Avatar
goatmeal
43fabde62ffea1aa0ddae7c0ac03b7017e2d864f8665784b00bbfa2f9114c06a
bitcoin maximalists are all lying scum

custodial and noncustodial lightning is a false choice. lightning was always a shitty idea. L2 does not have to use payment channels. why are your bitcoin devs so married to payment channels when it has such shitty limitations?

how do I start a brand new 100% non-custodial lightning wallet without causing an onchain transaction?

maxis always blame others and never self reflect. you used the wrong wallets, you connected to the wrong nodes, you messed up your channels... how about build a L2 that doesn't have a wrong wallet, a wrong node, or a wrong channel? a L2 that isn't for pocket change on a microblog echochamber? is that so hard? there is no second-best.

the circlejerk here is worse than reddit

another way to gatekeep fragile window licking bottom feeding reddit users is to intentionally use slurs and bigoted language. it works really well, and people who belong to "marginalized groups" are smart enough to know when it's not directed at them.

community points are cancer. reddit karma is already terrible. imagine karma convertible to money. sheesh.

>everything x tries to solve can be built on lightning

okay I'm still waiting. let's go tell all the curve users and people selling ecstasy pills that they should just stop doing what they are doing and twiddle their thumbs until one day lightning labs makes the heavens rain with cheese. this is one thing you maxipads do all the time that I just don't understand. do you really think the world is waiting on you to deliver the solutions to everyone's problems and network effects just don't exist? you will have to face the reality that you are late and everyone doing all these things have already moved on.

#[0] someone on 4chan was talking about nostr and they called you fat

very telling that drug dealers won't use lightning. you know, the people who need it to actually be private or they go to jail. I'll believe you that lightning is private when drug dealers switch to it otherwise you're lying.

in bitcoin that's called UTXO commitments. they didn't implement it yet. I don't know the ETH equivalent since they don't use the UTXO model. I do know that ETH has a lot more data that isn't settled history and it's a bigger challenge to do away with it, but there is a similar concept called state expiry that they are looking into.

it really really irritates me how bitcoin maximalists pretend that everything they have is perfect and everyone else is wrong. there are so many ways to do this and we still don't know the perfect way to do it. if somebody ever tries to sell you a solution and they won't ever discuss any of its flaws, they are trying to deceive you.

someone tried to do a second layer with channels like what lightning does. it didn't really get anywhere. then a sidechain was made and it's still really popular but people acknowledge that sidechains are not as secure. now there are optimistic rollups, but they're still not perfect and have some limitations. rollups secured by zero knowledge proofs are better. until just recently nobody had a working zero knowledge rollup that could do smart contracts.

funny thing is bitcoin devs are actually right that it's a good idea to offload some amount of transactions onto a second layer. they just don't know how to do it very well as evidenced by this link. ethereum devs have already implemented multiple ways to do this and they all work better than lightning network.

the current state of bitcoin is the result of the block size debate. people came up with the idea that if you add more than 2 mb of data to the ledger every ten minutes, it will cause all our dicks to fly off. they even invented a slur for people who don't believe this. they had to invent the rube goldberg machine known as the lightning network in order to uphold this dogma, and it took them so long to make something that looked sort of usable that many people just abandoned bitcoin entirely. a small number of people tried to fork bitcoin and were not very successful.