What data is available in an analysis of the Bitcoin Network?
There’s no doubt that Bitcoin makes a lot of information explicitly public. At its very core is the Public Ledger, which is the entire history of all transactions leading back to the original creation of each Bitcoin. This has to be public, otherwise there would be no method to prove ownership.
It’s this public ledger that makes the analysis of bitcoins different from the analysis of other networks such as Facebook and Twitter. With many social networks there is a method to keep some data private. Analysis has to be carried out with missing or incomplete network data. Not so with Bitcoin.
With social networks there’s also a legal question mark over whether analyzing data infringes the users’ privacy rights. Companies that include network analysis of their social network data in their Terms and Conditions have been challenged in the past and have occasionally needed to rewrite them to take into account the user sentiment. Again, this doesn’t apply to Bitcoin; the public ledger is held in common, and available to all.
The public ledger holds a wealth of valuable information. It includes the date and time of a transaction, and with information about time, comes data about the flow of the currency. It includes the value of the transaction and the bitcoins received in previous transactions and combined to pay for it. It includes context - the fact that it moved from one person to another, and that those people in turn have interactions with others. There’s more than enough data here to attempt an analysis.
How Feasible is an Analysis of the Bitcoin Network?
So how feasible is a network analysis of this type? Well it’s been done before. The community currency Tomamae-cho which was introduced into the Hokkaido Prefecture in Japan in 2004 included the space on the back of each certificate for the recipient to record the transaction date, name and address of the person receiving it. This information was successfully used to form a network of currency flow for analysis.
Where’s George? in the US was a crowd sourced research program to track movement of US dollar bills. It proved very successful and was used to approximate the movement of whole populations.
In both these cases information was complete by design. However for anyone analyzing the Bitcoin network much of the information is still secret. But analysis can still be done.
If there’s information available on how data moves around a network, hiding inside of that network is extraordinary difficult. Every time you interact with the network you’re leaving evidence that can be picked up and analysed. That evidence can be used to infer social ties for example. If a group of people in a network talks to each other a lot the fact that they’re talking can be seen even if you don’t know what they’re saying. That the group talks at certain times of the month, or prior to specific events, can also provide context. These observations then gives you information about that nature of that group and it’s cohesion.
So network analysis of this sort has been done before. Indeed, it’s common practice for Intelligence Services and Marketing Firms alike.