Vojtěch Strnad

Inscriptions were possible before Taproot | Make crypto cryptography again

[Vojtěch Strnad @vostrnad]

Last week I made a special Bitcoin transaction. Today I'll tell you all about it: https://t.co/UVxXFeKrcI

[mononaut @mononautical]

what did they mean by this?
https://t.co/4S8jM2grem

[Vojtěch Strnad @vostrnad]

@w_s_bitcoin @murchandamus @brian_trollz It's beautiful really. Both usage and non-usage would prove to them that removing the filter was a mistake.

[Vojtěch Strnad @vostrnad]

@SteveSimple @kcalvinalvinn He definitely didn't argue back then that the limits should be preserved in consensus, and later wanted to correct the "oversight" in policy: https://t.co/7EUKTbeFXv

Highly recommend reading the comments in that PR, lots of context for your question.

[Vojtěch Strnad @vostrnad]

@SohumJShah @michaelinioluwa @cc33_rl @_lilaurora So you're suggesting two typesetters should get wildly different compensation even though they did essentially the same work, just because one of them was lucky to be assigned to a book that ended up being a success. I don't think that's fairer than the current system.

[Vojtěch Strnad @vostrnad]

@Fundamentals21m Now ask it to find a transaction that uses all possible SIGHASH flags.

[Vojtěch Strnad @vostrnad]

@w_s_bitcoin https://t.co/Bo585k8b0U, which monitors for new signet blocks, implements exactly this approach:

https://t.co/m5EaIvKIry

[Vojtěch Strnad @vostrnad]

@w_s_bitcoin I have – the solution is to set up cascade deletes in postgres that ensure deleting a block also deletes all its transactions/outputs/etc. and then just delete stale blocks when a new block is detected and its parent isn't what you thought was previously the best block.

[Vojtěch Strnad @vostrnad]

@cryptoquick @w_s_bitcoin @projecteleven Sure, the one way to not leak any public keys during a spend is to not have any in the script in the first place 😂

[Vojtěch Strnad @vostrnad]

@w_s_bitcoin @cryptoquick @projecteleven Yeah that's what I thought, whatever conditions exist in your script, once you spend from an address you show the world how they can be satisfied, and all that's protecting you is that the signatures can't be reused.

[Vojtěch Strnad @vostrnad]

@w_s_bitcoin @cryptoquick @projecteleven AFAIK, there is no way to spend from an address without revealing enough public keys to allow a quantum adversary to spend from it as well. Mind giving an example?

[Vojtěch Strnad @vostrnad]

@SteveSimple You can do a 4 MB inscription using SegWit v0 if you split the data across multiple inputs. Ordinals/inscriptions could have just as easily become a thing before Taproot, it's merely a coincidence that they didn't.

[Vojtěch Strnad @vostrnad]

@imikushin @rodarmor This doesn't work because the internal key is tweaked with a hash that also commits to the internal key. We already have a paper with a security proof for the quantum security of Taproot commitments: https://t.co/wyn9XjodLX

[Vojtěch Strnad @vostrnad]

@MagicalTux Sorry, not happening. Bitcoin's brutal immutability may be painful at times, but in the long term it's its greatest strength and what really distinguishes it from other projects like Ethereum (see DAO hack). We cannot afford to give it up.

[Vojtěch Strnad @vostrnad]

@DudeJLebowski BIP-110 doesn't ruin any data embedding "use cases". At most, it forces some of them to slightly tweak the format they use, at a one-time cost of maybe one or two developer hours.

[Vojtěch Strnad @vostrnad]

@LukeDashjr @phyrooo @satofishi That seems like an insane position to defend. But okay, how do you fix a double spend once the coins have mixed with the wider economy? Who do you take them away from?

[Vojtěch Strnad @vostrnad]

@LukeDashjr @phyrooo @satofishi Just to be clear, you're saying that the ability to add arbitrary data to a transaction is a "more severe bug" than being able to spend a coin more than once?

[Vojtěch Strnad @vostrnad]

@punspotter @BTCfootsoldier This particular oversight has already been patched. It's still a fact that BIP-110 doesn't stop spam, only some very specific spam formats.

[Vojtěch Strnad @vostrnad]

@Du_Bi_Bo @stephanlivera @lukedewolf @BryanBSolstin Let me say that again: you can avoid mining spam yourself all you want, but once someone mines a block with spam in it, you'll have to choose which chain to follow, because BIP-110 nodes will fork away at that point.

[Vojtěch Strnad @vostrnad]

@lukedewolf @stephanlivera @BryanBSolstin You can only be "compatible with 110" by mining on top of the BIP-110 chain. If it's the shorter one, you're mining what could end up being a worthless minority fork. Not what I'd call "no risk".

[Vojtěch Strnad @vostrnad]

@lukedewolf @stephanlivera @BryanBSolstin It's not just about what transactions you mine yourself, but also what blocks you choose to mine on top of. If the non-RDTS chain has more blocks and you build on top of it, it doesn't matter if you yourself don't include spam.

[Vojtěch Strnad @vostrnad]

@SteveSimple I meant I got blocked by the first person in the thread, not you.

"Pruneable" can mean different things, but mostly means it doesn't end up in the UTXO set. In this case witness data and OP_RETURN are both pruneable, and >10 kB of contiguous unpruneable data was never possible.

[Vojtěch Strnad @vostrnad]

Replying here because OP blocked me to not have his echo chamber disrupted:

The 10 kB script size limit only applies to executed scripts, OP_RETURN outputs can be up to 1 MB. It was also not lifted in SegWit. And, executed scripts don't allow large contiguous data anyway.

[Simple Steve 🌌 @SteveSimple]

@vostrnad @adam3us @byronhambly @BTCtoOblivion @MartyBent @ODELLXYZ @LukeDashjr @GrassFedBitcoin How did one get around the 10kb max script size consensus rule before Segwit