Author: Matt Corallo 2019-07-22 05:01:58
Published on: 2019-07-22T05:01:58+00:00
A bitcoin developer named Andreas Schildbach expressed concern over a proposal to remove the NODE_BLOOM feature from the network, stating that more than 10 million wallets rely on it. He believes that pulling the rug under that many wallets is a disastrous idea for Bitcoin's adoption. The developer also questioned the reasoning behind the decision, claiming that DoS attacks using high hash counts to fill a node's CPU usage are possible and that there is no alternative to NODE_BLOOM. In response, another developer, Matt Corallo, clarified that he does not anticipate that most people will change their defaults, but believes that NODE_BLOOM publicly accessible nodes may not be around for at least three or four years due to general upgrade cycles on the network. While the filter-generation code for BIP 157/158 has been in Bitcoin Core for some time, the P2P serving side of things appears to have lost any champions working on it. Andreas Schildbach suggested improving the current filtering for segwit by testing the scripts themselves and each scriptPubKey spent by any input against the filter, which would fix the main privacy issue with server-side filtering where wallets have to add two items per address to the filter. He strongly recommends postponing the proposed change until an alternative exists and developers have enough time to implement, test, and roll out the new solution.
Updated on: 2023-05-20T20:41:15.116279+00:00