allowing the website to make any network requests. Of course, this begs the question of what should take the place web applications; a return to the desktop would just mean further proliferation of 'apps'. >mutability Links break or change their contents arbitrarily and HTTP headers act as a huge fingerprint. Addressing files via their contents a la IPFS would seem to fix these issues. >Javascript abuse/SPAs What counts as abuse? Where's it coming from? On the one hand, I want a webpage to be able to render a model or show a simulation, but on the other I'd like not to have to deal with devs overreaching. Scroll-jacking, breaking the back button, breaking right-click, triggering expensive repaints, "blank" websites. Would just removing features from JS be sufficient? I feel that the problem's deeper than that somehow. >JS/browser vulnerabilities Aside from data exfiltration, it seems the problem is that JS suffers scope creep and is far too complex to implement safely AND performantly. Reducing the scope is easy, but how could creep be discouraged? For safety and performance, a simpler language or VM could be employed, but I've actually had a hard time finding literature on languages designed to run untrusted code. >bloat/load times Page load times have ballooned over the years. According to the metrics I could find, it appears to be down to an increased amount of media in the page, slow Javascript logic and 3rd party resources like ads. Media could potentially be dealt with using progressive formats even for images, and 3rd party resources could just be blocked unless they're static - but what about slow logic? How does one enforce performance? As far as I can tell, there aren't even languages that calculate asymptotic complexity, let alone estimate performance. Perhaps the logic equivalent of a progressive format? How could that even begin to work? >ineffective URLs Since Javascript has its own state, it's possible to see two different pages at the same URL. Therefore, webpage state should be reified in a single location that can be incorporated into the URL. A simple implementation of this would be if Javascript worked as reactive transformation rules over the webpage, which would be referenced by its hash. I want to be able to say more, but the whole mess is far too complicated. Also, most suggestions I've found don't seem to propose anything solid that isn't a regression to plaintext - ignoring the problem instead of facing it.

>>31441 I've been very excited to see a few LoRa projects like this sprout up; p2p mesh text-based networks are just the throwback to pre-WWW internet that my soul craves. And as someone else said, it's great for rural areas. Kind of feels like a blessed fusion of oldschool internet community and ham radio. In fact I bet I could build a hell of an antenna for it. My question: does anyone know or have good reason to suspect that one of these projects is starting to pull forward and gain more momentum than the rest? Because of course a mesh network would massively benefit from having the most early adopters

>>31576 >that isn't a regression to plaintext - ignoring the problem instead of facing it The regression is a feature on its own. Keeps it pure. Providing rich multimedia doesn't actually help most internet communities. I think that it's less a case of ignoring the problem and more a philosophical disagreement on what the problem is. Just my one-and-a-half cents.

>>31578 >Providing rich multimedia doesn't actually help most internet communities I agree, but it's not just community that one needs to consider. I mention in my post the desire for rendering 3d models and simulations - something that can significantly contribute to education. While desktop software can fill this need, it is arguably less secure and encourages fragmentation by reducing portability. Styling, too, provides significant utility. Good presentation is essential to efficient navigation and comprehension. By having a universal styling framework, effective presentation can become more widespread (assuming a clean enough model-view separation). >philosophical disagreement on what the problem is That's reasonable, I suppose. I just feel that in general, we should want to empower the user not only relative to exploitative forces, but in an absolute sense as well. Plaintext, by virtue of its limitations, only provides the former. Also I fear leaving multimedia to DHTML will end up biting "us" in the ass when we inevitably need or want it, since Gulag and co. will have had plenty of opportunity to make the web worse. Already with Firefox falling out of the race, I'm concerned for the future of the Tor browser.

>>31584 I certainly share those concerns. Maybe what we're looking at is something requiring a few solutions instead of one monolithic solution. I do think that for now, keeping it extremely light and low-bitrate will help the mesh establish, since it will require less nodes and be more likely to perform all the functions in its scope reliably. I think that's the first big hurdle: getting a network effect to bloom and grow so we have an actual thing in the real world to work within.

>>31586 And then once there are enough nodes in the wild, higher bitrates and more complex applications might become feasible

from my experience trying very many mesh networks, the only one that actually works without internet and has a lot of people using it is ham radio

>>31588 The painful truth. And it's all just a bunch of extremely ornery old men jawing about their goiters most of the time

>>31586 >I do think that for now, keeping it extremely light and low-bitrate will help the mesh establish Definitely agree. The slimmer the tech, the more places it can spread, and we need to spread like crazy if there's to be any hope of real freedom. >I think that's the first big hurdle: getting a network effect Definitely. But what will be the "killer app"? Perhaps if we ever get trustable hardware, a holistic "product" that's marketed as glowie-proof would see success. I also feel that achieving a truly independent internet where compute as well as data can be fully distributed and therefore border-less would be a major win. Finally, maybe some organisational win like what Google search was crossed with what Xanadu promised to be. Regardless, I hate the idea of relying on webshit pushing people out; such a passive approach lets them control the playing field.

>>31576 >The web is about documents Right now I'd say that the web is about services. More and more functionality has switched from clients to servers, because companies don't want you to use (middleman) software. Phone apps are this, most of them are just frontends for a specific service. Browsers are becoming less and less relevant since less and less people care about navigation. In the future you will ideally not even have to click on a link or fire up a search engine to get on the site and then shop, you'll have everything built into social media or the device. I'm digressing but when I look at normal people I realize that this whole concern is about a side of the internet that's already dead and buried. normal people simply don't use the internet how I or anybody else here uses it. They only search something on google when they need a service e.g. buying a thing, and their only concern is finding the service as easily as possible in the first results and installing the app from the site. They hate browsers, they never use them. I'm sure that companies hate browsers, too, and they're doing their best to find ways to remove them completely. The ideal endgame is that when you "go" on a site, your phone downloads (with high speeds) an (approved) app and you tap OK once to sign in with Google, then you shop on it. This is probably why cancer like Electron is growing despite being garbage. My 2c

>>31577 I don't think any of them are particularly popular, it's all in the "attractive features" stage. Probably the most important feature is translation, assuming there'll be a bunch of different incompatible mesh networks, the one that can bridge all of them will eventually predominate.

>>31584 How about something like a rump kernel, but for browsers. Design standalone "apps" that are also defacto part of a browser. Then instead of cludging everything into one framework language, you just make more "kernel modules", and encourage people to make all efforts to stick to the default set. Kinda like a more intentional version of flash plugins, make the WHOLE BROWSER out of plugins, let anyone make their own, and then leave it up to the consumer/package manager which ones are enabled.

>>31441 >maximum network propagation range 270 kilometers >sender anonymity, but no server anonymity >separation between station(routing/stationary) and peer(non-routing/mobile) nodes Almost but not quite. Sounds great for a low-cost/low-power WAN, but lacking on some important features if you want mobile comms, anonymous hosting, regional networks, etc... That said, I don't know of another option better for low-cost, plug and play mesh networks. Could use it for IoT/home automation really easily with no insecure proprietary soykaf.

>>31595 Probably blockchain propagation or Freenet. Reticulum would be amazing for Freenet nodes, because they are designed for partial meshes and asychronous sneakernet darknets. You could even do realtime voice over 5 kB/s. It's realtime, local and free, the possibilities are endless. >>31577 Piratebox is the most popular and fully developed meshnet so far. Serval is still WIP. Wind is WIP. Locha Mesh is in preorder stage. Netsukuku is finished, but not widely deployed. Tinc is complete and operational. Packet radio over the ham was shutdown before they could present at DEFCON. Freifunk has 41000 nodes. Red Hook Wi-Fi is operational in Brooklyn. NYC Mesh has 200 nodes. Ninux has 352 nodes. sarantaporo.gr has about 234 nodes. Guifi.net has 35000 nodes. Peoples' Open Network has about 100 nodes. Village Telco is operational in 3 cities, more in progress.

>>31588 Speech recognition has advanced so much that you can control computers over ham, just by saying command phrases and passwords.

This might be to interest here; https://nethood.org/ > NetHood is about bridging the digital with the physical space; sharing ideas and experiences, affirming differences; claiming our right to the hybrid city; building together local solutions for local needs, including: DIY networking, neighborhood catalysts, complementary currencies, cooperative housing, and sustainable food culture. It looks a bit inactive atm though, but there are some nice projects listed.

>>31689 Wow. This entire site reads like fluffed out nothing. None of the projects describe an actual goal, the only point of interest was the discussion of "complementary currencies" but it was thrown into a salad of other subjects so I doubt how focused this discussion would be. And judging but how many times this project stresses words like "diversity" and "plurality" and other buzzwords I am certain every counter boils down to reciprocal fart-sniffing between the organizers and a bunch of snowflakes trying to leverage for money or power over their city.

Build an informal network composed of direct private connections. Base it around file-sharing. File sharing implicitly includes a wide range of people with diverse interests. A human-powered network like this can serve as a compatibility layer between multiple platforms and architectures including mesh networks. It's a social approach, not a technical one. You're not going to out-tech the Tech Giants but you don't necessarily have to. Instead build a manual network that makes use of existing infrastructure in unconventional ways. Think in terms of guerilla tactics instead of beating them at the very thing they're strongest at.

>>31616 you can broadcast digital signals over ham. it just has to be a public protocol which isn't encrypted. there's a good defcon (i think) talk where a guy figured out how to legally run an imsi catcher on ham frequencies.

>>31763 Excellent, then you can use chafing or steganography to provide the same level of security as encryption, with only a small bandwidth penalty.

>>31793 technically you can't, because of the public protocol thing. i believe the requirements are that it is "well known" and published somewhere. if you're willing to break the rules of course you could figure something out but that's what the FCC requires.

>>31795 Chafing and steganography are free protocols, but with steganography, they cannot prove that transmission is secured.

Thoughts on Lokinet / Lokinetwork / Loki crypto? The messenger Session seems to be getting a lot of traction as an alternative to Signal, seems very promising (like a better version of I2P) https://loki.network/ https://lokinet.org/ https://loki.foundation/session/

>>31799 just another cryptocurrency bullshit

>>31801 I haven't used their crypto currency so far, but the onion routing and session messenger are awesome. (Even though the latter is still experimental)

>>31803 The project as it is might be cool, but the integrated crypto currency is just for the hype. Not more than a simple buzzword, they could also suffer from a 51% attack, the marketcap is 32mio$ for a attack in the size of a country this is peanuts. you better stick to networks without that "value" they add with blockchain-technology

>>31804 PoW blockchain is the only effective way to counter Sybil attacks, which Tor, I2P and Freenet are vulnerable to.

>>31912 Tight tight tight Is there a way to turn off the artificial brightness overglow/scanline effects/screen distortion etc? So I can just jack it into a legit old monochrome CRT and use the real thing :) Suppose for that I could just go with Gopher as someone else has said. But this is cool, thanks for the rec

>>31916 They had created their own thread about it after posting it - So I deleted it in this thread to not split up any discussion. >>31913

>>31917 Appreciate the heads up :) thanks

>>31891 it is not ... a state can just buy enough GPU power to have 51 % of the mining power and your at the same point as before internet is not free anymore, we should better accept this and search for some real solutions towards this problem

>>31924 The solution is encryption for security/privacy and hashing for verification. Everything else is secondary. ipfs is a really cool framework...But the amount of work needed to make it actually function in the capacity that is desired is pretty insane. Now I have come up with a strategy as a temporary solution for ipfs and its security flaws, but it does not scale well. Not that it would matter on any near term, to think it would really be anything more than three autists is likely self-absorbed. But ipfs has one major flaw - It does show the IP address of every user. The work around is using something like vpn/vps/etc. and note that it does not work over tor (for now) -- Although, there may be a method to force it anyways. Those are my immediate thoughts on it, where my biggest area of interest is the cryptology aspect.

>>31924 I invite any state to try a 51% attack against the entire world.

>>31943 it's not any state against the entire world, it's any state against a bunch of neckbeards.

>>31943 >>31953 is right, crypto/privacy and all of this neat little things are valuable for us, but not the majority the majority will not use linux (i mean privacy orientated, not the android-scam), the majority will not use a cryptopayment and the majority will not use any obscure network to hide their traffic privacy does not matter to them, otherwise they would vote the same dickheads as every year or would inform themself about stuff

>>31924 Internet is just a giant infrastructure. Just wait for wifi6. As wifi routers get more powerful, the more potential for independent network there is. 5G could've been like this but then freedomland didn't want proprietary and unhackable wifi-like communication tech in their soil. *sad reverse engineer face*

>>31953 If you are small group, use Dining Cryptographers or sneakernets. Form cells to resist state oppression. >>31959 Privacy does matter to average person, they do keep a lot of information secret. Just try asking them if you can take their phone and read through all of their private messages. They just do not understand how to actually preserve their privacy in present day, because they fundamentally do not understand how technology or science works.

>>32005 I'm glad you brought this up. I think there's unfortunately a very popular idea that common people are stupid and want to be exploited. They really aren't! Attend a sports bar during a big game sometime. Watch the people around you—it might not be your cup of tea (and indeed it may be something you look down upon) but please recognize that common people use strategy all the time. They aren't stupid, they just don't have certain kinds of specialist knowledge that matter most in this recent and emergent technological culture. I think that's forgivable, considering the learning curve and the fact that most of them would have had to be prophets to see it coming decades ago when they would have had to start that training process. Maybe it will change in the future. But they aren't stupid. They're just specialized in the wrong things. And people want privacy back. They really, desperately do. We're all debased and dehumanized by this current infoculture. But most people don't see any way out. I'm not sure any of us see true ways out for even ourselves, let alone the greater culture. That's the challenge for us all in the near future: how do we even fight back? And I mean effectively at scale, not symbolically or just personally. Maybe remind yourself that you can think strategically as well. Be humble enough to realize that all the people around you might have novel ideas that we might miss because we're too deep in the net. Maybe there's a strategy outside that could have a chance. That's just my hunch, so take with salt of course.

>>32006 There is a simple idea to prevent us from being exploitet. Shut it down. As soon as you have soykaf like social networks someone will use it and the operator wants to make money with you. You would have to change people, whats not possible in the long run. The idea that all people could be smart and act smart is a nice fairytale. Humans are emotional and when emotions come in play there will be dumb decisions.

>>32015 Shut it down isn't a reasonable solution. We all love lain and dont want to stop. The correct solution is to build the alternative 'small tech' we want to replace the megacorps big tech with.

>>32006 I would like to point out that most people actually are stupid, however it has almost nothing to do with how educated you are. Most people really don't know how to interpret a statistic.

>>31924 Exactly, and it's way easier to pull things off like eclipse attacks when relays are just automatically put into directory because you own some token. At least Tor has a public relay directory that doesn't rapidly change and a large number of relays run by non-profit motivated actors. Even without considering deliberate honeypots, if Lokinet takes off, everybody is just going to cash in by running 100 "relays" in the same data center with fake latency timings and such, making trivial to spy on traffic. Not saying it's completely worthless, but it has some issues lurking.

>>31576 Some sort of animation language ala flash (with only ability to play animation) and a standard for describing control elements for such language and also any other video or audio file without involvement of JS would be nice.

>>32006 If you examine how VIPs handle secure communications, they do not understand how it works, they just trust the person who provides it to them. We can do the same with our own networks, by providing everyone you want to talk to with a dedicated secure device for contacting you only. Alice and Batman want to communicate securely, so Batman gives Alice a secure communicator that can only contact Batman, who is responsible for all the security. Implementing a secure communicator can be done using an Arduino LoRa pager with OTP. Basically a pair of devices that can send text messages to each other only. An Arduino with a small text display and an SD card for preshared OTP decodes each text message received over a LoRa network like Reticulum. With a built-in TEMPEST-hardened keyboard, it can also send messages to the other unit only. Addresses and keys are all preshared, and cannot be changed on the device. This is to idiot-proof the device and prevent sending messages to the wrong address or receiving messages from a fake address. The cryptography is unbreakable, and user error is prevented by limiting the user to safe actions only. To improve security further against hidden cameras, the display should be a peephole or VR goggles. Hardware backdoors can be countered with discrete components or libre chips. These secure communicators will give you perfectly secure communication with those you want to talk to.

>>31693 This

>>32238 >will give you perfectly secure communication not how that works bud.

>>32238 What if one party loses the device, how do they tell the other that future messages should be untrusted? What if an adversary peeks at messages while Alice/Bob are asleep or don't have the device with them? What if someone loads malware onto the flash storage which leaks messages in the clear? What if someone dumps the OTP from the device and is able to decrypt all futures messages (because it's wireless). How do you rekey? >over a LoRa network A wireless connection only Alice and Batman use will make them both very trackable via directional antennas. That's probably not what you want.

>>32254 >A wireless connection only Alice and Batman use will make them both very trackable via directional antennas. That's probably not what you want. Want to find the batcave? Steal Alice's device, send a message to Batman, use a directional antenna to track where the response comes from. This is why you need a threat model when designing secure systems.

>>32254 >What if one party loses the device, how do they tell the other that future messages should be untrusted? >What if someone dumps the OTP from the device and is able to decrypt all futures messages (because it's wireless). How do you rekey? Use a local password to secure the device. Local password should encrypt OTP. Use a backup pair of devices for a secondary channel. >What if an adversary peeks at messages while Alice/Bob are asleep or don't have the device with them? Keep the OTP symmetric encrypted. Never store plaintext. Delete OTP for already read messages for Perfect Forward Secrecy. >What if someone loads malware onto the flash storage which leaks messages in the clear? Every computer is vulnerable to some form of physical attack, an adversary physically tampering with the device, not just imaging the SD card, is a security breach. If you do not have the device with you at all times, it should be considered permanently unsafe. Enforce this with needing to enter a local password every 12 hours or device self-destructs, which also sends final automated message to Batman. >A wireless connection only Alice and Batman use will make them both very trackable via directional antennas. That's probably not what you want. Security does not imply nor require untrackability. But if you also want a secure and untrackable communicator, you would need to use something other than omnidirectional radio on a single frequency. At the minimum, you would need frequency hopping, with burst transmission too. A LoRa pager only emits radio during transmission, which means location of the device cannot be tracked.

>>32006 >We're all debased and dehumanized by this current infoculture. That's exactly what people deserve. Even I contributed to today's situation, I just woke up a bit earlier but still too late.

>>32302 >Use a local password to secure the device Now I just need to watch Alice type in the password or beat her up until she tells me. >Security does not imply nor require untrackability. Security does not imply anything. Security only makes sense in the context of a threat model. What you're doing is trying to make a "secure" system by countering all the attacks you can think of. Classic beginner mistake.

>>32256 >Want to find the batcave? Steal Alice's device, send a message to Batman, use a directional antenna to track where the response comes from. Neither Batman nor Alice need to transmit from their own location. They can use unlimited repeaters and disposable transmitters. >>32321 >Now I just need to watch Alice type in the password or beat her up until she tells me. Alice should cover the device and herself with a towel before entering the password. For plausible deniability, Alice should have 2 passwords, the real password and the compromise password. 2 versions of OTP are stored, one in a regular volume and one in hidden volume. The real password unlocks both, but the compromise password only unlocks the one in the regular volume. If entered, the compromise password provides the same message sending access as the real password, but uses the compromise OTP for all sent messages which automatically alerts Batman that the device is compromised. Alice must be required to send new messages to Batman on a regular interval. Old OTP must be deleted so that adversary can never decrypt old messages with compromise password, only send and receive new messages from Batman. Batman should send messages that require both versions of OTP to decrypt, but still yield an intelligible message if only decrypted with compromise OTP. >What you're doing is trying to make a "secure" system by countering all the attacks you can think of. Unconditional security has a distinct meaning from privacy, anonymity, plausible deniability or freedom. With this secure communicator, it is physically impossible for any third party to ever read the plaintext between Alice and Batman.

>>14586 Alice, you should check out the N-O-D-E zine number 1 article "Mesh Networking Basics" . It gives an introduction to popular meshnetwork protocols and deployments. You might find it useful. >https://n-o-d-e.net

>>28182 Politicians should not be using private, centralized social media. A good start would be move them over to running their own ActivityPub instances. Pleroma for example. Then they do not have to worry about getting censored. It's sad that companies in America, are getting away with literally censoring their president. God we're so fuarrrked. Internet is going to soykaf Lains. Is Lainchan mirroring to IPFS or anything? Not too familiar with that technology. But I hope we have an outlet to go to in the ever possible case this site dies for whatever reason, which I hope it never does. I love it here.

>>26732 >Wireguard Can you elaborate? How exactly is this relevant? It's just a VPN?

>>32451 I don't agree with censoring posts like that, I'm just stating I heavily enjoy the content here. I would much prefer Lainchan then nothing. Currently I have not found any other imageboard like Lainchan, 4chan is craped to soykaf. Google scripts, and just "why don't you use apple" or "why do you use apple", etc. It's become taken over by normal people to be honest. Don't get me wrong, I still go there only because. It feels like the father of imageboards. But I'm trying to limit my time there because I don't find a whole lot there enjoyable besides some of the general boards. The fact that it has proprietary dependencies pisses me off quite a bit and is one of the reasons why I'm not active there much. 9chan/8kun seems fairly dead in comparison to 4chan still. I wouldn't mind adopting 9chan to leave 4chan but right now that doesn't seem extremely likely until a lot more of the /g/ generals end up on 9chan. Also even 9chan is using cloudflare. At least 8kun just uses Letsencrypt. :/ Sushigirlus isn't too bad but is pretty damn dead. Has general threads from 2017 still up with little conversation. Here though it feels like early internet, it's fairly quiet, but not dead enough to be unable to converse and stay entertained. It's fairly computer heavy oriented, kinda comfy feeling idk. Again, not a defense for censoring soykaf, but it is nice here. Can't even say a ton of words because of the word filter. lol. I don't want to see Lainchan go away. At the very least, if it does go down, I hope someone brings it back up and administrates it with a more hands off approach.

>>32452 7chan and 4taba are pretty ok. I personally like the computer club (/cc/) board on 4taba.

>>32454 Didn't even know these existed to be honest. 7chan is running Google Ajax and Analytics.. blehhhh 4taba looks fairly comfy. Doesn't look as /comfy/ as here but still looks pretty coimfy. Saved.

>>32457 32454 here. yeah..... I forgot about 7chan's googleshit.

just use APRS glitterboy