Author: Timo Hanke 2017-04-06 17:43:56
Published on: 2017-04-06T17:43:56+00:00
In a discussion on the bitcoin-dev mailing list, Luv Khemani asked why ASICBOOST was considered an attack on the network. In response, Bryan Bishop referred to a Reddit post by Gregory Maxwell, arguing that it is "an attack" since it significantly reduces the amount of work needed to prove a certain amount of work has been done in Bitcoin's Proof of Work. However, Timo argued that making individual operations cheaper does not qualify as an attack. He explained that attacks usually refer to something learned from prior evaluations that shouldn't be possible to be learned, whereas ASICBOOST involves making the individual operations cheaper by a constant of ~20%, rather than changing the number of operations. Therefore, Timo said it doesn't qualify as an attack in the sense that Bishop meant. The proposal posted by Bishop does not prevent the technique, only its covert form, and is narrowly targeted at interference with upgrades.
Updated on: 2023-06-11T23:43:35.235252+00:00