Shareable

Have you looked at Freenet? It is the decentralised highly anonymous network that is the most censorship resilient form of communication available today. Well worth looking at.

Nice article. Good point about 'net neutrality automatically being lost the moment it's brought into question.

One idea might be for a government to write it into its constitution that a network connection must be provided to every household, much like running water, and that a government body must maintain any centralisation of such a network, and paid for by the tax payer.

You may think that it would be dangerous for any government to control its country's internet, but considering that they own and control the water supply, surely a consensus of trust could be made. And again, legally, its owned by the people.

Just one idea.

Well you might have to start with community projects to provide subsidised access to the poles. Otherwise you're going to be leasing your lines from the very corporations you seek to avoid contact with. And at least in Canada, the CRTC has already ruled that they're allowed to regulate traffic on their network that they reserve for whole-sale access. It pretty much killed the last of the competition in the telecommunications industry here.

You're going to have to lay cable if you want to cross the distances to reach rural communities and send packets across continents. It cost the telco's tonnes of money over many decades to get where they are now. The employed hundreds of thousands of people over the years to do it too.

I used to dial-up Fidonet BBS's too dude. I'd much rather have a 25Mps connection and a VPN that spans across the world. The cost of any such large project is that it's the work of many people and that means regulation governing its use so that we can all try and get some benefit from the effort in building it.

If it's just pervasive advertising that's ruining your culture and free exchange of ideas, there's ad-blockers and dark-nets of all sorts out there where you can exchange your radical literature free from the bonds of "the man."

The problem isn't with the Internet, it's the people.

Surely the most obvious constriction which has to be overcome is the wire? To be really free the people's network needs to be totally ubiquitous, with as few landing spots as possible?

Mark's Connective project looks like a great initiative, but perhaps we need more out of the box thinking to make it really work?

Fairly simple to do, a series of point to point wireless radios and head units with Alpha, Beta, and Gamma faces to distribute through neighborhoods. The big issue is to decentralize electrical distribution, lastly you can transmit data through electrical wires too.

You pretty much summed it up. All we can really do without building our own infrastructure is subvert the existing one... which has shown itself to be a difficult thing to do.

Private networks and darknets are one way to go. There are lots of them. WASTE, Freenet, etc. Basically impromptu distributed and encrypted p2p networks. They have the problem of isolating content to your immediate group of friends with whom you have shared and signed one anothers keys. In other words: complicated to set up.

Another way to go would be to build a new DNS service. Perhaps one that can resolve names across islands of encrypted p2p networks. The only barrier to entry with that is maintaining consistency. Caching would be hard. And again, encryption scares away end users.

But then again, all of this can be subverted by the people maintaining the network infrastructure... so if they don't like what you're building, they'll just deny access to their networks.

The only real way is to literally build your own. Form policy to subsidize access to the poles and blocks of wireless spectrum for community projects. Invest in the commercialization of space and launch satellites. Etc.

I think it's probably cheaper and easier to try and subvert the policies that are eroding the ownership of the public investment in the infrastructure in the first place. It's going to be a while before it's viable for a grass roots organization to launch a network of satellites to provide cheap wireless access. Not saying it won't happen... just that it's not here yet.

P2P is a very active research topic and maybe someday we'll create a distributed censorship-resistant naming system. In the meantime we can support alternative DNS roots and TLDs, independent TLS CAs, etc.

On the other hand, I don't find very interesting the idea of wireless networks parallel to the Internet, like netsukuku. Firstly because I don't like wireless networks so much, their impact on health are still unsure. Secondly because you can hide on the Internet for as long as cryptography is not strictly restricted, while wireless access points can be located and taken down by police, for whatever phony reason they can think of.

In the end, the real problem is censorship, not the fact that the Internet can't stop it, because a censorship-free network would be destroyed by governments anyway. So what can we do against censorship ? We have to try and take control of governments, by proposing a better societal model.

I've been thinking about this for years, including recently. I sat down a few nights ago and started jotting down notes for an ad-hoc mesh network protocol for such a network. And FidoNet was an example for me, too. This new network would be store-and-forward, Unicode text only, much like FidoNet, with e-mail and forums. I'm of the opinion that this new network should use multiple possible transmission channels—modems, ad-hoc Wifi, optical links, and even things like USB sticks and CDs that could be mailed (write the outgoing mail to a file, put on USB stick, snail mail to receiving node, they grab the file, process, and the mail is delivered). There's a lot to think about, but yes...there are those of us who are thinking about it.

And not all FidoNet sysops were kids (although I was :) ). And yes, FidoNet is still in operation.

Invest in Quantum Teleportation research, this is a real area of science that could see us within the next few years capable of transmitting point-to-point data instantly over any distance. Thats when you create a truly distributed network..

A simple layered solution over an IPV6 network would make sense, you'd simply sign up to provider who would let you on and you'd go from there. Whatever was in between would be inconsequential. We can have our internet over top of their internet, they cannot stop us lest they stop the internet.

signed an original anonymous contributor, not a 4channer.

i used to belong to a group in amsterdam who worked on a project creating a kinda of local shared internet, we built our own antennas, hooked them to discarded old pentium II or IIIs (running debian linux) and shoved the contraptions into the attics of all the local squatted buildings, we used it mainly to distribute a handful of dsl/cable connections outward, but as well to trade files, movies, music whatever on our own network which we controlled, this was done using open mesh software creating a kind of giant ad hoc network, which often messed up, but it worked more often than not. i am not so up to date on the development of the software behind this, but i know there of meshlinux and the b.a.t.m.a.n. project, which are working still on improving the mesh network idea, see http://www.open-mesh.org/ its certainly a way of creating your own network, and cheaply.

There is something horribly wrong with this proposal.

You are abandoning the "real". The issues threatening the web are already in full force in the "real" world doing much more serious damage than they ever will on the net.

The inventiveness, strategy and will to build must be taken to the streets and against those who oppress you IRL. Inventing a parallell universe where you are free but utterly trapped by the confines of the medium is ridiculous.

I could not imagine a greater copout! Read the text again and consider the horror of the defeatist argument.

In the above suggestions, they are many interesting ideas. I'd like to propose the following: ubiquity through simplicity.
Create an open project for making the hardware (say Arduino + communication shield + solar power shield), code the software so an easy download would be sufficient to assemble the device. Design the whole thing so that this device is real cheap. Keep the device real cheap, make it so that the device-to-device is fully self-booting and distributed.
Now let the people put these wherever they want. Got some fund money? Order a large quantity of these, and distribute them widely. This ad-hoc system is hard to disrupt, easy to maintain. Now to figure out how to handle spammers, etc. and the system can be kept clean.
The point of the above brain-fart is that a cheap self-maintaining self-sufficient
system can be made, and can provide valuable service. But it may not allow one to compete with a full fiber-to-the-door service; nor would it be able to provide a robust security/self-cleaning service.

You're right, you can't change human nature. There's always going to be a thick layer of sludge over whatever utopian ideal we come up with. The way to solve (or at least mitigate) that problem, however, has already been demonstrated in other areas of society.

For example, look at music piracy. Back when the best tech available was large vinyl discs, the music industry had almost perfect, greedy control. When cassette tapes came out, copying became popular, even though its spread was limited by the problem of imperfect copies. When CD burners became available at low cost to the public, file sharing got big - suddenly there were inifinite copies possible with no degradation! The music was still somewhat tied to the hardware, though - you either had to buy CDs to burn, or you had to sit at your computer to listen.

When digital music players became widely and cheaply available, file sharing *exploded*. Perfect copies, unlimited playability, and world-wide availability became the perfect storm. Sure, there were people who uploaded fakes and viruses. Sure, people scammed others and sold bootleg copies on the street, but it flourished anyway because *the bad apples couldn't stop the good ones*.

In any solution we come up with, there will be the dregs of society doing their damnedest to destroy, subvert, or exploit the system, but as long as everyone has equal and irrevocable access, there will always be free transfer of data and ideas between the people who will use it properly.

The solution is cheap, ubiquitous, un-throttled access. I think mesh networks have potential. If you screw around too much, your neighbors will simply stop forwarding your traffic, but they won't be able to know *what* you're sending, so they can't discriminate except by bandwidth usage. If your neighbors block too many people, they won't get their own traffic forwarded. The problem will solve itself - at least, it will once the mesh reaches critical mass, which is always the first thing to worry about.

I have been thinking about this as well as FidoNet. People did have to pay here and there to keep it maintained but, it worked great and was run by mostly home users.

When will the revolution start? Count me in.

Great points, Douglas

I was a Fidonet Point back in the day (last half of the 80's). Talkin' slow modems over POTS. It is very doable, albeit asynchronous. But what's the hurry?

As a ham operator, I can tell you that running any kind of network like this on amateur frequencies would be blatantly illegal. You can't use encryption of any sort. So unless you want the entire world reading your email, that's no good. Plus you can't use them for commerce, so it would be personal and public sites only. Not a terrible idea, but you're going to have a hard time keeping a network like this up without commerce. I'd imagine most public frequencies are the same way. You could _maybe_ use CB, though that would be damn slow and the amount of traffic would choke it up pretty fast.

So, you can't use public frequencies. You also can't use anything wired - I mean, good luck getting even a single neighborhood wired without any corporate control, let alone the nation. That costs real money. Huge money. And you have to convince a lot of people who will probably not want to even be part of your network. So that's out.

So what you're left with it buying spectrum. But that's extremely expensive as well. Plus whoever buys the spectrum has control. So that's pretty much out.

So you're left with using existing (corporate owned) wires and just rewriting the software. Which is all the Internet really is anyway. Forget about replacing the infrastructure. Anything physical will always be able to be controlled. Even with a wireless system, the government or corporations can deploy signal jammers and packet sniffers and such. But that doesn't matter. What we need is just more people interested. Get more users on Freenet, FCON, or I2P. Encrypted, anonymous, distributed networks built right on top of the regular Internet.They can't filter traffic they can't read. The problem is that, at the moment, these networks are small enough that they could just block all such traffic and get away with it. But if we got a sizable enough user mass, that wouldn't be so easy. Admittedly there are problems - Freenet and FCON, for example, rely entirely on links. There's no real domain name system. But these can be overcome, and it would be a hell of a lot easier than overcoming the problems of building new infrastructure.

Yes, lets fork! I'm down!

I'm an independent consultant (IT Architect) currently working for a client that is investing billions to create a private network covering most of the US. Spectrum has been bought, custom radios are built, and the custom messaging platform (which abstracts messaging from the transport layer) is on the way. It can use radio, cell, satellite, WiFi or wireline IP transport (based on availability and cost).

KB3RTKs comments are right on... To truly fork the Internet you need spectrum. Otherwise you'll have to piggyback on existing infrastructure. If the latter, building a logical network based on a private PKI/Certificate Authority might be appealing.

Part of the solution would be to build 501(c)(12) cooperatives encompassing a group of mutually interested individuals who build their own last mile infrastructure and negotiate wholesale deals with existing telecoms. They could start with existing wireline technologies such as VDSL, HPNA, HCNA, MOCA, or others, and eventually perhaps fiber optics. Eventually, a cooperative formed of these cooperatives could develop a national infrastructure to replace that of the existing telecoms.

First I have to say that the future looks a lot less dark knowing there are this many people with similar thoughts.

I have thought about this for several years. Looked hard at the ham radio idea since my father has been a ham for 60 years. I agree with those that have already posted - ham is not it.

The main idea I want to share is one that is half way between the two current ideas. i.e. Rebuild a whole new internet and build an internet over the top of the current one. My theory is that we don't really need to rebuild the whole thing. The corps have control now because they have almost complete control of the last mile. If we were to build local only intrAnets that didn't use corp owned methods to distribute the data that would be enough to kill most of the corps power. We could use the corp lines to share the data between local nets. Now the corp lines are dumb pipes again! You could also inject internet content into the PRIVATE local nets. This way you can start small with one small community at a time. There are generally enough geeks in each community to get things started especially with some help from some "super-geeks". If the communities are kept small enough could we not just use dd-wrt access points/routers to build the local net???

A peer to peer system will get rid of central control. The OLPC already uses this method.

We do need a censorship regime to eliminate the spam and other evil content, we just need to make sure that it's not run by thugs with guns. We need an organization that is capable of making these decisions that is respected by the community and not subject to takeover by corporate interests or governments. It needs to be capable of standing up to these forces even under extreme pressure to reveal the identity of a user.

I've proposed creating a Trust for this purpose (see my homepage). You would need to choose the initial set of trustees from people with a strong record of supporting free expression and human rights.

-Pete

Here is what we need:
P2P DNS (from what I heard already in process of being developed) decentralizes dns.

We need a way to bypass routers on the internet and setup our own routing scheme where peers route data back and forth through a P2P like routing system as well. Somehow, a tracker needs to be setup and certain hosts assigned to act as routers. The computers with the fastest bandwidth and fastest cpus (whih could be determined by client software and sent to the tracker distributed via P2P) would act as routers and somehow the app would need to load balance the traffic so no computers on the network got overloaded with routing.

We either need a whole new network separate from the Internet or somehow a new network with possibly a new protocol within the current Internet. The problem with using a protocol within the existing Internet is that it could be filtered. But we definitely need some way of either bypassing the routers used by the big telcos/isps or else we need a new network entirely. The more distributed and P2P like it is the harder it will be to shut it down.

It would be best if the P2P dns system somehow forced you to create a url and then checked to see if it was registered in P2P dns already. P2P dns could work similar to current dns only would need to be P2P enabled so there is no central authority controlling it. Also with P2P dns, there would be no need for limitations on a URL. You could have a url like:
blahagagsgsgsgsgsgs
no need for .com .net .ca or anything like that though there is no reason those domains couldnt exist either

An alternative net is something I've been thinking about for a while as well.
The number one issue, is how to implement a physical layer that isn't dependent on any centralized, corporate or government influenced entities. Secondly, it would be ideal if the presence of such physical infrastructure could be effectively hidden from 'official' efforts to destroy, suppress or spy on it.

Obviously, the original DARPA concept of a distributed web of connections, with no central points of vulnerability, is a good idea. Too bad the present Internet doesn't implement this - since it depends on the major telco carriers and ISPs for everything from the backbone right down to end-user connections. Also, there are central administrative vulnerabilities, such as the DNS and IP address allocation systems.

One way to implement a fully distributed lowest physical layer, would be to market high bandwidth point-to-point IR laser coms links, able to be set up surreptitiously between private dwellings. Note that this is actually illegal under present telecoms laws - but screw that. Together with a more distributed, peer-to-peer routing system, city and country wide webs of such links could actually bypass and eventually replace the present Internet.

Other key requirements would be for a node identification scheme that did NOT require any 'official' allotment from a name-space. So it couldn't use unique numeric addressing like the current MAC and IP addressing. Something that might work could be based on geographic (lat.long) location plus personal identifiers - which are individually chosen. This would allow long distance message routing based on destination area, with final local look-up by personal node identifiers.
Node identifiers should be human-readable text and of extensible length. No more 'running out of address space'.
Full anonymity and free use of aliasing must be allowed.
Strong public key encryption of all message traffic should be the default.
All link endpoints should generate random (encrypted) fill traffic when idle, with no way for eavesdroppers to differentiate between fill and real traffic.
The system should support real-time voice and data links. If something like this is going to be implemented, why not put the (government lap-dog) telcos out of business while we're at it?
It should NOT depend on RF links for anything system-critical. Too easily jammed, not to mention revealing of activity and location. And increasing already bad RF pollution, with (arguably) negative health effects.

Here's a name for it: Annet - the anarchist net. Because although a degree of social order is desirable, when it comes to the right of free communication, nothing less than total anarchy is acceptable.

In every age, the dominant social hierarchy is determined by the weapons and information technologies available to the people. Today we (the people) are getting the pointy end of the stick, as governments tighten their control over both weapons and information. Time to flip it around and start poking Elite eyes.

TerraHertz

This was a great article. However, I think it was a very flawed move to even suggest taking money from a philanthropist, especially a politically-motivated one such as Soros. He's every bit as bad as, if not worse, than the government-corporate cooperative that we have today.

The way to a free Internet is through free markets. It is, in fact, the *only* way. There must be competition at every corner, free from regulatory influence that serves as a preferential meter for those that fund the ruling class.

I have learned a hard lesson - being "pro business" is not the same as being "pro free markets". I believe that a capitalist solution is the only way to make this work, the dreams of a shared electronic utopia are nonsense that should have died with Marx. The idea that Obama's buddies at Google are part of the solution is equally flawed.

If you want your Internet back, take your government back - and cut that mother down to size and make sure we never, ever make the mistake of choosing a security blanket from our nannies over being free ever again.

I don't know if you see the real issue...connection.

Phone lines are operated by regulated companies, coax lines are operated by regulated companies.

I believe what the author is implying is this:

The only way we'll be able to change the internet is to not use the internet and rather focus our attentions on creating a new method of distributed computing which doesn't involve any controlled means.

We do make choices, whether to have one particular set of breakfast or to have Blackberry over Iphone. Why can't we have the choice to choose which Internet we want? We already have chosen our connectivity so Internet itself should be no different. In fact we should have many Internets; i.e, the current Internet, the new Internet, the educational Internet, the political Internet, the pornography Internet, etc..etc...
Just like we have many cables or channels for our TV...

How about switching completely to mandatory ipv6+ipsec in transport mode? Will solve LOTS of problems once and for long time (ISP will have no way to tell what we are using our connections for, beyond fact we contact with some IP - everything almost encrypted: cleartext, tcp, udp, port numbers, urls). Unfortunately for ipsec we need some kind of certification system. But this should not be hard to achieve in some simple ways of proving you own (or at least) have access to IP. IPv4 will not work well in transport mode due to the lots of NAT. Fortunately Ipv6 will probably kick in in not long future. Having enabled ipv6, we can enable ipsec, and the try developing global tor-like network (with one hope). This will hide also the IP we (as a person) are connecting to (at least to ISPs), and make blocking IPs somehow unpractical.

Only way to regulate this will be to completely connect somebody off (or block ipsec, or make it all slow, which are all essentially the same and pretty agressive). Unfortunately there is practically no possible way to prevent this. One need to choose proper ISP to know what to do. Hobbyists will not build physical network infrastructure easly on global scale (yes, in big cities it is possible to build good MAN, especially using wifi and meshing, but who will have money to connect two such cities). Business is needed, and it already do the right job quite cheaply. The problem is with policymakers, not business.

By software/hardware we need to help projects like openwrt, etc, to enable more secure connectivity for home users. Start enabling ipsec with opportunistic encryption by default (desktop and server) in all major Linux and other open-source OS distributions.

Ah yes...Fidonet. How well I remember. It was 1988 and we had Evans BBS, in our small west Michigan town of 1,000. It worked beautifully, and brought us all together, long before Windows and ISP's. Your article, and the international conscience it speaks of is a powerful indication of how we are evolving beyond this failed paradigm. "Life finds a way."

What's this about running cables between individuals' homes being illegal? and the same being true for IR laser links (and, one assumes, point to point microwave links too.)
I've never heard of this law, please can you tell me exactly where I can find the details? I'd like to know what, if any, possibilities remain, and what's not allowed (sigh...sigh aimed at the massively nanny and big brother state!)

I too have thought that a network that was not owned by cooporations would be less biased toward certain types of traffic and more open, that is, supporting free speech and freedom of information.

I too (like many others here) find it hard to see it working without a infrastructure. It seems that almost any method that works on top of the existing infrastructures is compromised at some level or point in the network (the most obvious being at the ISP.) My (naive?) idea has been to use a cheap and basic method of connecting up the houses on a street using coax and a basic IC at each premisises - or even use bus ethernet or a modded version of this to better control anonymity. Using a type of 'IP masquerading' on a wide scale could also help to build such an anonymous network.
These 'local' networks would then be connected by enthusiasts, using laser or microwave point-point based connections.

I have got the impression that networks such as TOR and Freenet are not all that they seem, because they are also flawed - probably due to man-in-the-middle types of attacks, along with the fact that authorities can often force such groups into opening up their systems... however, this brings up problems, because policing is useful. It is a worry that such networks would be jumped upon by those wishing to carry out crime, and use them for selfish reasons. From what I have seen/heard of Freenet and TOR, whilst they are used for purposes of freedom it seems that this usage is far outstripped by people using them for pornographic and even worse. Then there's the problem that SPAM will bring when such a system really takes off.
It seems to me (naively?) tha the only way to get around such problems is by technical 'trust' mechanisms. Without policing o f these trust mechanisms it seems that all that would be possible is trusted individual-individual transfers, which is essentially the same as encrypted&authenticated email. Maybe the whole new network could consist of sub-groups (e.g. Special Interest Groups), each of which could govern the trust-status of their own members...

A people have pointed out that what is actually needed is a change in real society, not in the not-really-existing universe of cyberspace. I agree. There are many issues in the real world that need fixing, but in fact it is for these very reasons that people are here on the 'net thinking about creating an alternative network. It is because they are not entirely satisfied with all that is going on that a revolutionary group/system is required. It seems that the authorities are ignoring the people, and that democracy is disappearing to an even greater extent than ever. Such an anonymous and free speech network would facilitate a group to challenge the government on these issues, that is to form a powerful group of like-minded individuals that would be much harder for governments to ignore, as they have done to date. For me, the issues that are at stake are freedom-of-information (massive reduction or even full removal of copyright/intellectual-property, for the promotion of education and to allow the faster advancement of commerce and associated technology especially), and privacy. I think that if the authorities want to know what I'm saying/up-to then they should contact/investigate me with my knowledge, not be automatically and surreptitiously scanning my communications etc. In other words, rather being in some kind of technological panopticon it is much preferable to have human-interaction with authorities which secures our dignity as humans *and* reminds us of the social contract that we have with our fellow humans - to look after one another (which is why I believe we should of course support police/authorities this way.)

Something between having a nanny state and a system completely overrun with cyber-criminals and annoying spam would be great. But I think there are still a lot of fundamental questions that need to be thought through first - policing, trust, criminality, intefacing with external authorities. Then the ways in which a system can be legally created - if this is even possible.

"Is it time now to start thinking about a new and possibly non-existant public utility ... for the transmission of digital data among a large set of subscribers?" -- Paul Baran, On Distributed Communications Networks, RAND Corporation, 1962.

The internet is deeply flawed, there's no doubt about it -- and some of its flaws are becoming more important as they're recognised and exploited by those who desire power. But before we start building a better internet, let's ask a couple of awkward questions. First, why do we want an internet at all? And second, how did we get the internet we've got now?

I ask the first question because I don't think the reason we want a new internet is the same reason we wanted FidoNet. Until perhaps five or ten years ago, the main attraction of the internet was that you could do things online that you couldn't do anywhere else. You could talk to a stranger on the other side of the country (or the world), read obscure (or just banal) information that wasn't available in your local library, play the game of masks and nicknames, enjoy the still-new thrill of seeing your own crazy ramblings published for the world to read. Nowadays? Google Maps and Facebook. The internet is an extension of your offline life.

But the thing is, the powers that be never gave a damn about the old internet -- and they still don't. We could reinvent it and they wouldn't care. But they give a damn about the new internet, because it's part of our offline lives, part of the world they control -- and if we reinvent it, they'll want to control that too.

So let's ask, are we building a new FidoNet that isn't controlled by government and corporate interests -- or a new Facebook? Because one may be possible and the other not.

On to the second question -- how did we get the internet we've got now? I ask this because I'm not sure it's an accident that the internet started in the military-industrial complex, where it enjoyed decades of funding and political respectability before a bunch of dirty anarchists hijacked it to share porn and bomb-making instructions. If those lefty weirdos had started the project back in 1962, do you think J. Edgar Hoover would have let them build a nationwide communication network to escape government and corporate power?

Let's assume, pessimistically, that no social progress has been made since Hoover and it's still impossible for shady beatniks like ourselves to build national infrastructure. What, then, can we build? We can build on national infrastructure. We can use the internet as scaffolding, like FidoNet used the telephone network. And we can use the telephone network, too -- and wifi, and Bluetooth -- thanks to the miracle of packet switching, invented by the RAND Corporation in 1962.

We don't need to reinvent the internet. We can use the internet to help us build the kind of network we really want.

"Escape always takes place in The World As We Know It." -- Brian Massumi, A User's Guide to Capitalism and Schizophrenia, MIT Press, 1992.

There are a lot of good ideas in the comments and it is encouraging to know that folks are thinking ahead of time about alternatives. In the mean time, what I'm hoping for is a few corporate "good netizens" who share our goals to the point that they put their future on the line to support them. I may be dreaming, but if a few big carriers and content providers (these would need to be real heavyweights) would sign some sort of manifesto, and agree to take real action to protect or increase the openness of the net, policy makers and coporafascists might think twice. Honestly, if they would think it through, such a move might well represent rational self-interest. Too many companies think short-term. This is a long haul issue.

The internet , was never designed from the ground up to be a publishing platform for the rich content we se today. Many of the technologies were originally hacks on top of the base stack in order to solve a need.

Today its a sophisticated bundle of protocols, scripts and software that deliver the content we see today.

Its a good point - What if we were to redesign the internet to solve all the issues we have today?

My number one gripe is the hacks we have to put in place for websites that provide multiple downloads.

(I run a digital downloads website , and there is no way with current browser technology to pick a destination folder on a users machine and have 20 mp3s download one by one into the folder) There are various Java (which we use (which feels horrible!)), possibly also flash based hacks that provide this ability. This is just one example of an everyday problem that is limited by the technology infrastructure in place today.

I think the closest thing to this would be the piratpartiet which isn't very big or even capable of supporting a world wide infrastructure like what we are aiming for. but they have a good start by having their own ISP, I believe they either have or are planning on having their own satellite as well.

indeed, wireless mesh networks, already deployed in a number of cities are the future of local connectivity, when big corporations will loose their tight on average people. We should organize best and concentrate in building these communities in people's neighborhood, not only in our labs or garages ... !!
look at babel routing protocol and AHCP for network config. it looks promising...

I also cannot help but periodically spend time trying to dream up a solution to this problem. I'm glad someone else recognizes that you ultimately will not be able to stop powerful interests from censoring the Internet.

I usually refer to my dream of a FYETI (F* You Evil Telecom Industry) device--a free (as in freedom) digital communicator. But the problem with actually finding a wire/wave on which to communicate globally is extremely challenging.

(1) A WiFi mesh can work in urban environments but there are choke points all in between.
(2) Stenographic methods could mask communications over various FCC regulated channels but requires layers of trickery expensive on CPU to retain stealth.
(3) I hear the new bands from between TV channels are possibly opening up to a longer range WiFi. Are there possibilities there?

Being fully P2P and stealthy is entirely feasible. One legitimate reason governments won't like that is its usefulness for illegal activities and terrorism. We need to think of ways to police such things without handing out dictatorial censoring control to governments.

Overall, I think the Stenographic approach is the only fundamentally workable option. But that can work over a full mix of other communication mediums.

Matthew

Not long after this fabled new Net is live, you'll start getting porn and all manner of vile crap on it. Unless that's what you want?

First I've heard of this one. So, reading between the lines, I take it that enough proliferation of such local area networks would pressure ISPs to allow uncensored traffic in and out of them, in exchange for access to them? But Internet access remains from each home through ISP, unless you share connections through a mesh network which they can merely ban. Actually, aren't local communities already legally not allowed to create their own communication networks without FCC approval? Local wifi mesh networks are common in Turkey and some other countries that allow it, but that doesn't stop the censors at the ISP and backbone levels.

A P2P stenographic protocol is one way around this. It's just very resource hungry process to encode (when done in diversely sourced layers).

I think the first steps should be pretty clear:

(1) New internet over old internet. Like IPv4 over IPv4, we should be able to connect to the new network over the infrastructure of the old. That doesn't mean that we have to use the old infrastructure, but that we can if we have the capability and inclination. This is necessary for mitigation and/or migration.
(2) Tor-ified e-mail. It should be a simple matter to set up a mail client that works over Tor and that incorporates full public key encryption. It might take some jiggering, but you should at the very least be able to set up a makeshift listserv that has RSS feeds that update with the latest messages. Publish it on your computer in an RSS feed the listserv is set up to check, cryptographically sign it and encrypt it with their public key, and the listserv decrypts it, reencrypts it with each recipient's public key, and the recipients retrieve it via RSS password protected by HTTP basic access authorization. You now have a message that you can be sure came from the sender and has not been tampered with--so if it's spam, you know who the spammer is, and you do not know who is sending messages unless you're the recipient. You would probably also want a list of message-IDs for the messages downloaded to be kept on each recipient computer, so that the messages can be removed from the queue once the other computer receives them. I'm sure this could be streamlined, but this method works now.

(Please do not construe this opinion as representing that of my employers)

Something to add to my previous comment:

We don't have to make a decision quite yet about the kind of technology to use for our evolving last mile - someone jokingly but quite appropriately said we should call it the hinternet - infrastructure.

If we push that decision down to the local cooperatives, then the best solutions will emerge through practical experimentation and they will be emulated. There may well be a mix of technologies to use, depending on what environment the community operates in and how distant its members are from each other. There shouldn't be any insurmountable technical problems.

We have choices.

There is normal Wifi, the thing used to make in-house networks. With better antennas and smart drivers that do beam-forming, those things are quite sufficient to start building up a mesh in a city or suburb.

Rural areas that have greater distances to bridge can use WiMax which, I believe, reaches farther than WiFi.

TV spectrum in the (I believe) around 700 MHz bands is becoming available and is being labeled white spaces in the US. This would be ideal for rural-type connectivity that needs to bridge greater distances.

Ham Radio can bridge huge distances like half way around the earth, but it is restricted in bandwidth. It's quite sufficient for emergency use though, sending written communications long distances.

In the Czech Republic there is a guy who has made, and who describes how to make, an optical line-of-sight interface that could go rooftop to rooftop, using red light or infrared. It has trouble only in very bad weather conditions such as a snow storm. The advantage of optical is that it doesn't add to the electromagnetic pollution. Those microwaves used in WIFI, WIMAX and mobile phones may not be the best choice health-wise, as they are modulated by packet frequencies (about 200 packets per second are sent in mobile phone traffic) and those low frequencies apparently lock onto and disturb intercellular comunication in living organisms.

The military has been looking into optical mesh networking, and they have produced some papers, perhaps even hardware that can be used, but nothing of this has been filtering out into the civilian area as far as I know. Optical has even been tested for mobile applications. It is routinely used as infrared optical in convention hall translation devices.

We can even use the wires of the soon-to-be-abandoned fixed phone network. All we have to do is wire the neighborhoods together with existing phone cables (once the phone companies get out of the way) and we have a basic network that's quite acceptable for any use we might want to make locally.

Basically, to construct our own net we aren't bound to any one technology. Whatever fits best can be adapted. Once the hackers start working on this and get some support, I believe we'll have all kinds of options.

The question of software is still an open one. What to put on those local networks, how to cache data, how to allow everyone to publish their own data to the network needs to be worked out, but it's also being worked on. Starting the local communities (cooperatives) communicating will stimulate software development to invent uses for that new part of the net and to handle/address traffic.

There is already a new-internet-over-the-current-internet. It's called i2p (http://www.i2p2.de). I'm surprised it hasn't gotten more attention. It's not an anoymous gateway to the existing internet, such as Tor. You can only access i2p sites and you can tunnel any traffic in to it. There is no central DNS so anyone can register an address and host their own "website" called an eepsite, for free. And the source and destination of traffic are hidden. Even with an eepsite, users could not trace the location of the eepsite back to you. Don't ask me to explain how that works though.

Some ideas from #NewNet:

1. Use part of the unlicensed WiMax spectrum (yes, there is such a thing!) for this new network

2. Focus on getting several people in one localized area using it but use the Internet (at first) to provide connectivity between distant nodes (as Sepp suggested above)

3. Release a forked version of DD-WRT that enables a home router to become a node that connects local wireless users to the rest of the network over the internet (P2P VPN?)

Live IRC chat about these ideas: http://webchat.freenode.net/?channels=#newnet. Chat log, to get caught up: http://bit.ly/NewNetIRC1

ingredients:

1. wireless mesh network
2. identifiers based on unique MAC address
3. GPS style relational position mapping using triangulation of signals.
#3a (used to help determine fastest routes)
4. software layer to deal with above

future addons:
naming layer to put a name to the identifier (like a domain name)
linkable identifiers to allow distributed hosting for high traffic sites
ex: google.free could allow 100s of identifiers so that when you enter google.free into your device, it sorts the google.free identifier list by relative distance and goes to the first one that gets a good "ping"

We are very close to being able to use fiber optics from house-to-house, with every computer a router.

Use geographic addresses (lat/long) and routing is easier. There was a patent on that in the early 1980s, should have expired.

Neal,

Am completely with you on the role of free cities and alternative networks as centralized systems recede - @paragkhanna in Foreign Policy also has a good sense of their emergence in a new Hanseatic era (http://j.mp/fNe9GP), as does Paul Romer based on his work to seed Charter Cities.

You may be interested as well in slides from this 2009 talk at Princeton, on how virtual networks can help catalyze creation of free zones and free cities in coming years:

"Social Networks And Free Communities" - http://j.mp/aSKLHX

Look forward to further thoughts on turning the vision into action.

Best,

Mark Frazier
@openworld @peerlearning

#newnet irc.freenode.net

Discuss!

"2. identifiers based on unique MAC address"

you do know that it's trivial to spoof a MAC address, right?

identities have to be based on asymmetric cryptography; MAC addresses are pretty much useless as a guarantee of identity. identity is ensured by cryptographically signing things (usually hashes of other things, actually).

just sayin... ;)

What we need is a totally transport-agnostic store-and-forward (like FidoNet or usenet) or distributed cache (like Freenet or Tahoe-LAFS) system that can be tolerant to high latency and basically run over all kinds of transport you can think of:
- encrypted links over the "corporate" internet (why waste precious bandwidth?)
- sneakernet
- telephone systems
- bluetooth/wifi meshes
- HAM radio (steg+fun_with_markov_chains can make encrypted data seem like structured data, like spam)
- steg-over-youtube-videos
- steg-over-twitter-spam
- steg-over-flickr-spam
- steg-over-whatever
- hard drives over postal mail
- pigeons
- you name it

That way, each node forwards messages through their available channels and everyone can choose which types of links they prefer to maintain in order to maintain acceptable levels of latency/bandwidth/availability.

It's easy. Now, start working everyone! ^^

Very exciting to see the surge of interest here.

This comment thread is getting a bit unwieldy, so I've opened a forum over at the connective where we can focus these discussions. You can log in using twitter, google, or open id, so no need to create a new account.

http://theconnective.net/talk

peaceandlove
/mark
http://theconnective.net

The original Web was on a single computer, unix

Truism - Cannot defeat human nature.
Question - How to use human nature to advantage?
Answer - Better content served with better relevancy to time and place.
Discussion -
Content is the first holy grail of VC - the best content wins.

Geolocation, relevance to current time and location may be second, it is getting chased hard by todays VC.

Local serving can win on both fronts. First step is every day people believing it is possible. Fanvision, Starbucks local, the tech is there, receiving limited exposure.

Raising awareness among every day people, so they will understand it is a possibility that can be delivered by their neighborhood geek is the difficulty. Articles like this are a first step toward raising awareness. Pirate radio? Ah, too few remember the good ole days of pre-digital out of band communications.

Truism - No free rides.
Question - Who will pay for it?
Answer - Local merchants who benefit from it. The same people who have always paid for media communications like magazines, newspaper, radio, tv.
Discussion -
Just Google revenue last year was $24 billion. Far too much of it was local merchants banging heads with each other only to get killed at the end game by better financed corporate conglomerates.

For local digital media services to make sense, advertising must pass the test of viability in a business sense, or in a community service context. Local businesses are your friends and neighbors, they do have something at stake.

Truism - Hardware and software are required for local serving.
Question - Which standards to apply?
Answer - Let free market capitalism work.
Discussion - All required tech, hardware and software, is basically commodity level stuff. Anything available is far better than anything from the early days of the Web and net. My own praescope.org, began in July 2010, is an attempt to frame a few ideas, while waiting upon the need to be widely perceived.

Local serving is easy. Whatever the local geek community understands well enough to deliver a product people enjoy and use works fine.

Truism - It is not a Web until it starts connecting everything together.
Question - How are the edge connections going to work?
Answer - ????????
Discussion -

The availability of 32GB and up SD chips, multi-terabyte raid arrays that fit in a ruck sack, makes storage capacity and raw information transfer not much of a problem.

The edge connections and overlaps are standards yet to be defined. But how to define these in advance of knowing what is going to happen when local serving becomes widespread? Will it all be text? Will it all be video? Will it ever make it off the ground?

Worrying about step two when step 1 is yet a distant glimmer of hope is pointless. Sneakernet worked before ethernet. Similarly sneaker net will play a role while edge connectivity waits to see if local serving gains enough traction to work. Speaking from personal experience, it ain't easy. Expect lots of blank stares. I tried a catch phrase, hyperlocal new media, more blank stares. This is from Southwest, MO, hopefully your mileage varies upward.

The Internet! ... fun while it lasted

He is right about the problem. But the solution can't be to try to compete with The Internet. It must be in some way to build on top of it. Examples (though not final solutions) are things like Tor https://www.torproject.org/
and
GNUnet https://gnunet.org/ .

Before I leave the world ,wish more people know the truth in China, please stop the stupid and evil continue to exist.

I like what I'm seeing here. Maybe we should form a grassroots wiki to discuss our options for this project.

In addition, 1 idea I immediately thought of while reading this article was that each person who wanted to be "connected" front the cost of a customized multi-homed "transceiver" comprised of components hacked together from a wireless router, cellular GSM repeater, HAM radio, Satellite phone, etc... and whatever means they have to connect to an existing architecture (IE: cellular services, wired internet, etc) be used to allow long distance comms while the shorter distances can be handled by a free wireless means of communication. If every "node" communicated basic data such as public key, services available, and ports via a predetermined frequency/format... one could wardrive and map overlapping signal hosts to determine dead zones where we could focus on a community driven investment in a new node.

Basically the idea is to take advantage of existing networks while attempting to roll out a proprietary encrypted wimax net as a fallback should services or connections fail on existing networks.

Anyway... good article. Got me thinking a bit...

Time to bring back Usenet?

isnt the phone system that fidonet uses a coprorate/centrally controlled system? So even that model was still dependent on some centeral system

The answer, I think, is plurality. The internet is fragile from a democratic (and I think, ecological) perspective due to it having an effective monopoly of this kind of information transfer; you can choose service providers; but they are ultimately the same kind of source. Corporatised and regulated by the government (rightly or wrongly). This isnt bad, as such, but it IS open to abuse, as the article points out. But if there is a plurality of sources/providers, AND a plurality of independant communication channels; then we build some serious redundancy into our communication system. We use ALL the proposed methods, with a focus upon the modularity of the systems, and encourage (through design) the use of multiple networks simultaneously; such that a user might be browsing the web through his ISP, whilst also sending any sensitive information through more secure but slower local networks (wifi, some form of cable and/or utilising mobile phone networks); his ISP may be a local communities efforts rather than a distant corporations, or indeed a pirate effort (from offshore, perhaps?), many of these networks would be ad hoc. Any one of these systems is vulnerable alone, but much more resistant in concert, to monopolisation by significant powers. The costs for governments to censor this kind of methodologically plural, ad hoc, network is going to be orders of magnitude higher than just censoring the internet, which is hardly entirely successful even in China's case; whilst the cost to the average consumer should, after some significant outlay, decrease due to the competition of paid-for and free & secure(r) but slower networks. There will always be some level of interferance by those with the interest and power to do so (and not just governments and corporations, look at anonymous, or the various 'digital' crimes), the point is to create a system robust enough to tolerate a large amount of that, and still deliver the basic service, not to create a system that will never suffer those kinds of abuse.

Just launched a knowledge base for the connective ...

http://theconnective.net/fire/

part wiki, part Q&A.
seeds for a grassroots citizen owned internet.

Another compendium of related projects posted in an article here:
http://www.datelinezero.com/2011/01/29/3-projects-to-create-a-government...

We need a decentralized internet at least for emergency situations or when the big government decides we no longer need to communicate with each other.

We can build our own infrastructure between these diverse nodes. Each node can have its own organisational ethos and currency.

Just to be pedantic, almost. If you are a ham, you can use encryption for your spacecraft telemetry etc. And here is the interesting part. You can use encryption legally at home too. The trick is that the encryption must be public. That allows "obscure". And perhaps we could flood library of congress with 100 trillion keys? Hmm, just blue skying, but I really like corner case stuff.

Speaking of corner cases, I am unaware of restriction on laser signal encryption. Also, you can cheaply use a kitchen microwave to generate a tight beam. This is not the sort of stuff the FCC spends time looking for. And anyway, what they mainly look for is stupid interference, largely as reported to them by the victim.

My best take on this sort of thing has been that we should have multiple transmission methods and the default is the oldnet. But if you are playing serious then you do not need a lot of bandwidth. Maybe a hologram in the US Mail? A USB stick at a drop point? Keep an eye on the corners.