They’re not preventing anyone from doing anything. But they get dangerously close to causing a chain split.
Discussion
If they validate and relay the same blocks as core nodes, how do they cause a chain split? Besidea, chain splits happen naturally, and don't last many blocks.
Wait, nostr:nprofile1qqs9qp2dql3vmuetzq6hw77eeaeej2j2ughers22wch0m2jm7c05w4gpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhszythwden5te0dehhxarj9ekxzmny9uq3zamnwvaz7tmwdaehgu3wwa5kuef0mr8vev, you think k it's is going to lead to a permanent, deliberatw altcoin chain? Who is advocating for that?
nostr:nprofile1qqs9qp2dql3vmuetzq6hw77eeaeej2j2ughers22wch0m2jm7c05w4gpz4mhxue69uhhyetvv9ujuerpd46hxtnfduhszythwden5te0dehhxarj9ekxzmny9uq3zamnwvaz7tmwdaehgu3wwa5kuef0mr8vev, would you please tell me what knots has to do with an alternative chain split? You're, I guess, a journalist? Have you gather'ed any evidence to support this concern?
Yes, look into the 2013 accidental chain split. It came from an otherwise uninteresting Berkeley DB upgrade. I think it was Luke who noticed it and reported on the IRC dev chat.
So you are concerned that there is code in knots that can cause a chain split? Can you point to the code that raises these concerns?
I’m glad you asked. This is the first giveaway: the Knots repo is 1923 commits behind Core.
You can argue it’s not essential and that older versions of Core are compatible and in consensus. Which is correct.
But it’s the 1757 commits that are ahead of Core that are concerning. Who do you think has enough time and energy to review them all? Bitcoin Core is already difficult to review and takes months to consider every update.
Knots has 1 guy who lost 200 BTC, a bunch of unreliable volunteers and an army of cheerleaders who are in a position of absolute trust.
As more divergent code gets added under what looks like the benevolent dictatorship of one guy, it will get much worse. It only takes one bad mistake to cause a split. And Ocean will be the only pool mining the new chain. Good luck with that!
https://github.com/bitcoinknots/bitcoin
From my experience of working across multiple branches, these divergences in commits have a tendency to be the same code, but merged manually or differently. I have a suspicion that these commit metrics are a faulty heuristic on which to base your fears.
A more reliable tool would be to audit the diffs across the repos to see exactly how divergent they are from each other.
I get that the fork is really old, but it isn't really a good look (as Vlad pointed out). I was a bit overwhelmed trying to audit the 37K+ 14K- diff.
It would make me feel better to see only a few commits rebased on Core's release. I suppose I could just cherry pick the important commits myself and maintain my own.
Honestly, as I'm writing this, I might just start running a pruned node and dropping all OP_RETURN data instead 🥲
Vlad didn't point out that it wasn't a good look. Like the bimbo larping as a smart person that he is, he started hyperventilating about how a bad heuristic is reason enough to panic.