The Internet is a marvelous thing, but it requires an extensive infrastructure of expensive equipment. Because of the investments required to build and maintain the network, the Internet has come largely under the control of media and telecom corporations, who have financial interests in controlling the way people use the Internet to increase their profits.
The explosion of wireless technologies, including cellular, wifi, and even bluetooth have demonstrated that network communications need not depend on wired equipment. In this article, I am going to attempt to outline a basic description of how a free (in both senses) wireless communication system will function, and the technologies required to implement this service.
Two key principles behind this effort are open source, and cooperative interoperability. This means that all the hardware and software is open to study and modification by any interested party, that all the disparate systems and implementations can communicate with each other, and that each owner of the equipment in question sacrifice some of their own resources to support other users, while receiving support from all the other users who sacrifice some of their own resources.
Fundamentally, this network will take the from of a wireless mesh, however unlike existing implementations, the network will not be restricted to a single frequency band or set of protocols. The infrastructure of this network will be provided by the inter-operation of numerous individual nodes owned by businesses and private individuals. Users will not be an end point like in the current design of the Internet, but another interconnected node in the local network. Unlike the Internet, ownership of the infrastructure will be extremely diverse by its very nature and design, and so not subject to control by a handful of large owners. Even if a single owner controls a portion of the network, their control will only exist locally, can be easily bypassed by other local users, and cannot be used to exclude or monetize access.
Any individual node will have a coverage area modulated by its emissive power and power supply. Mobile users will typically cover a smaller area while hard-wired, power-line emitters will be able to reliably cover a larger area.
Each node will poll other nodes in its range to create and maintain a network map of static and mobile nodes. This map will be correlated against the physical geography using GPS coordinates. Nodes will maintain data about the reliability and connectivity of other nodes so as to route their transmissions through the appropriate paths to reach other nodes. This will include path-finding algorithms to ensure quality of service concerns such as throughput, lag, and number of hops.
Mobile nodes present some interesting problems with regard to power usage, storage constraints, and the ever-evolving network map. As such, mobile nodes will handle a relatively small portion of network traffic in comparison to static nodes, which will be preferred for infrastructure duties. However, mobile nodes will retain the same capabilities as static nodes, and in many circumstances mobile nodes will communicate directly with one another. Transmissions will prefer the least hops and the lowest power consumption, so when mobile nodes are in relative proximity, they may communicate directly, but if there is an intermediate wired node, they will modulate their signal strengths for the intermediary and thereby reduce their own power consumption.
When searching this network for information or a specific node, the requesting node will poll its neighbors for a match, who will then direct the request to their neighbors, and so on, until the information or node is found, or the network is exhausted. Nodes will attempt to map and preserve path information and content from other nodes to increase speed and reliability for future searches; but also update and modify this information when it changes. For this reason, static, hard-wired nodes will be preferred for infrastructure duties.
Data on nodes will be identified in several ways, including public, private, and restricted. Private data will not be visible or available to the network, whereas public data will be locatable by links and searches for access and downloading. Restricted data will only be locatable by parties with the appropriate keys or credentials. Regarding downloads, public and restricted data will be identified by hashes and transmitted in a BitTorrent swarm. To enable streaming audio and video during download the data will be preferred in order, for static images and pages it won’t really matter, but in any case, local sources are preferred to minimize network activity. For example, someone who requests a distant file will receive the hash, and download the content from the most local available sources.
The software and hardware to enable this type of network is largely extant. Open source operating systems such as GNU/Linux can implement the software packages needed with little modification. The hardware is another story – most of the components exist but not in a popular or integrated implementation.
Software requirements are as follows – and keep in mind it is important that the implemented technologies are all open source – an OS like GNU/Linux, mesh networking, software defined radio (SDR), protocols for a wide array of wireless standards, a BitTorrent style distribution system, a wallet-style personal information repository, full disk and packet encryption, network mapping, web, email, and domain servers, and the ability to maintain and represent network information such as web pages even when active communication with the originating site has ceased, similar to Google Gears. This list is not exhaustive, I’m certain some necessary technologies have been unintentionally excluded.
Hardware requirements – also required to be open source – are ultrawideband (UWB), interference-aware electronically tunable antennas for multi-input multi-output (MIMO), and compatible with the software defined radio. These devices will require a large storage capacity to maintain the network maps and content information, and their processing capacities must be capable of providing their share of routing, page serving, and other services typically provided by existing infrastructure service providers.
These requirements, while not typical of existing network requirements, are not onerous, are feasible, and are well within the ability of current technological development. The capabilities that emerge from the integration of SDR with tunable MIMO UWB are beyond anything currently offered by any technology developers or service providers.
A UWB SDR could receive and transmit in any existing communication protocol, from WiFi, Bluetooth, CDMA, GSM, GPRS, PAL, NTSC, AM, FM, HDR, HDTV, ham, emergency channels, in fact, any type of signal within the bandwidth of the tunable antennas. It could also implement any other protocol we might develop, to maintain full backwards and forwards compatibility. The device would monitor the frequencies and not transmit on a frequency that is being used, to avoid interference or disruption of existing services. Signal strengths would be modulated by the distance to the receiver to reduce power consumption and further reduce chances of interference.
The technologies I have described would provide free (as in speech and beer), universal wireless communications services for anyone within range of other nodes. This technology would not require an extensive, centrally owned and managed infrastructure. The necessary infrastructure would grow virally, piecemeal as individual owners added their resources where these resources are needed. The design of the network means that each portion of infrastructure provides more resources than it uses, so it is fully scalable and becomes faster and more robust as additional users are added. Initially, users may still connect to necessary services through existing wireless and cellular services, but as more users adopt these technologies the size of the free and freedom-respecting network would grow and spread through daisy chaining. Local networks would continually grow in size and subsume each other resulting in ever larger networks, until enough nodes exist for nationwide coverage. Specialized repeaters could be installed in areas of limited population or coverage, or in high traffic areas for better quality of service.
I believe the unique capabilities of this technology will be well worth the investments required to develop them, and that within a few years integrated chipsets will be cheap enough for widespread consumer adoption.
In addition to the inherent qualities of this technology, there will be a widespread social and economic impact by its adoption. Many existing industries will undergo great changes or they will face extinction. ISPs, telecoms, cellular providers, and other forms of service providers, as well as media and content owners, will be forced to adopt new business models. In upcoming entries, I will discuss possible solutions for these industries.
What can you do to help? I would like to announce a new open source community driven initiative to develop these technologies, to build prototypes, and to eventually design retail products for consumer purchase.
We need hosting services, websites, and forums to promote these technologies and attract others to participate in this initiative. We need designers and engineers of all stripes. We also need public support. This project will have to overcome legal, political, and economic resistance from the entrenched industries and financial interests. We must take heart and persevere. All great change comes from the outside; true innovation never comes from within an existing industry.
I hope you can see the potential of what I propose. It will not be an easy task, but it is well worth doing.
If you have any questions, please post them here and I will do my best to address them. Thank you all for reading, and if you are interested in helping, please contact me at chaos.motor at gmail.com.
Note: I have edited this post in response to concerns about my ambiguous use of the term free, and to append GNU to the term Linux.
January 11, 2008 at 10:01 pm |
I’ve had the same idea in the past, too bad I have no programming abilities or knowledge to make this work.
I do have one question. Do you think land lines would be integrated into this idea in the future?
January 12, 2008 at 8:10 pm |
In a continuation to my thoughts from the other post, I was musing about it, and something you said in this one struck me. You state that “ISPs, telecoms, cellular providers, and other forms of service providers, as well as media and content owners, will be forced to adopt new business models.” I think what we’re seeing right now with the music industry is a good indication of how such a transformation would go. The RIAA knows that their business model is sinking, and so are trying to milk things for all they’re worth, with their slews of lawsuits that do little more than extort people.
This makes me feel that though corporate interests will likely resist any attempts to reduce their hegemony over information, there is a sort of inevitability to change–they’ll try to profit from it, but there’s something of an inexorable march of shifts in the paradigms of technology and business. Let’s hope it works out for us.
January 12, 2008 at 9:59 pm |
You’re absolutely right as usual. The only issue really with the modern state of things is that the economy has gotten out of balance, favoring a minority of interests to an extreme degree and thereby over-representing them. All I’m suggesting is a natural evolution of current technology that will rebalance the equation. If we don’t burst the bubble and let out the pressure, the imbalance could topple our whole system of democracy.
January 13, 2008 at 5:12 am |
I certainly agree. Perhaps what I’m struggling with is to envision a way in which such an evolutionary or transformative step could take place within the current paradigm in a way that it doesn’t engender opposition, because everyone– us, the economic and political elites, and the economically and politically disadvantaged– gains from it. Of course, such a step would be for the long term benefit of everyone, but if history has taught us one thing, it is that rarely do humans operate with the interests of the future in mind.
I wish that I had the technical expertise to give any commentary on your specific plans and ideas. If it will function in the manner you posit, I think it’s wonderful. But my areas of specialty tend more towards the social analysis of my disciplines, so I can only really comment on the peripheral issues for such technological developments.
January 13, 2008 at 4:02 pm |
This is a very clever idea indeed.
My only concern is “This map will be correlated against the physical geography using GPS coordinates.” which affords much less privacy than the current interwob.
I’ll contact you via PM on SA with a few more thoughts later.
January 13, 2008 at 10:50 pm |
Mr. Mnemonik,
Thanks!
As for your privacy concern, you may know where the nodes are, but unless you have their permission, you won’t know /who/ they are unless you’re within visual range. But yes, there are issues to be addressed and concerns to be allayed, as with any new technology.
January 14, 2008 at 5:29 am |
[...] Chaos Motor Driving Toward Change « Designing a Free, Universal Wireless Communication System [...]
January 14, 2008 at 4:06 pm |
I’m really trying here Chaos.
It’s not “free” but wireless from anywhere is available. I use EVDO, and there is no need to find a hot spot.
I pay for the service, but that is how one insures they keep upgrading the technology, I like knowing I am providing work for others, that they can continue to support their families.
Having to pay for a service is not necessarily a bad thing. I in turn need to keep working, and that keeps me out of trouble. It also forces me to give my best.
January 16, 2008 at 9:49 pm |
avenuegirl, I appreciate your continued attempts to understand what I’m promoting. I know it’s tough to think about things in such a different light.
You can get wireless anywhere there is cell coverage, but that doesn’t provide a tenth of the benefits I’m describing, and it does cost you. As for finding hot spots, you do have to find them, as you have to be within range of an EVDO-equipped cell tower, but cell towers are fairly ubiquitous in modern cities so it’s rarely an issue.
But I’m talking about wireless that doesn’t depend on cell towers or subscription, a true P2P wireless service that connects any machine capable of it, to any other machine, directly, and uses each person’s machine as a node for transmitting past your own horizon.
The technology I’m describing won’t require nearly as much centrally-owned infrastructure, so it’s upkeep and upgrade costs are spread equally among all users without the overhead of also supporting a central organization.
Having to pay for something isn’t inherently wrong, but having to pay for things that don’t require paid service as a fundamental aspect of that service ranges from silly to stupid to totalitarian. Now, I admit we don’t currently have a way to avoid paying for wireless data services or at least the connection from wireless to the larger network, but that’s what I’m trying to fix.
But further, paying for something that doesn’t have to cost doesn’t support anyone or their families, it only enables your own exploitation. You drive a car, right? You don’t pay a horse trainer to keep a stable of well trained horses simply to support the trainer and their family, do you? But cars replaced horses, right? Just as what I’m describing will replace paid wireless services.
And what if every time a new technology replaced an old technology, the public had to shoulder the cost of continuing to support the old technology just because, well, those people have families too? It’s not a justifiable position. The people employed in these businesses will find work elsewhere, and new businesses will pop up surrounding the new universal wireless service I describe, just like mechanics and tuners and customizers and oil change shops and car-washes and drive-throughs sprang up surrounding the popularization of the car.
Technology changes, and some old technologies die. We don’t subsidize horse trainers, we don’t subsidize carriage makers, we don’t subsidize telegraph men, we just let those businesses naturally respond to the new condition. Things grow old and die; other things are born and mature; that is nature and nothing is more natural.
February 4, 2008 at 6:45 pm |
What I find interesting is your “software defined radio.” You are aware, of course, that transmitting on different frequencies requires different physical hardware?
The simpler solution – and more technically effective – is to remove the government sponsored monopolies in the broadband sector, and encourage healthy competition. Capitalism hasn’t failed us yet.
February 4, 2008 at 7:07 pm |
Alex, sorry, but you’re wrong. Thats kind of the entire point here is that SDR means you don’t need different hardware to use different protocols. Go read up on SDR. I’m talking about a great leap forward in networking technologies, and your response is, “Nah, what we have is fine”? You obviously didn’t understand this essay.
Your trust in the government to do anything useful is ill-founded, as is your belief in capitalism. Capitalism fails us every day, while our elections are shams, our dollar is collapsing, our economy is in tatters, children starve and millions go without health care. Capitalism is what has gotten us to this point, you don’t honestly believe that it’s going to solve the problems it created, do you?
March 23, 2008 at 6:08 am |
I have dreamt of exactly the same kind of internet infrastructure based on an autonomous open-source wireless mesh network that completely dispenses with central servers owned and controlled by a few corporations. chaosmotor, you seem to have the relevant technical knowledge to implement the idea. I am very glad to find people like you with such a great vision. My major concern, however, is that certain elements of the governments, more than the corporations capitalizing on their current business models, will try their best to prevent this free universal internet idea from coming true because they will perceive a completely ubiquitous self-evolving autonomous internet as a grave threat to their “control of information” which is absolutely necessary to successfully run their behind-the-back businesses as usual. I think the realization of this free universal communication infrastructure can be one of the key prerequisites to the realization of a true democracy.
May 5, 2008 at 8:37 am |
Somebody already has implemented and commercialised a similar idea. But they took it a step further by integrating it with the internet. So it does not have the inherent flaws of relying on unreliable end user nodes. i.e. an option to bypass a critical choke/lost node.
All we have to do now is wait until a significant number of router owners on earth have fon routers then we officially launch a firmware fix for a free alternative to isps one day.
Aside from the advantage of cost free information, ISPs still exist for a reason, reliability and speed. People with gaming consoles/pcs want low pings. You can’t have that with adhoc networks with its nature of many short ranged packetlossy nodes. This idea may be popular with file sharing folks but bear in mind the RIAA is moving to charge us a $5 monthly fee through our ISPs so we can download all we want with no legal implications.
The argument of trust in the government is flaky, wireless networks are the easiest to snoop on. Encryption only requires time to crack. All the government needs to do is place “wifi data” collection boxes outside one of the nodes in you local neighbourhood network and send the data back for processing. Open sourcing the constructs only makes all these easier.
On top of that what if I want to buy something online without my pimply neighbour watching every packet that leaves my home? I’d rather use my wired links.
Here’s the fon router initiative if you’re interested:
http://www.fon.com/
cheers
July 3, 2008 at 7:10 am |
Netsukuku?
http://netsukuku.freaknet.org/
January 2, 2009 at 2:20 am |
Great idea and great vision. I hope you’re able to implement this project.
As mentioned by another commenter, the GPS requirement does not sit so well with me. For a mobile node this might not be an issue, but for stationary node (read somebody’s home) you could then map an ip address to somebody’s residence. This would definitively have some privacy ramifications.
But, I also don’t really see the reason to have this physical GPS based map. Aren’t we interested in finding the fastest path between two points rather than the closest physical route. Although in many cases they may be similar, this might not alwyas be the case depending on different node’s equipment and current load.