0 sats \ 8 replies \ @Murch 23 Jan \ parent \ on: Bitcoin Filters Work By Default, and That's a Good Thing! bitcoin
There are people that want to send inscriptions and are paying dozens of bitcoins in fees to people that mine them. Given their incentives, do you think that at least one person from these two camps might manage to come up with the obvious idea of simply not turning on the filters on their own nodes?
It's called scamming or money laundering. It could even be an attack similar to empty blocks, just flooding the chain with spam/useless tx.
Btw, what in the whitepaper and network design makes you think this is a valid use case? And before you start with muh free market: Bitcoin is not a free market, never has been.
reply
I don't think they're useful or interesting. I just think that the people who participate in inscriptions have incentive to continue relaying them and the miners have financial incentive to include them and observe that inscribers let it cost themselves fees worth paying for a significant proportion of the blockspace.
reply
And for individual node runners who don't want their nodes to participate in this activity as much as possible within consensus, to reduce local compute resources and moral hazard? they're just told/supposed to use blocksonly=1?
Do you care about these users?
Your silence is telling.
reply
I don't think it makes a significant difference to forward the transactions with the blocks or when they're relaying unconfirmed. The node might save a few inv messages and forwarding a tx once or twice. If that bandwidth savings is of concern,
blocksonly would be more effective at reducing bandwidth use.I don't see any non-marginal benefits to be achieved by filtering unconfirmed inscription txs beyond virtue signaling, and I don't think working on enabling virtue signaling is a good use of my time.
reply
No one's making you do anything. You are though, volunteering your time to voice your opposition to what you view as virtue signalling. While parroting the same tired false goal post I make a point of calling out in my article. baffling.
I view this anti-spam activity as reducing moral hazard and overall risk, while preparing for the worst case scenario.
RAM, CPU, Bandwidth and energy costs are all of concern here.
If normal people can't run a node because of this activity in 14 years, we have a problem.
reply
I don't see how RAM or CPU are negatively affected. I don't see how it costs more energy either. Inscriptions have fewer signature operations per data than other transactions. And nodes that filter their mempool still get the whole block, only marginally reducing bandwidth use.
So, I don't see why filtering is helpful for any of those four concerns.