Furthering An Open Internet with Open Wireless

To build a completely secured and decentralised Internet infrastructure requires looking at the problem from many angles and on many layers. Currently, MaidSafe focuses on decentralising the Internet on top of existing networking layers using as many modes of connectivity as possible such as WiFi, or Ethernet to connect to the SAFE Network’s global, peer-to-peer Internet. It is extremely important for individuals using the network to have access to such connectivity methods thus it must be an area of consideration. That is not to say MaidSafe is looking to build these protocols, we have our hands full already, but rather support efforts already in development which align with our mission to decentralise the Internet for everyone.

Of these connectivity protocols, the most compelling is called mesh networking and is based on peer-to-peer connections between devices like routers, computers and smartphones. The basic principle of mesh networking comes down to removing dependency on single access points like a home routers by directly connecting them to other home routers or devices which relay traffic. In the chance that a home router is no longer able to connect to the Internet, traffic will route through the connections of its neighbours. Alternatively, situations exist where ISPs or governments take aggressive measures to censor communications by blocking access to the Internet and having p2p connectivity allows communities to use local services independent of the greater Internet. Disaster scenarios also present justification for having p2p infrastructure in place for reliable communication.

Various organisations are pursuing the design, development and implementation of mesh networking infrastructure and the efforts of community mesh projects are particularly inspiring because of the conditions in which they tend to form and sustain themselves. For a majority of the people in these communities, sheer drive to bring Internet connectivity and the resulting access to knowledge to neighbours and greater community is a driving force. Interest in profit is rare and most groups are comprised of volunteer engineers and community organisers. By recognising the alignment in creating a censorless Internet, we can see the direct relationship between mesh networking and a secured, global communication and storage network like SAFE.

However, given the mesh protocols that exist today, the general consensus between these communities is that privacy should be considered for layers above. While decentralisation can increase overall security by removing a central point of control, it is not enough and a mistake to assume inherent security in decentralised systems like mesh networks. By building secure layers above, MaidSafe is in a position to offer a solution to security problems of these communities in a way that compliments the general ethos of decentralisation.

It is often the case in these networks that VPN servers are used to protect the node operators in the network from potential legal consequences of others using their Internet. Additionally, servers in various countries are used to circumvent censorship such as the situation in Germany where many YouTube videos are blocked and the only way to view them is accessing YouTube services from a proxy server outside of the country. Decentralised, secure technologies like the SAFE network can be used to remove this need for a server completely. Even more exciting is that storing data on SAFE and accessing it through a mesh network enables possibility to bypass the use of the IP protocol completely. By having complimentary efforts such as mesh networking and SAFE, we can begin to see a digital world with much less opportunity for central control in large corporations such as existing ISPs and cloud hosting providers and much more freedom for the individual users.

6 comments

  1. How does MaidSafe and Mesh networks go together though?

    As far as I understand and recall how MaidSafe works, data is indexed and stored by its hashes and distributed to the “nearest” nodes in a “XOR-space”, which could be anywhere in the world. Thus it would rely on the whole global internet network to be available.

    A mesh network on the other hand is inherently local. It might have nodes that connect to the classical internet through ISPs passthroughs at some times, it might be totally isolated at other times. Then suddenly data might become inaccessible (even with the 4x redundancy).

    The goal of mesh networks is for ISPs to eventually be abandoned altogether, all content would be eventually stored on servers that would be reachable (also) through the mesh infrastructure. This means data that is more local would be more accessible than, say, data in China. But that’s also how nature works if you think about it. (Or imagine an interplanetary internet, it will be limited by lightspeed, so realtime access will be limited to one’s own planet.)

    To sum up, I believe in a mesh network there would be too high churn that MaidSafe isn’t designed for (and neither would be blockchain-based crypto-systems for that matter).

    1. Clients on MESH are no problem. Nodes (routing nodes) will have high churn but as long as the mesh is actually connected to the Internet then all is well. A fully disconnected Mesh is an issue as it would be a smaller network. The churn from a mesh will be fine, addressing on mesh is interesting though and there may be some side projects to look at SDR etc. with XOR addresses at the network layer instead of as an overly, but that’s a long way off. With the advances in mesh though there is debate to be had as to whether satellite and other off earth solutions may even leap frog the need for mesh or push mesh into more local faster communications networks, perhaps. It will be an interesting foray when we get time to more fully engage with some of the mesh projects.

    2. It is true that MaidSafe is a global network and therefore needs peers with a reliable connection to it but that is not to say there cannot be updates to the core or implemented as plugins to allow for nodes to accept chunks via a local storage routing protocol (perhaps this is where a collaboration with IPFS would be very useful!).
      The toughest problem will probably be maintaining security in local networks or preventing insecurities being introduced in the interaction between data stored in SAFE and a local network.

  2. I have a simple question. When will MaidSafe be ready for common use?

    Thanks for your answers.

    Ruediger

    Profile on LinkedIn

    1. It’s hard to be very specific with timescales, we tend to work on tasks. By the time we reach DB2 on our roadmap (http://maidsafe.net/roadmap) it will be possible for end users to install a vault and client software and connect to the network and run some early applications.

Comments are closed.