Internet Freedom

I've mentioned a virtual internet SubWeb for quite some time, and have been pimping the idea. I just want to share with people what I am doing personally to bring about it's creation:

The past few days I've worked hard on completing a working sample deployment under Linux, I would like things to work the same under a variety of platforms and there are still a few loose ends and optimization to do, but now I've added a member to the team and so I've got to work on making a utility to replicate the thumb drive software.

Essentially when you plug your thumbdrive in, the software is launched. The software is stored inside a virtual box image and thus is isolated, designed to bypass the hard drive IO stress that tor, torrent and other modern p2p protocols produce. The IO stress issue has been long-standing with software of this nature, and the launcher for vbox images will be used to make deployment of virtual machines more prolific. The use of tiny core in this case as a standardized light weight linux platform for running Apache2 and php and extensions is where the core of the software will ultimately reside.

Essentially completion of Windows, Linux and Mac platform deployments are nearing completion and I am seeking developers (more like script kiddies) to develop methods for Solaris and OSX.

The resultant package will be released to encourage traffic for the ultimate idea here.

Digging deeper into the code I have discovered that a variety of web software bundles can be easily converged using php, and over p2p can be run fairly efficiently. CPU usage can be easily managed and there will be software for managing a variety of deployment types.

The idea is to configure the machine to operate in simple peer mode, backbone node mode, and Mesh Wifi node modes.

The idea of the mesh wifi node is that I don't feel like begging for donations to keep critical server equipment running as for a time period there will be required anchored centralized servers. The software will be carefully designed to make disconnection of the anchor point possible as proliferation of the software hits a point that finding peers is easy to do for initial connection with the network. The wifi product will be an "enhanced" router, basically a micro atx pc with several wireless cards to enact as an inter-meshing open network hub for this software. I will leave the long-distance links up to other people to figure out; for now I merely would like to establish a networking method that can fight censorship and can bypass the difficulties of modifying a multitude of router bios's to accomplish the same feat. These will be a little more expensive than typical routers at a couple hundred bucks each but will function as dedicated nodes and will be cheap enough to encourage manufacturers to embrace the standards that will arise from a successful piece of software.

Backbone nodes will serve to reduce the complexity of the model involving the un-anchoring of centralized services. I am seeking assistance, two of us at this job is still going to take a while to get these results, and with no finances it will be difficult to afford hosting for the anchor points. This is where source forge comes in and as to why I am fixating on establishing a presence there with the launcher and traffic encouraged by it. Oracle could have made it easier to handle vm's but they left all of that scripting and management to be done manually through a poorly designed api. If Virtual Box is not present it was nice to find a .run file which automates the installation across most linux platforms.

Anyhow, onboard this PHP server is an open source torrent platform already. There are IM protocol plug-ins though I think they would work better for backup connections and IM migration into a new client that is more secure via the local PHP server. Mesh Wifi protocols will be migrated into a PHP plug-in, as well as an implementation similar to onionskin routing but made more secure through lazy math branching of encryption (to make decryption take a little work for a piece if you have the right key, but a hell of a lot of work if you are trying to guess it).

We need developers for this project -- I would like search engine integration, a service similar to facebook though more like a personal website, a software management hub where the node administrator effectively can manage the converged php software.

I have investigated software such as Onion skin, which works acceptably well (though still a little slow). Freenet is borderline unusable. I feel so sorry for people in China who have no other choice but to use that piece of crap; the culmination of almost a decade of distraction from something that actually works and does the job. I believe it too, to be a compartmentalized distraction setup by the system just like Wikileaks is in my books. I don't trust anything centralized and closed source. There are some aspects of distributed cloud computing from seti@home which could be adapted for this project.

If you can't contribute skills but would like to donate be my guest, though I am not really expecting anything until I get the graphics and descriptions up and running on my source forge page, for now I've spent maybe an hour or two setting things up initially. If anybody would like to volunteer any artwork that would make this go by a lot faster; my laptop was repaired but I still have to re-install everything and I don't really use my netbook for photoshop as running windows in a vm to run photoshop with (because wine sucks) would be dirt slow.

The benefit of thumbdrive installations means that nodes can move around and change node hosts or cache included on already running hosts.

I was thinking of posting the source forge page but that will wait I'd hate to post it till I've made it look sharp, however, that doesn't preclude the reason for my posting of this thread -- it's intention is to encourage some dialog and feedback into my project as it develops and of course if I started this thread with something that is already totally made it would be pointless, considering I would like people to see how easy it is to get something even this complicated rolling. The reason I chose PHP should be obvious after reading this, much of what is needed is already out there, and can be tweaked or used to model something new and deploy it quickly.

The pace of this project is going to be accelerating quickly so I'm stuck at a point where I am not entirely ready to toss things out there; but have no choice because PHP does not take very long to develop with the sheer volume of people out there who can help code this (even C has a fair amount of support).

I will note success in my attempts to get freenet to serve PHP documents which I had curl running on Apache2 retrieving and running; though it ran flaky because the core of the software; freenet itself, is crap. The server worked fine; though freenet would frequently crash because it was not designed to work with a proxy webserver. I got curl to work reliably on all sorts of things, though freenet times everything out and bogs everything down and hence why I believe it develops unstable behavior and eventually shuts itself down. It was a huge disappointment to me to observe just how useless freenet really is.

Now in terms of outreach I also believe it is possible to run linux on smart phones, and therefore it is possible to create easy integration of my software with smart phones. This will be the last stretch once there is a working backbone -- my intention is to give people the tools to setup metropolitan-wide Open Mesh Wifi networks, this way providing all with better connectivity and cheaper services (as all ISP's do is charge a multiple times the rate for fractional services). It is my belief that we must push for public funding of infrastructure for a free internet, but at the same time we must keep that kill switch away from government hands. Right now they could order ISP's to disconnect everyone and there'd be very little anyone could do about it, period. We needed a defense to the system's censorship tyranny yesterday, do you people really want one or are you more interested in a song and dance look-alike system with embedded traps the system has engineered inside to keep them in control.

This is a project that could take months to complete; not the HUMILIATING years that freenet has spent getting basically nowhere (I mean they do make improvements but their software is still fairly useless).

I want ATS to take part in things. I don't need you specifically to contribute programming abilities or donations, I specifically need open dialog and just general discussion about the problem and how conceptually to flush out a system to achieve the goals of this project. With all of the wikileaks topics and time spent speculating about what is happening (and what will happen next) I figure it would be a good time to mention -- you all have energy that could be directed towards a productive solution to the issues at hand.

If you feel you have no power to fight back consider what helping this project will do. Even if the help is small, or a suggestion, or something cool you saw that might work into things somehow. I'm great at problem solving and coming up with ideas but from what I've observed so are many people in the ATS community -- let's take some of that mental energy and do something to make us conspiracy guys look better. We need an example of good; so far the media only has bad things to say of us, here at ATS (about our connections to the ariz. shooting for example, to cite the most recent example I've noticed). The key to the truth is to make sure that it's journey from source to you is freely unimpeded and does not distort the original content. If you think the truth is a piece of paper with black lines all over it to "protect" people then you are living in a lie.

Just as an interesting side-note, I noticed there is a check box for software in this category to be alerted to the US government as it employs encryption. Well they can get contacted all they like, I'm in Canada, I couldn't give a crap if they one day come along and thinks my software is technology that shouldn't get distributed around the world; they have no jurisdiction here and I will be setting up mirrors to ensure that if I get taken out someone else will be around to pick up the pieces and carry things forward. This will get built regardless -- nothing would give me more pleasure than to get us to a point where we can disconnect the ISPs, disconnect the cellular providers and telco's and cable tv providers. We don't need them anymore, they've designed technology specifically to empower them and their fascist corporate models. They come up with a technology and then inject dependency on them into it so you'll have no choice but to play by their rules; this has to stop, though the window to make it stop is short (determined by how much farther ahead technology-wise the system gets relative to the general populace -- currently we have 1000's of times more computing power and resources than all of the government's and businesses of the world combined, we must put this to use before our strength in numbers is diminished by the system). Are you going to wait until there are vulcan cannon sniping drones in outspace capable of headshooting someone with a smart phone, you seriously going to wait till there's an army of these things before you get off your ass to do anything to fight back?

Personally I hate violence, this is the only thing I can think of doing till the SHTF.

I am looking forward to your comments.

Ryan,
SubWeb Project Co-ordinator

Ummmmm...I'm in.

Do you pay at least minimum wage?

wow! good luck.
I found a linux free soft ware site.true free.
you need to make shoure no one can claim it later.
or parts of it. or will you get paid for the soft ware?

um! can you say all of that in a very simple form.
or lots will just skip it. like they do.

hmmm I don't really intend on making any money; unless if it's to get paid doing work assembling these box's and selling them but honestly I've figured if I can put a team together and everybody can pull their weight they can take anything extra that arises from this aspect of the project. I honestly needed something to give people what is due if all this hard work pays off for humanity. I don't actually want people to pay anything for the software -- that is entirely open sourced, but with all of the traffic that will be directed towards the anchor-server and the main website for the project once it goes live, there'll be a mechanism to generate revenue through massively parallel computing and sales of routing equipment. I am not interested in the least in advertising and I definitely would like to make sure there are mechanisms to get rid of them. I am tired of corporations invading my house through my mailbox, television, computer and phone. I am planning on having kids soon, and I am proud to say that they will never know who ronald mcdonald is if I have anything to say about it. My children will not be exposed to the corporate world until they are adults and have learned to handle the social manipulation and engineering produced by it.

I plan on publicly accounting for all funds, I have 0 interest in making money I've learned how to live well off of little. I can buy months worth of food for a couple hundred bucks and I'm happy with it; my rent is mostly getting paid but the unemployment hurts as I am not useless; I have talents and I'd rather not sit around and have them go to waste and unused. I have an opportunity to learn more about computers and establish a portfolio again, that is enough for me. I lost my previous portfolio over the course of moving several times and through a career change into tower crane rigging; but there's no going back to rigging there aren't enough tower cranes out where I moved to and the employers don't give jobs to people who haven't lived here long enough in their books. It's tough but what can you do; I'm just trying to make the best of an all around hard situation. I think that's what we all need to do -- reconnect and relearn what it is to be part of a local community again. Modern society has done much to damage the sense of local community; it barely exists anymore unless it is pre-screened by the corporate media (so it is their version of community, global, not local).

Thanks for the quick reply, it's nice that you asked about getting money because nobody likes to work for free. Personally I think taking part in something so ground breaking ought to be enough but then again if that ground does get broken then the people doing it definitely deserve something; I've considered it and decided that there is nothing wrong with some sales on the side -- t-shirts and what not, doesn't ats do that? Alex jones does it. In fact it is a fair and viable way to make an otherwise profitless venture a little more rewarding to those who spend so much time with no return making it happen.

Originally posted by buddha
wow! good luck.
I found a linux free soft ware site.true free.
you need to make shoure no one can claim it later.
or parts of it. or will you get paid for the soft ware?

um! can you say all of that in a very simple form.
or lots will just skip it. like they do.

aha well see my previous post; obviously it's difficult to pay for an anchor server; no worries about that yet however.

To answer things more directly however the entire software is open sourced. This means that if we integrate anything into it, that component too needs to be open sourced. I don't really need to make sure about claims, copyrights, patents etc. I am willing to put my ass on the line for this -- I am the one on the hook and I have no problems with that. They can sue me in court but they won't stop development, nor will they stop torrents of this software from getting out there, nor can they shut down every last node -- even if there were patented software inside it would be unstoppable once released. Hence why it is essential to un-anchor things as quickly as possible, because the vulnerable spot will be someone attempting to take over the anchor server.

Very little of linux I have observed is closed source. About the only thing I've noticed are nvidia's drivers.

Virtual Box - open sourced and developed using python made by oracle, www.virtualbox.org...
Tiny Core Linux - open sourced and not sure how it was developed, probably using all sorts of languages for differing aspects of the package, tinycorelinux.com... (scroll down to the bottom of the page, under about our project, a good paragraph to read). It is free and licensed under GNU I believe; either way no issues there.
information on tiny core - www.h-online.com...

Apache2 - open source, www.apache.org...
PHP5 - open source, www.php.net...
Torrent extension for apache2 - open source, www.torrentflux.com...
onion routing - open source, www.torproject.org...

That is only scraping the surface. There are many extensions I haven't included in this list that come by default with PHP, as well as many which do not (literally thousands of them). All the tools needed are mostly built, we just need convergence of these open sourced projects in a transparent process to make use easy for people of all sorts.

reply to post by stealthc


I've never heard of this, can you cliff it down a bit for me?

What is the actual aim? Subweb?

Are you attempting to implement an 'on the fly' system of communicable 'servers' so as to use the current internet but with an undefined, yet coherently routable, method?

I ask about it being on the fly as you mention thumbdrives. but I am unsure as to what you mean by OS HDD stress.

Confused, but I've had ideas similar to the concept in the past without any way to implement it alone...

reply to post by badw0lf


Long story short -- the notion of a "server" should be rendered a dinosaur model; we are all peers on the internet and we should be migrating away from "server" centralized models. This is the trojan horse implanted by the system to ensure that they keep a lid on control of the internet. They were considering this right back to the original days of DARPA, IP and the original OSI model. I spent a long time following w3c and their ipv6 development and clearly to combat the new features introduced into this we need something now. It is only a matter of time before they (TPTB) demand IPv4 be made incompatible.

Indeed it is confusing it's taken months just to wrap my head around this. I've been using linux for years but never really got into scripting and development under it so I've got a huge learning curve; but have discovered PHP is as easy to pick up as ASP was thankfully

Now to explain things a little better for you in relation to your questions:
Of course you haven't heard of it; nobody's really proposed the idea of just outright virtualizing the whole net -- there are limited implementations of it but nothing quite so tightly nailed down. It's an idea I've mentioned around the past few months but that is *the* only place it would have come from so I am not surprised that you haven't heard of it. I was surprised when I ran into freenet I thought it was going to drastically speed things up but I was wrong it lead me down the wrong track for a while.

To explain things further, if you are running a node and have a cache it becomes a nice feature to just bring the cache around with you. Keys can be securely stored on the thumbdrive to make security seem transparent to the user of the software. The epic IO stress that is produced by torrent and by freenet is due to the fact that you are constantly reading and writing information to the medium and there is never a break from it; it is cheaper to burn out a thumbdrive and lose the software than it is to burn out your own hard drive and lose everything and have a broken computer. When blocks of data are constantly being written, erased, read, re-encoded, etc, the point of failure is the hard drive mostly due to it's mechanical moving parts constantly moving. Thumbdrives are very resilient. The higher-end backbone node and Mesh Wifi node configurations will come with a small SSD to handle storage, hence part of the slightly higher costs associated with it. Let's say you are in the middle of downloading something but you are going to your friend's house -- but would like to bring it with you and finish downloading. Merely pop the thumbdrive out, bring it with you, and pop it into your buddy's computer and it will configure it to work as a node automatically, no stupid settings to play with (if you don't want to).

The on-the-fly notion you presented somewhat matches the concept. Nodes are not meant to be stationary. Their cache is meant to roam about as 32kb pieces; the node itself is not meant to be necessarily pinned down to a single geographic location or computer configuration. When a node moves it automatically reconfigures itself to mesh with the network using the resources available (depends on configuration type).

The thumbdrive will also be bootable, so that it will work even on computers with no operating system.

Torrent's will relay decryption data for global file, but individual pieces will be handles and onion skinned on a per request basis. The anchor server will manage piece migration within the cloud, the peers will negotiate decryption mechanisms with each other; both of which if employed using onion skin should protect privacy.

The problem with torrent is that it does not always encrypt connections, web seeds are not common, pieces are usually not well dispersed, and there are way too many leechers. At least if we all dedicated a gig or two of thumbdrive space (or more) it is easy to see that you'd be giving nothing up and a vast virtual drive will result.

1 million users = 5,500,000 gb roughly (if they all used 8gb thumbdrives). 5.5 petabytes from such a small number of people is no laughing matter. if everybody in america joined in, there would be 1650 petabytes. If everybody shared 20% of their cpu power (during idle time) of one core of their multicore cpu arrangement, then we are getting (let's say idle is 50% of the time) 100 Tflops. That is from a million people, again times that by 300 million and you'll arrive at a figure which is truely impressive, 30 Pflops (recently breaking 1 or 2 peta flops was considered a feat). Considering many people now own multiple computers, every smart phone, every netbook, every laptop, every desktop, and even some things which may surprise you (smart appliances even) can contribute. This is entirely what I am talking about; convergence can create something truly remarkable. Now who out there is game to see this happen?

I'll be on here a lot tonight, these questions are getting interesting. Thanks.

Just as an update, there are two developers signed on now.

One is an expert in cloud computing.

I am very interested in the idea, still a little conf.. Heck a lot confused

My concept of a 'server' in this case is the individual - not a centralised service, but the users point of communication, similar to how using small free webservers like Abyss and a free dynamic name resolution service, such as dyndns would be.

But dyndns relies on already established services which use name resolution servers to translate the dynamic IP to a static name for coherency.

So as every machine is unique on the net with a unique IP determined by the hosting service they connect with, to break free of that there would need to be a method of identifying each user/machine somehow, or at the very least, a broadcast of your unique ID to the other members in this system in order for communication - this is where I think I'm most befuddled.

Most methods would be fraught with issues as there are so many levels of blocking of non standard traffic going on from one isp to the next.

I'll keep an eye on this to get more info, I think I'm still standing on the outside peeking over the fence and missing a lot of what the intent is!!!

Good thread!

Originally posted by badw0lf
I am very interested in the idea, still a little conf.. Heck a lot confused

My concept of a 'server' in this case is the individual - not a centralised service, but the users point of communication, similar to how using small free webservers like Abyss and a free dynamic name resolution service, such as dyndns would be.

I'm not surprised. Ultimately dyndns requires a central service because that is how the web is meant to work, to organise things from tld down; at least that is how the naming service works. Yet another internet killswitch in the cname entries for ip addresses.

Ultimately though if you have nearby nodes why would you need anything to tell you where they are? I can find a torrent client in a few seconds with a port scanner; so if it's that easy because the software is so widespread that everyone is using it; it then becomes quite simple decentralizing things -- you simply unplug the reference to a central server and send people to find their own way into the network (because the software on their end can transparently do this). We can use IM to assist this process and make establishing connections to the network even faster via the buddies list. All I will be doing is offering an im service that is tunneled over the network and universally used; nothing more I won't have anything to do with automated php software that defines the structure of the network. Ultimately the central server will have to operate like discrete symmetric blocks of scripts executing across many machines in order to verify integrity and consistency across the network via. random selection of geographically distinct and dissimilar nodes to perform such operation in parallel; this way it is easy to run the central server across many nodes. If the code is small and compact; the group(s) of nodes randomly chosen to do the job will work for a few seconds, then the serving portion will mirgrate off to a new collection of nodes -- thus everybody participates and that which is centralized is dispersed within the p2p cloud.

But dyndns relies on already established services which use name resolution servers to translate the dynamic IP to a static name for coherency.

Do you now see how concentric the design of IP and related protocols are to dependency on a centralized model? And yet to establish a connection across IP this is totally unnecessary.

So as every machine is unique on the net with a unique IP determined by the hosting service they connect with, to break free of that there would need to be a method of identifying each user/machine somehow, or at the very least, a broadcast of your unique ID to the other members in this system in order for communication - this is where I think I'm most befuddled.

have a look at onion routing. Pieces of data are packaged in a shell of encryption, as the data passes from one random node to the next pre-chosen by the client using global node keys. The speed comes from doing this many times over in parallel for data.
Also note -- everybody doesn't have a unique ID, there are many overlapping ID's obscured by NAT.

Most methods would be fraught with issues as there are so many levels of blocking of non standard traffic going on from one isp to the next.

I'll keep an eye on this to get more info, I think I'm still standing on the outside peeking over the fence and missing a lot of what the intent is!!!

Good thread!


onion skin routing can get through this; I don't see why we can't use standard https connections to achieve this end; throw in port stealthing and give the software a list of commonly unblocked ports to try. That is about all one can do. If the ISP blocks this traffic, buy or setup a Mesh Wifi node and start coaxing your neighbors to do the same, bypass your ISP as a community because of their sucky service. This concept is designed to take ISPs and ultimately the last remaining corporate control out of the net too. As far as your isp is concerned it won't be-able to make heads or tails from streams of encrypted traffic (the only thing it can do is read the source, and the port and perhaps know it is encrypted data; that is about all).