This paper by Dorit Ron and Adi Shamir from the Weizmann Institute of Science provides a comprehensive quantitative analysis of the Bitcoin transaction graph, which is publicly accessible. The authors downloaded and analyzed the full history of Bitcoin transactions up to May 13, 2012, to answer various questions about user behavior, transaction patterns, and the distribution of balances and transaction sizes. Key findings include: 1. **User Behavior**: Most bitcoins are not in circulation but are stored in "saving accounts" or dormant addresses. The majority of entities and addresses have received very small amounts, with only a few receiving large sums. 2. **Transaction Patterns**: Many small transactions are common, but there are also large transactions, with almost all large transactions (over 50,000 bitcoins) tracing back to a single transaction in November 2010. 3. **Graph Structure**: The transaction graph contains complex structures such as long chains, fork-merge patterns, and self-loops, which may be used to hide the true nature of large transactions. 4. **Most Active Entities**: The paper identifies the most active entities in terms of incoming bitcoins and transaction volume, with Mt.Gox, Deepbit, and Instawallet being notable examples. The authors conclude that while the Bitcoin system is widely used, it remains challenging to obtain accurate practical information about its usage due to the complexity and anonymity of its transaction graph.This paper by Dorit Ron and Adi Shamir from the Weizmann Institute of Science provides a comprehensive quantitative analysis of the Bitcoin transaction graph, which is publicly accessible. The authors downloaded and analyzed the full history of Bitcoin transactions up to May 13, 2012, to answer various questions about user behavior, transaction patterns, and the distribution of balances and transaction sizes. Key findings include: 1. **User Behavior**: Most bitcoins are not in circulation but are stored in "saving accounts" or dormant addresses. The majority of entities and addresses have received very small amounts, with only a few receiving large sums. 2. **Transaction Patterns**: Many small transactions are common, but there are also large transactions, with almost all large transactions (over 50,000 bitcoins) tracing back to a single transaction in November 2010. 3. **Graph Structure**: The transaction graph contains complex structures such as long chains, fork-merge patterns, and self-loops, which may be used to hide the true nature of large transactions. 4. **Most Active Entities**: The paper identifies the most active entities in terms of incoming bitcoins and transaction volume, with Mt.Gox, Deepbit, and Instawallet being notable examples. The authors conclude that while the Bitcoin system is widely used, it remains challenging to obtain accurate practical information about its usage due to the complexity and anonymity of its transaction graph.