Author: Mike Hearn 2015-02-20 16:54:06
Published on: 2015-02-20T16:54:06+00:00
In a conversation between two individuals, Mike had posted a detailed response on why improving privacy in the blockchain becomes complex and bandwidth inefficient. Although it is possible to improve privacy while still preserving high performance, this requires complicated programming work. The original intent of Bloom filtering was to allow for both privacy and compacting queries for sets of addresses. However, being sufficiently smart in this regard has never been a priority, as users tend to complain about other issues. The problem with downloading block bloom filters for complete privacy is that every block will match at least a few transactions, making it necessary to download every block. Eventually, wallets must stop doing linear scans of the entire blockchain to find transaction data because this approach will become too slow, especially with larger blocks. Wallets currently scan the chain to show users time-ordered lists of transactions, but by combining Subspace and BIP70, it is possible to replace the payment list UI with actual proper metadata that isn't extracted from the blockchain. At that point, non-scanning architectures become more deployable.
Updated on: 2023-06-09T17:30:46.979469+00:00