Michael Graaf, from the Right to Know Campaign (South Africa) is a lifelong activist in anti-apartheid, antimilitarist, green, and information liberation fields. His education background is some science, some humanities, some media studies, and some IT. He is the author of this post on peer-to-peer (P2P) name services as a potential game changer in internet governance.
Much of the first day of the African School of Internet Governance (AfriSIG2014) was taken up by exploring the regulation of “names & numbers” – domains and the IP addresses that they map to. The system now overseen by the Internet Corporation for Assigned Names & Numbers (ICANN) evolved over the years along with the internet itself, and now constitutes a large part of internet governance. One of the interesting facts we heard was that governments have been known to block Domain Name System ( DNS ) lookups for websites they considered subversive, thereby crudely censoring such sites.
During the presentation I asked the presenters what they thought of a recent development: peer-to-peer (P2P) name services. The gist of their replies was that it’s too early to say what the significance is of such projects. From the little that I knew, I felt that we should look outside the status quo which we studied so diligently, so I’m offering this piece to get the conversation going.
Background
There are several projects to provide alternative DNSs; in general they are deprecated because they create the possibility that different users clicking on the same link could be directed to different destinations; this could seriously harm the usefulness of the internet as a whole. However this general concern does not apply to all of them, as will be seen below.
Two concepts need to be understood before P2P name services can be fully explained: peer-to-peer architecture as such, and secondly blockchains.
In essence, P2P architecture removes the distinction between client computers and servers. Each machine in a P2P network has similar status as a “node” or “servent”. Early examples were file-sharing systems, where once a file had been seeded from the first node, it would be stored in multiple locations and each time a new request for it was made, it could be sourced from several of those locations simultaneously, according to the traffic load in each part of the network from moment to moment.
The concept of a blockchain is an innovation within the P2P paradigm. First introduced as part of the Bitcoin project, in essence it is a distributed open ledger/database which is constantly synchronised between all the nodes of the network. Hence it is almost impossible to make a fraudulent alteration. Although originally created as a record of payments, new uses are constantly being found and the administration of domain names is one of these, first deployed in the Namecoin project (being free/open-source software, anyone can duplicate and modify it).
How Namecoin’s service works
The domain name serving function of Namecoin relies on its cryptocurrency nature. Because a massive amount of processing power is involved in verifying and recording a transaction (e.g. the leasing of a name domain), the uniqueness of each URL is guaranteed and fraudulent or abusive activity excluded. This processing is performed on the computers of some members of the P2P network, referred to as “miners”. They are rewarded by units of the Namecoin currency, which are actually encrypted digital code. These same units can be used, among other things, to lease domain names.
A user wishing to use this service can install a special add-on in their browser, which enables them to access URLs having the “.bit” suffix by communicating with the P2P network which then returns the relevant IP address. Other methods such as proxy servers are available.
Ways to work around censorship
The main “selling point” of P2P name services is that they prevent the kind of crude censorship mentioned earlier, where a government simply compels the operators of DNS servers to remove certain websites from the lookup list. However, there are other ways of working around such censorship.
Many people may choose P2P name services because they like the do-it-yourself aspect, which leads to collective self-reliance among a community of users. Conceivably, such services could continue to function if the main structure of the internet is disrupted, especially when combined with mesh network infrastructure, which is becoming more common and easier to implement. Such resilient communications have repeatedly proven to shift the balance of power between civil society and repressive regimes. They also offer a cheap and co-operative way to make telecommunication accessible where the market has failed to do so.
In conclusion, I submit that roleplayers in internet governance should be aware of P2P technology as a potential game-changer.