Hacker News Reader

Comment by tudorconstantin | Hacker News Reader

Why can't people be paid directly on one of the layer 2 systems?

(Author here)

They can, but how large a part of the world's work force do you think is able to get their employer to pay out their wage within a Bitcoin layer 2 system? If layer 2 systems only support people who receive their wages within that layer 2 system already, how useful is it really at increasing the adoption of Bitcoin?

Also, as far as I can see, it would require us to redefine what a "bitcoin" actually is. Right now, it constitutes something in the Bitcoin blockchain, whereas some proponents of layer 2 protocols seem to wish that we redefine it to include unconfirmed Bitcoin transactions. I'm not saying this can't become the case, at some point, but right now no one considers the holder of an unconfirmed Bitcoin transaction to be in possession of bitcoins. So, at one level or another, something needs to change.

If an employer is willing to pay the employee bitcoin in layer 1, how much less willing would the employer be to pay in layer 2? Not much, I think. It doesn't seem that it should be much harder to use layer 2 instead of layer 1. If it is, then isn't this a usability issue affecting all users and not just employers?

If the employer is willing to only pay fiat, then there is no new problem, is there?

A bitcoin is just an entry in the BTC blockchain ledger. And I think you are misunderstanding the concepts in what you call "layer 2". The idea of the Lightning Network (which I assume you are referring to) is that you can execute off-chain transactions without needing to trust anyone. Its not the same as an "unconfirmed transaction" which leaves you open to a double-spend attack.

> A bitcoin is just an entry in the BTC blockchain ledger.

I agree.

> The idea of the Lightning Network (which I assume you are referring to) is that you can execute off-chain transactions without needing to trust anyone.

Being off-chain, a LN transaction is not a bitcoin by your definition. While LN might be trustless, it's not the case than an LN-transaction is as good as a bitcoin in the blockchain. With LN, there's a race against time, whereby a node can lose funds received through LN, if it fails to settle on the blockchain. So the two (blockchain-tx vs LN-tx) are definitely not equivalent, but it's possible that future schemes can close that gap.

Timeouts are a necessary part of payment channels, since if you just send your bitcoins to a 2-of-2 multi-sig address willy nilly, the owner of the other key can hold your funds hostage unless you give him half. Hence the timeout, after which funds sent to the 2-of-2 multi-sig address can be redeemed by just a single public key (the sender's). The receiver needs to publish a transaction (that redeems the same 2-of-2 multi-sig output) to the blockchain before this point in time, or the funds received through the channel are up for grabs.

At that point, aren't you just reinventing fiat money?

No, a blockchain 2nd later system still gives strong security guarantees that are similar to the 1st later system.

> still gives strong security guarantees that are similar to the 1st later system.

Like Mt. Gox?

No, not like Mt. Gox.

The "second layer system" that people are actually talking about is Lightning Network, the basic idea of which has been around for a few years now. Basically, you open a "payment channel" of X bitcoins with someone else in the network, and you can trade up to X bitcoins with them almost-trustlessly - cheating is easily detectable within the protocol, and results in the party detecting the cheating settling the payment channel on the blockchain. Payment channels can be linked, so you can send money from A to B through payment channels A-X, X-Y, Y-B, without actually trusting any of those entities.

So blockchain transactions only ever happen when someone wants to open a new payment channel (which is only relatively rarely), when someone they directly have a payment channel open with attempts to cheat, and in the rare occasions that someone wants to hold "actual" bitcoins rather than a cryptographic promise of bitcoins.

There's no entity within this system which has central control over currency that goes through this system, and all cheating is detectable with rollbacks in place when it happens, so it's not like a bank, exchange, or even a traditional payment processing network at all.

Note: I have my own qualms with cryptocurrencies and the human layer, but you've got to admit the technological backing is sound within reasonable parameters.

Mt. Gox was a centralised custodian. Layer 2 systems like Lightning Network are not, because they can't run away with your money. If they try to cheat, you can always withdraw your money.

But detecting cheating and withdrawing requires you to have client software online and monitoring. Which individuals can not do and thus will outsource it to their payment handling company which will probably be the same as the LN hub.

It can be outsourced to trustless third party, who also has no control over your money and cannot cheat.

It depends on the purpose. For example: You can't hype fiat money anymore, too much liquidity.

Layer 2 is not money and so can’t be fiat. Lightning network is just a clever way to construct bitcoin transactions such that it is possible to send sub-transactions via different communication channel(not congesting bitcoin network) but preserving crypto currency guarantees.

> Layer 2 is not money

Ehrm, what? So Layer1 is not money for you either? Explain...

PS: I'm not agreeing with that "this is reinventing fiat money" comment.

layer 2 is the payment processing network, money being transferred remains bitcoin.

Reply on news.ycombinator.com