Author: Ryan Butler 2015-12-09 04:44:09
Published on: 2015-12-09T04:44:09+00:00
In a Bitcoin development mailing list, Gavin Andresen suggested creating a 1-megabyte transaction with all of its inputs spending segwit-spending SIGHASH_ALL inputs to fit more data into one megabyte. Gregory Maxwell responded by stating that witness size comes out of the 1MB at a factor of 0.25, making it impossible to make a block with signatures that have the full 1MB of data under the sighash while also having signatures externally. He also believes that nothing he has advocated creates significant technical debt and combining functional changes with structural tidying is bad engineering practice. Martin Fowler's belief in refactoring code in conjunction with adding new features was also cited as a counterargument. The discussion ended with a disagreement about how delays were caused by a strong desire for inclusivity and hearing out all ideas, while Gavin Andresen pointed out that worries about theoretical future problems have prevented the consensus on simple solutions.
Updated on: 2023-06-11T01:44:04.284235+00:00