Hacker News Reader

Comment by Yizahi | Hacker News Reader

And extremely centralized in the so-called Lightning version. Due to ridiculously absurd requirements of the Lightning layer (bidirectional channels on L1, locking funds in advance, solving NP-hard problem thousand times per second a scale etc.) everyone basically resorts to using very few centralized entities as a pseudo-bank, who issues virtual paper, IOUs, which which wallets trade on L2. Basically all negatives of the banks and all negatives of crypto-tokens combined, with no positive sides at all.

View on HN

Okay this was what I thought would happen. I looked into Lightning a few years ago and found the whole concept to be quite dense and not something that a "regular person" does. I don't have a strong opinion on the Bitcoiner vs Shitcoiner debate and have been happy to participate in BTC/BCH/XMR style chains, and the ETH/SOL/EVM-style chains as well. But I felt like Lightning ended up acting a lot like the Ethereum L2s with weaker guarantees.

It's still pretty affordable and not-hard to run your own Lightning node; The pseudo-bank hosted wallets people use (e.g Wallet of Satoshi) is purely out of convenience.

The real lesson is that most people don't care enough about the underlying risks - they care about convenience.

How much BTC do you need to run a node? And what are the failure modes if the node goes down or becomes network unreachable or something? I'm not trying to be critical, just curious myself what happens if I run a node. Would be happy for any resources you have on hand if that's too much for an HN comment.

I’m with you. It’s too complicated to run self-custodial node. I’m an engineer and can do it and yet I gave up after a year of constant baby sitting my node.

Hopefully future versions of Start9 or Umbrell make it easier. Or some hybrid solutions like Greenlight/Breez or Spark.

If I'm not mistaken, running own Lightning node would mean opening channels to every single merchant you want to trade with AND those same merchants would need to open a symmetric channel back to you with the same amount of locked funds. And then during the transaction this system would need magically solve traveling salesman problem per each transaction in the system, taking into account that after each completed transaction system state changes. Oh, and those channels are all on L1, which has 7 tps on best days.

Lightning can't work distributed, by design. It's a silly architecture. I guess the problem that the proponents of the alternative were slightly mental, sealed the fate of the BTC. It really should have increased block size, multiple times by now, and don't bother with stuff that can't work.

Wouldn't repeated block size increases lock out "normal people" from participating too? Hosting a local BTC node is already hard, and completely impractical on a mobile device.

Nope, because SSD/HDD tech has evolved in capacity while the costs have gone down.

If I'm not mistaken, you don't need to host a full node to access and operate BTC network. But hosting full nodes certainly adds resilience and speed to the system.

Point 1. Increasing node storage size would not be very expensive, people are buying NAS storage for fun, some of the more idealistic ones would certainly do that for BTC. BTC chain today is less than one terabyte. Even increasing chain size x10 times would still mean that it would fit on a single cheap consumer HDD. Increasing it x100 times would require something like a 6-8 HDD stripe, under 1-2 thousand bucks in price. People do that for memes or hosting torrents every day, all across the globe.

Point 2. Increasing chain size by a lot would mean that the amount of full nodes would inevitably drop, lets say by half, or even by 3/4. Still a lot would remain and the system would be very decentralized. While with lightning crutch the decentralized part is clogged permanently and is unusable for any serious currency replacement use, while users are forced on a broken by design centralized Lightning.

In my opinion the answer is obvious, if we want to have a truly decentralized currency, and not some casino replacement for gambling and law avoidance.

you just need to open one channel to another node (better if well connected) and all your payments will be routed through that node.

If you have multiple channels you have more routing alternatives and can choose the best one for every single payment.

So, for a normal case where multiple people want to trade with multiple other people/merchants we would need a centralized node with connections to every single one of the clients and every single one of the merchants (so a "bank"). Both clients and merchants need to open a channel on L1 (so it's goddamn slow or expensive) to that node and freeze in the channel the full amount of tokens they would wish to spend with every single peer or merchant in the future (so basically a "deposit" of all or at least a lot of the funds in a "bank"). And every merchant would need to do that too for the full amount of anticipated transactions. And then this system will work on L2 somehow, if the system will correctly calculate paths and and sums in all the channels there are. To exit the system, channels must be closed on L1 (slowass or expensive) and if there are not enough funds in the channels, create new channels with more funds, again on L1.

Even in the most absurdly centralized scenario, the system is practically unusable as soon as there are more than a handful users. Or people just stop pretending and use custodian tokens, aka bullshit IOUs.

Reply on news.ycombinator.com