BIP Proposal: Compact Client Side Filtering for Light Clients



Summary:

On June 1, 2017, Olaoluwa Osuntokun proposed a new light client BIP on bitcoin-dev along with Alex Akselrod. The proposal is available on Github. Another light client type was also suggested by Tomas van der Wansem of bitcrust that has different security characteristics from the proposed BIP. This alternative client would send each client the TXID (uncompressed), spent outpoints (with TXIDs compressed), and pubkey hash (compressed to reasonable amount of false positives) for each transaction in each block, making it about 2-2.5 times larger per block than the proposed BIP. However, this alternative client could verify the TXIDs against the merkle root with a much stronger (PoW) guarantee compared to the guarantee based on the assumption of peers being distinct that the proposed BIP seems to make. The alternative client removes privacy and processing issues from server-side filtering and retrieval of all txids in each block can serve as a basis of fraud proofs and (disprovable) fraud hints without resorting to full block downloads. The author does not completely understand the benefit of making the outpoints and pubkey hashes (weakly) verifiable since they only serve as notifications and therefore do not seem to introduce an attack vector. Omitting data is always possible, so receiving data is a prerequisite for verification, not an assumption that can be made. The author questioned if it is worth compressing away the information to verify the merkle root.


Updated on: 2023-06-12T01:31:23.227718+00:00