Nostr Web Client

I think it’s intellectually consistent to say segwit and taproot were mistakes. I don’t agree but I respect the position.

There’s near universal agreement from devs that there’s no way to prevent data we don’t like w/o prevention “useful” use cases, which is another way of saying that actually the protocol needs to be value neutral in terms of content of data. The way that we filter for usefulness is the market, block size limits, and limited supply of currency all of which are reasons why bitcoin is probably a better platform to have experiments like NFTs than alternatives. The reason is if they have no value then as fees go up but people are less willing to pay them their use will go down. The same could be true for more complex spending transactions of any kind. If my business doesn’t want to pay the fees for a more complex locking script that might provide more security at the cost of block space, then we’ll opt for simple multisig.

clr 2y ago

I agree there is no way to completely prevent data we don't like, but segwit and taproot have enabled the inscriptions' abuse like hand in glove. In two ways:

1) inscriptions benefit from the segwit fee discount so they pay less fees for the same amount of bytes than other types of transactions and

2) having a limited OP_RETURN size and a limited amount of data per transaction on segwit means that inscriptions would have to be split among several transactions, therefore increasing the fees they have to pay for the same amount of data.

Without the segwit discount and without taproot the same inscriptions would be paying several times the fees they are paying right now.

I don't care that there might be "near universal agreement from devs", like there was near universal agreement from "experts" during the "pandemic". That is an appeal to authority. Devs aren't gods and they don't represent bitcoiners.

Reply to this note

Please Login to reply.

Discussion

buck 2y ago

Limiting data on witnesses makes no sense and would break many use cases of bitcoin including those that are “purely monetary”. I’m not appealing to authority as an opt out. I’m making the point that

(a) this is not as simple as “just get rid of things I don’t like” and

(b) you would need someone to implement the type of limit you describe and if everyone that could says that’s a bad/dangerous idea, that’s a problem.

The difference about the pandemic is that experts that disagreed were silenced so it wasn’t universal agreement.

Lift the cap on OP_RETURN and no more witness discount. If it’s not just about the discount then you’re really just asking to censor data you don’t like.

buck 2y ago

By the way, do you know why there’s a witness discount?

One reason was to encourage migration to segwit. That was desirable. It came with tradeoffs but probably still worth it.

Second reason is because the witness doesn’t need to be persisted by all nodes. So this is actually better for those that don’t want to persist inscription data on their node.

clr 2y ago

I wonder what would happen if the nodes stopped saving and verifying the segwit transaction signatures. Can those UTXOs be spent by anyone? I still haven’t found a convincing answer to that question. But that’s another story.

buck 2y ago

Utxos haven’t been spent yet so they haven’t been signed. You can verify by verifying all of the blocks.

clr 2y ago

I would be pleased if the segwit discount was removed. About the OP_RETURN cap, I’d rather keep it unless there is a justification for increasing (not removing) it. You mentioned the “expresiveness”. I definitely don’t want to censor data, but I believe that transactions should be as concise as possible rather than expressive/verbose. The bitcoin blockchain is, first and foremost, an accounting ledger, not a blank canvas to paint or write novels. I am OK with short strings of text, such as the Times’ headline in the genesis block or the miners writing their name on the blocks they mine. It’s not about the contents, but about the length/size.

buck 2y ago

At some point what is concise will become a subjective value. The way we make that objective is by scarce block size and the free market. Costs encourage users to be as concise as possible.

clr 2y ago

Yes and no. The protocol is opinionated: whether to keep, alter or remove OP_RETURN, the segwit discount, and so on alter the conditions of the “free market”, i.e. the playing field that the protocol enables. Even the block size limit is opinionated. For example, with an unlimited OP_RETURN, it would be cheaper to make an inscription using it, because you wouldn’t have to split it among several transactions.

buck 2y ago

Great point. What you’re describing is a change to the protocol, which by definition would not be how it works today. But yes, removing OP_RETURN limits (which I believe is just policy anyway) and the segwit discount would make the market more free. Discriminating against data (a la current segwit discount and OP_RETURN limits does) is would not.