pull down to refresh
144 sats \ 5 replies \ @optimism 2h \ parent \ on: Supertestnet's gotcha: BIP16 also confiscated coins bitcoin
Have to call this out because I think that this is false, sorry. If you weren't an economic participant at the time of BIP16, there is nothing you can do now because consensus enforces it. Any coin received after BIP16 activation will be reversed if you want to challenge BIP16, so you have no choice really; if you allow these to be spent, you're the one hard forking. These coins are lost. Did anyone complain?
you make a good point. I want to agree, but I'm nervous because it feels like an easy out.
Why isn't it the case that we ought to put up with the nightmare that is a hard fork in order to fix this problem?
reply
reply
This is Vojtěch's response, as well. It looks like him and Super have continued to discuss.
As of an hour ago, this is where Vojtěch was:
I'm most definitely okay with a soft fork like P2SH confiscating a dollar of value as collateral damage, I just don't think confiscation itself should be the goal.
I think this is reasonable, but I want to take it further: I want to have the strong stance embodied by nvk at the top of my OP: utxos are sacred, don't fucking take them from anyone.
reply
I'm most definitely okay with a soft fork like P2SH confiscating a dollar of value as collateral damage, I just don't think confiscation itself should be the goal.
I'm not fully okay with this, much like you, but I am sensitive to the point that spending to explicitly unimplemented opcodes to do anyonecanspend, is an appeal to undefined behavior being static. That means no more soft forks ever. So to protect Bitcoin, in case we are confronted with something that absolutely requires a softfork, if you want to donate coin, spend to
OP_TRUE. Any undefined behavior is undefined.If inscriptions were using undefined opcodes, witness programs or other things to put data on the chain, then I'd be okay with dropping support for these. But since they are not, it's just not feasible to prevent past usage.
reply
reply