Author: Jonas Schnelli 2016-06-28 12:13:27
Published on: 2016-06-28T12:13:27+00:00
In this email exchange, the discussion centers around the use of Bloom filter features in the Bitcoin network's peer-to-peer (P2P) protocol. The writer expresses concerns about the security implications of applying identity to the P2P protocol, rather than keeping it in a client-server model. The responder argues that the Bloom filter SPV use case is not purely client-server and could involve trusted nodes with shared identities/keys. However, private P2P extensions would need to be discussed before implementation. The writer also argues that the desire to secure against the weaknesses of Bloom filters should not be generalized to the P2P network, as it may actually weaken the protocol's security. They suggest moving the Bloom filter features to a client-server protocol instead. The responder disagrees, saying that they do not see how the proposed BIP151 would weaken the security of the P2P network and asks for specific concerns.Lastly, the writer notes that the BIP does not make a case for other scenarios or address the significant problems associated with key distribution in an identity system. The responder clarifies that BIP151 does not rely on identities and only uses ephemeral keys, but agrees that an authentication/identity management system needs to be described in another BIP to prevent MITM attacks. They suggest focusing on the "pure encryption part" of BIP151 to avoid scope creeping and overspecification.
Updated on: 2023-06-11T18:56:13.660973+00:00