Topic: POLL: Have You Ever Used the Lightning Network? (Read 751 times)

It is possible to build more layers on top of the Lightning Network. For example, Sphinx Chat takes advantage of the way LN transactions work and provides a decentralised encrypted messaging service.


The Lightning Network was not designed for one-time payments. You are supposed to use your channels multiple times before you close them. You have to pay an on-chain transaction fee for both channel opening and closing transactions. Since LN fees in most cases are extremely low, those channels will quickly pay themselves off if you use them more than once.


I am not sure if I understand your question correctly. Lightning Network payments are not recorded on the blockchain. Instead, both participants of the channel update its balance without broadcasting anything to the Bitcoin network. If your Lightning payment is routed through multiple nodes then it becomes nearly impossible to track your transaction. Once you decide to close your channel, the closing transaction will include only yours and your peer's address (or either of them).

I have not used the LN before. But I have a few questions.
1. Apart from the timeliness of transactions procession, are there any other advantages of the LN?
2. The cost of opening/closing channels and the minimal cost of transaction fees, if summed together, will it not equate to the original transaction fee on the blockchain?
3. In terms of fraudulent transactions or transactions complications, if the channel is not closed, can the transaction also be tracked in the blockchain?

Bump

The few times I've had a LN-transaction that failed, my wallet tells me I have to wait. That's the annoying part, I can't just try again instantly, the transaction is in limbo. That also means I can't just try another wallet to try another route.

I'd say less than 10% fails nowadays.

I'd say it's about 90% reliable in my experience, high because the other end of my channel is connected to what they call "Hub", aka a well-connected lightning node.
Most of those failures are from instant exchanger from time to time, specifically Fixedfloat, but I think it's their channels' capacity since they usually 'fix' it after a while.
In jochen's mempool stat and some users of custodial lightning wallets, I haven't experienced any payment failures.

How many of your Lightning Network payments fail to reach the intended receivers percentage wise? Do you consider it to be a cause of worry personally?

---

Regarding donations via the Lightning Network. After a quick search, I found that the Bitcoin Development Fund accepts LN donations.
Wikileaks does the same, but I think they accept it in their shop, not exclusively for donations.

Thanks, apparently google search isn't reliable when it comes with "lightning donation" queries.
I have found a list but most of them are inactive or have very high minimum donation amount.

lol

You can also "donate" anything from 1 sat to at least 2 online casinos. Or feed some chickens Last time I did that they charged 1000 sat, now it's 50 sat.

Yeah, I've been sending some lightning satoshi to the site since September last year for testing purposes (to try if my channel is the problem when I'm failing to send LN funds) aside from appreciation for being an early adopter of lightning.
Actually, it's the only site that I've found that accepts low value LN donations, so there's no other choice :D

I don't know if lightning donations to that site has been available much earlier than Q3 of 2020.

I just noticed jochen-hoenicke.de, my favourite Bitcoin fee estimate site accepts LN donations:

Big spender as I am, I paid for an hour

If you're new to LN, this offer still stands:

No I haven't. I only read the whitepaper. And I'm unhappy with the pace of LN protocol development.

Sure we have the contract part of the protocol completed and ready to build (and there are at least two protocol implementations of it), but even the whitepaper itself notes there are a bunch of holes that need to be plugged.

For example, 8.4 Payment Routing:


We do not currently have a specification for this - every implementation is doing what it wants - and I think this it is absolutely necessary to have standard for this kind of stuff to resolve idiosyncrasies and quirks in what the description above has. (Eg: "The core Tier-1 routes can be online all the time" we haven't even defined what a tier 1 LN node is!)

[on a side note - even my ISP has crap routing, evident by my multiple attempts to load the Reply page and post this message: obviously we do not want to allow for that kind of behavior in LN.]

Another example of out-of-scope concepts is in 9.4 "Data Loss":


Neither how a data retention 3rd party would work nor how to store transaction data on a node redundantly has been defined, and quite frankly nobody can make such a service if they don't even know how the protocol would use it relative to their service. Not that I like the idea of a 3rd party intermediary in the first place - making data loss from lossy UDP connections extremely unlikely is much better.

Yet another unfinished business is in 9.2: Forced Expiration Spam, when someone expires many channels or activates too many timeouts at the same time. One solution is proposed for the former and two for the latter (the other is in section 10) but none of them are elaborated well enough to actually design a standard mitigation for this. And IMO this one is so bad that clients shouldn't be allowed to run in production without a patch to this.

I am not anti-LN (I also do not think it's an altcoin, but at the same time I feel that I can't call it canon bitcoin either since it allows for sub-satoshi payments) but frankly it should be clear that our work is not finished yet. We risk having different incompatible standards of LN implementations that segregate users to the same client if we don't have specifications for everything. Some anonymous guy was bankrolled by Square in tenths of BTC which is 2x its worth now to develop LN and we need to see some results because we aren't seeing any major additions. People are talking about increasing the block size as a solution for some of these vulns but this'll just introduce more drama. There are already companies like LN Labs who have their own view of how the network should be constructed and stuff like this is going to faction the network.

Let's make things simple.

- Mitigations for spam expirations and
- a proper, lossless, routing protocol for LN contracts

has to be standardized first, and maybe then we'll see some real adoption.

Then wallets need to make it easier for users to "move" BTC between mainnet and LN.

Then the majority of wallets and especially exchanges have to implement the protocol too (do not underestimate the influence of exchanges).

You aren't the first one to suggest that the Lightning Network isn't Bitcoin. Casascius coins are Bitcoin as much as Electrum, Core, Ledger or any other implementation is Bitcoin. They are not, but they store your keys and allow for easy transfers of digital values. I can't agree with you that LN isn't Bitcoin, because as PrimeNumber7 mentioned, you are transferring BTCs over the network. More precisely, two parties are signing a balance sheet where they agree what each of the participants gets when it's time to close the channel. Although those transactions don't get broadcast on the blockchain, they affect what the final balance (in bitcoins) will be. 

I used it a few times a couple of years ago when the first Neutrino clients were released. Lightning App was the first wallet that I used. It was really buggy and I had a lot of failed payments. I think eventually they stopped developing the wallet which is odd because Lightning Labs entire existence revolves around the Lightning Network.

I also tried out Zap Wallet, Blue Wallet and tippin.me. Now that we're seeing new services that support lightning payments I am thinking of giving it another try. The new Moon browser extension is something I'm interested in trying next time I do online shopping.

You are using irrelevant examples. None of the things you named rely on bitcoin. For example shitcoin (WBTC) built on top of another shitcoin (ETH) has nothing to do with bitcoin. Some day the ethereum foundation could decide they wanted to spend all the WBTCs on their chain and they can do that easily regardless of what is locked on bitcoin blockchain. Same with BSV shitcoin.

On the other hand LN is built on top of bitcoin and relies on bitcoin and uses pretty much the same protocol without the "blockchain". Users are still in control of their funds and they rely on Bitcoin protocol. Maybe we can't call it "bitcoin" but it is not an altcoin either, it is a second layer built on top of bitcoin protocol to help the scaling.
And I believe that using a second layer is the only way to scale bitcoin but as long as we also increase the on-chain capacity.

When two people open a LN channel, they are having bitcoin held in a multisig address, and are also pre-signing a closing transaction in which each party will receive the same amount of bitcoin they had (less tx fees) when they opened the channel. As transactions done between the two parties of a channel, each party will exchange new signed transactions that will allow the other party to close the channel while receiving an agreed-upon amount of bitcoin. Having coin in a LN channel can be compared to having an unconfirmed transaction that was sent to you, except there are additional safeguards against double-spending.

The closest thing to bitcoin in your list is WBTC, and with WBTC, the end-user has to trust a third party to follow through with their obligations to allow for WBTC/BTC redemptions.

Yes, I am aware. Being that this is the B&H section, I wanted to minimize the technical details. See the bolded part of my comment:

I've never used it, i don't usually do many small payments so i didn't need it.
But I maybe use it in the future if i'll need it.

I would. I did. I’ll do it again because that is what it is. I don’t care what cryptographic magic or guarantees make users feel like Lightning is Bitcoin, it isn’t...  Anyone could make an altcoin with the same cryptographic guarantees, but nobody would call it Bitcoin and rightfully so. Do you think ETH’s Wrapped Bitcoin is also Bitcoin? Do you think USD loans given with Bitcoin as collateral is also Bitcoin? Are Casascius Coins bitcoins or are they holding BTC? What about leveraged BTC on exchanges or call options, are those Bitcoin too? Where is the line drawn? Just because Core maintainers developed it doesn’t give anyone the right to call it Bitcoin any more than CSW can say his altcoin garbage is Bitcoin. It should be pretty obvious that Bitcoin exists on the Bitcoin blockchain but I figured it was obvious that having a dick means you should use the men’s restroom, so maybe I’m just not with the times and LBTC identifies as BTC.

"Double-spend" in this context sounds inaccurate. It suggests that the other party can freely create any transaction which is not the case since the funds are locked in a multi-signature address. Instead, the other party can perform an uncooperative channel close and broadcast an old state of the channel (an old commitment transaction). Such a transaction can be easily revoked by the other participant of the channel if they come back online before the locktime expires. Also, a penalty transaction can be broadcast if that happens.

You can read more about it in this blog post.


There is a risk even if you don't deposit your own coins. For example, when someone asks you to route their payment through your node then you receive some coins from them in one channel and you need to use some other channel to send the exact same amount (minus your routing fee). Now, if your node went offline and that person broadcast the previous state of the channel, you could lose money this way.

Also, both parties need to maintain a channel reserve, which is equal to about 1-3% of the channel's capacity. So, if you open a channel to someone, transact with them a couple of times and the other side ends up with the 0 balance. The other party can still lose money because they have to maintain that unspendable channel reserve on their side.

The person who created and opened the payment channel is risking losing his whole deposit if he attempts to manipulate the off-chain transactions. The risk is not the same for the other participant in the channel because he didn't deposit his own coins. What you described will be a great manipulation prevention method when dual funded channels get implemented. That way, both parties can fund the channel and they risk losing their money in case of fraudulent attempts. 

I would not describe LN as an "altcoin network". LN is a 2nd layer above the bitcoin protocol. If you use LN, you are transacting in bitcoin. If you have an open LN channel and receive a LN payment for 0.001 BTC, you are cryptographically guaranteed to own that 0.001 BTC, the only difference is that the transaction is not settled on the blockchain, however, the risk of losing the coin is near zero because if your counterparty were to attempt to double-spend this transaction, they would be risking a near-certain loss of far greater funds.