pull down to refresh
0 sats \ 3 replies \ @ZezzebbulTheMysterious 16 Jan \ parent \ on: Why Bitcoin Knots instead of Bitcoin Core ? 🧵 bitcoin
I fundamentally disagree and think that Satoshis words are timeless here. Knots does not add anything here. If “core” aka Bitcoin does something else, knots is not in a position to enforce consensus. It’s a minority fork, and we have dealt with these before.
Knots is nothing but a hard fork risk created by a sick man.
This is all the part of the evolution of any memetic structure where false prophets lead the unwise into poverty.
I think you're missing the point though. Knots does add something here. At minimum a faster IBD then core, and filters to exclude spam transactions from abusing my nodes mempool resources.
It's not a fork in the way you describe. It's a within consensus fork, still part of the consensus majority.
I don't believe luke is a sick man. Even if he was, that should be beside the point. Knots is open source bitcoin software. you can verify it for yourself.
As for being a hardfork risk. That risk is vastly overstated and could actually be a good thing. If we find a hardfork bug through some oddity, when the design spec is to be within consensus, then we've discovered a bug worth discovering, and can patch it.
reply
You are right in that I cannot stop you running it, nor do I really care to stop you.
I can just warn of the risks. If you have been around a while, you know the history for that developer. I lost trust when he messed with the gentoo package.
Bitcoin does not have a formal spec. It has an implementation, and someone elses interpretation of that implementation called Knots. There is no 'within design spec but wrong implementation in Core'. There is just the core implementation.
If there is a bug in Knots, the rest of the network will reject its blocks and it will hard fork. Its not an 'opportunity to fix core'.
reply
well it could be, since such a bug would reveal a consensus rule previously unknown, that would then likely need to be accounted for by core with guard rails. theres been examples of this in the past.
its interesting you bring up the gentoo issue too. someone else said the same thing to me recently. https://www.reddit.com/r/Bitcoin/comments/2iuf4s/lukejrs_public_apology_for_poor_gentoo_packaging/
from my reading, this was an unintentional fuck up, that had no effect on anyone? that he maturely owned up to.
nullc even states in that thread "OTOH, while I strongly disagreed with the 'anti-spam' approach (and had long been nagging luke to do more pure behavioral matching on the abusive transaction behavior (censor-magnet, and UTXO bloating)), I am a little sad to see many people criticising a different distribution of Bitcoin Core not just for its own policy decisions but for being different at all.
There is no mandatory official Bitcoin, and when it comes to node policy-- (not consensus rules, of course)-- diversity is valuable, and people should have the ability to control what their computers are doing, how their resources are being spent, etc."
reply