seriously corrupted organisation at this point.

out of the picture means no longer having so much clout.

and i personally have wanted to build a #golang based option that has the nice wallet GUI and fast IBD. btcd is pathetic. i'm too busy with nostr dev at this point but i'm intimately familiar with the btcd codebase.

the best outcome is that bitcoin has like 5 major clients making up 80% of the userbase, and making these kinds of arbitrary changes is a lot harder to do than it was when 80+% of nodes are running one implementation. that's why they have arranged to get spam-friendly devs into position on their team. and lunatics like peter todd. that guy alone screams red flag that someone completely non-representative of bitcoin culture has been inserted.

Reply to this note

Please Login to reply.

Discussion

I agree on diversity in implementations even if only "patched core" variants would be the best outcome for Bitcoin's resilience but manipulation of minds is not harder with 5 implementations than with just one. Non-technical people really have a hard time to tell what are legit concerns and what are straw man arguments.

yeah, the patent manipulative, fallacious arguments being presented makes me sick.

a lack of diversity in a peer to peer network is a recipe for the apocalypse. i literally saw will just say "bitcoin is bitcoin core" like, bro. stop building damus and notedeck then. we only need primal 🤡

Bitcoin Core fell out of sync with itself. It's damn hard to get consensus right and I hope all miners at least will stick to core and its forks. Devs love to build stuff from scratch and it's great to learn all the details of the stack but Bitcoin consensus is just too fragile to run alternative implementations reliably. Accidental hard forks are just damn scary.

ridiculous

the protocol isn't specified properly if there is only one implementation.

you probably know about how bad NDK is, applesauce is much better and simpler. NDK is buggy, it's a black box. the specification people are the same people building it. it's not a coincidence that the specification is so vague as to be ridiculous in places, they don't care about other people building clients or client libraries or relays. strfry is enough. primal is enough. whatever.

we aren't living in a communist dictatorship. diversity is part of how you have a robust ecosystem. bitcoin core now runs less than 80% of the p2p network and likely that's going to decline even further from this point because people have realised that bitcoin core is a single point of failure and that is fatal to a p2p network. attacks that work on 80% of the network are cheaper to execute than trying to cover 5 different clients that have different vulnerabilities.

What's the TL;DR on btcd codebase?

it's not terrible but the configuration system is buggy it's hard to figure out where stuff is and its IBD is slow, or at least it used to be. and the signature verification code in it is 1/4 the speed of the one in bitcoin core (i use the bitcoin core signature algorithm in my relay).