Oddbean new post about | logout
  @LynAlden asked a great question yesterday: why no temporary softforks?

Here's Base58's niftynei on the difficulty with adding opcodes to bitcoin
https://m.primal.net/HTwk.mp4 
 one minor correction: the video says you can't tell what opcodes are used in unspent scripts.

this is true. you *can* however tell what version of scripts bitcoin is locked up to.

so theoretically you could remove support for a version of script once all bitcoin locked up to that script version had been spent.

there are currently 3 versions of scripts in bitcoin; none of them have ever been deprecated and all of them currently have funds locked up to them. 
 Don't core devs test out new versions for a few months before releasing it to the public? 
 New versions are tested on Testnet, signets, or locally with regtest Bitcoin. But until it's live on mainnet it's effectively just like testing it on a completely separate fork 
 Well, the rule could take into account the block number the transaction was confirmed in. However, it still has many problems:

- the handling code cannot be removed
- if someone doesn't get their tx into the chain fast enough, they lose
- incentivizes miners to reorg

It could have some more, that's just what I came up with now. 
 Would you mind giving us an example of some things that this would work for? 
 You spark another thing than my LN wallet ! Lisa you are a bitcoin rockstar , take care of you. One day I'll take base58 courses.  
 Thanks for the detail! 
 nostr:nevent1qqsxnv4xut7fg58ysq0hc2zlza0dank07vnlpzzaww0meg2kt9fkhxsppamhxue69uhkummnw3ezumt0d5pzqekzj2lwpf5uygdwh5hj3rl7rma2x93jzutpe9fexdlzkrw42ym2qvzqqqqqqyamnvrz