I’m not sure it’s that simple
Knots is bitcoin core with some filters that don’t really work
Also if people run different versions of Bitcoin, it screws the lightning network
I’m not here to defend Core 30 but I don’t think we have a choice
🤷♂️
I’m not sure it’s that simple
Knots is bitcoin core with some filters that don’t really work
Also if people run different versions of Bitcoin, it screws the lightning network
I’m not here to defend Core 30 but I don’t think we have a choice
🤷♂️
Depends on what "work" means. Not having random unencrypted data in my mempool is already working. Not relaying that data is already working.
If it's returned to me in a confirmed state by miners, then the responsibility of that confirmation lies with them. Law enforcement / society will target them exclusively. And it will exist in an encrypted state in my node, which is less troublesome.
“Work” being the confirmed state by miners
Your mempool does not reflect the fee environment correctly
It probably does now because fees couldn’t be lower but you’re not going to be using that mempool to base your fees on whenever things pick up again
Knots is not the solution to this problem imo
The easiest choise is Core 29, if you're not onboard with knots, and don't have a use case for a larger op return.
Since I understand datacarriersize is essentially depricated now.
I agree
However
You will not able to host a self-custodial lightning node with people running v30
I don’t anyways but it’s a major issue for lightning
To be honest I haven't really heard this said before. Maybe cause I'm not following intently.
I suppose it's only a risk. That transactions are not caught by the lightning instance because bitcoind filtered it out.
My mempool was often full when the network was busy, and I felt comfortable enough relying on the CSV. I guess it would be the same now, that I would only know after confirmation.
Lightning is also not Bitcoin so if it has to go for a while, so be it
This is as far as my knowledge has taken me
Lots of bad and hyperbolic actors
Different implementations are non-partisan and perfectly fine as long as consensus rules match. Like you said, Knots is basically Core with extra relay options, and for Core/Knots team to deviate into hard-fork territory would be an insane move, and bad for Lightning.
The good thing about Bitcoin is we don’t have to speculate on compromised teams or people; the code speaks for itself, and everyone is free to choose which open-source software to run.
Knots is already bad for lightning, not when it hard forks. Same for older versions of Core
I think it matters more than you’re stating. I’m not going to pretend I understand the ins and outs
I also don’t think it’s a pressing existential issue now but it could become one in the future
Yeah, that’s true with the relay policy mismatches.
You’re right, Core 30 will roll out no matter what with its own policies too, and hopefully it doesn’t escalate further