Author: Me 2015-07-15 15:49:13
Published on: 2015-07-15T15:49:13+00:00
In a discussion on Bitcoin-dev mailing list, Simon shared his tests and was asked by someone to share the transaction hashes. Another person reminded him that what he did can be classified as fraud and advised him to keep his record clean by sending them money post-mortem. Peter Todd, who has been critical of Blockcypher on Twitter, was asked by someone to share evidence of their Sybil attack on the network. Peter responded to this by stating that the double-spent transactions had near 100% propagation on blockchain.info, but unfortunately, the relevant data has been purged. He explained that Shapeshift.io depends on Blockcypher's "confidence factor" model which is another one of those sybil attacking network monitoring things. However, those measurements don't actually tell you what you need to know because miners frequently use customized Bitcoin Core codebases that don't follow normal policies.Shapeshift confirmed the attack and said they're going to "improve" their system. It'll be interesting to see what that actually entails. Blockcypher's "confidence factor" model works by estimating transaction confirmation probability by looking at the percentage of nodes a transaction has propagated to. The discussion also included a request for Tom Harding to include the TXid's to show that he actually performed the valuable service he demonstrated. Furthermore, it was advised that anyone relying on an unconfirmed transaction must pay a good fee and ensure that it is highly relayable/minable.
Updated on: 2023-06-10T02:45:29.232956+00:00