Segregated witness softfork with moderate adoption has very small block size effect [combined summary]



Individual post summaries: Click here to read the original discussion on the bitcoin-dev mailing list

Published on: 2015-12-20T03:37:26+00:00


Summary:

In a discussion on the Bitcoin-dev mailing list, Peter Todd highlights how the implementation of Pay-to-Script-Hash (P2SH) in wallets can be seen as a vote for an increased block size. This is because P2SH leads to smaller transactions, similar to Segregated Witness (SW). However, Todd acknowledges that SW's client-side adoption may not be as effective as a direct proof-of-stake vote and can be viewed as coercive due to lower fees being paid.Another participant, Douglas Roark, responds to concerns about the adoption rate of SW by stating that P2SH and SegWit uptake should not be conflated. Roark believes that wallet developers will recognize the value of SegWit and enable it by default, leading to widespread adoption. While optimistic about aggressive timeframes, Roark expects more conservative adoption rates. He also suggests that a 40% adoption rate within a year of deployment is possible.Santino Napolitano disagrees with the notion that client-side adoption of SW indicates a preference for larger blocks. They propose alternative methods such as attaching OP_RETURN outputs or providing hand-signed letters to explicitly state business preferences. Napolitano raises concerns about English-language measures being misleading and the significant economic activity in non-Western countries. Centralized forums like Twitter and Reddit are also seen as easily censored and manipulated. Peter Todd adds an anecdote about a Bitcoin-related product that had majority sales in non-English-speaking countries, emphasizing the need for inclusivity in consensus-building processes.In response to jl2012's calculations on the effect of a SW softfork on the total block size, Peter Todd argues that client-side adoption of SW demonstrates the ecosystem's acceptance of a blocksize increase. Another user counters by stating that SW adoption merely indicates implementers seeing greater value in it than the cost of implementation, not necessarily a preference for larger blocks. They suggest alternative ways for businesses to express their desires. Todd acknowledges that SW adoption is not as effective as a direct proof-of-stake vote but finds it to be an interesting side-effect.Another user on the mailing list expresses doubt about SW being the sole short-term solution for scalability. They believe fraud-proof systems are crucial for network security, suggesting the addition of fraud proofs to blocks to make features like committed UTXO sets less risky. While recognizing the value of solving transaction malleability, their main concern is buying time for continued usage growth while working towards eliminating the maximum block size as a consensus rule.On December 19, 2015, jl2012 calculates the effect of a SW softfork on the actual total block size. Client-side adoption is necessary for an actual blocksize increase, which can be seen as a positive sign that the ecosystem has embraced larger blocks. Although this method is not as ideal as a direct proof-of-stake vote and has some coercive elements due to lower fees, it is an intriguing side-effect.The author of the text conducts calculations on the effect of a SW softfork on the total block size. They define various terms and propose a discount factor for witness size. They present a table showing the maximum total block size based on different variables. However, they note that P2SH has only been in use for 3.5 years and accounts for only 10% of Bitcoin storage, making a 40% adoption rate for SegWit within a year over-optimistic unless transaction fees significantly increase. The author also questions whether SW should be the exclusive short-term solution for scalability.Overall, the discussions on the Bitcoin-dev mailing list touch upon the implementation and adoption rates of Pay-to-Script-Hash (P2SH) and Segregated Witness (SW), the relationship between client-side adoption and block size preferences, alternative methods for expressing desires in the Bitcoin ecosystem, concerns about scalability solutions, and calculations on the impact of SW on the total block size.


Updated on: 2023-08-01T17:13:01.547714+00:00