Lightning Network has Failed

The Lightning Network cannot scale Bitcoin to mass adoption. Payment channel technology, like Lightning, requires management via the L1. So that means users must make L1 transactions in order to manage their L2 channels. Therefore the L2 is bottlenecked by the L1. Therefore Lightning *cannot* actually handle global/mass adoption.

Not only that, Bitcoin itself cannot handle mass adoption. Even if it were only being used as "digital gold" (buy once and hold forever), it would still take 100 years(!) to onboard the world to the Bitcoin L1 — which is a pre-requisite to moving to the L2.

So the L1 is too slow to support its L2, and the L1 is too slow to handle mass adoption in any case. The current trajectory then is to "scale" Bitcoin via custodial services that offer users IOUs. These are banks! This is our current system! If this is the outcome then the whole project will have failed, because there's no way to stop these custodial services from engaging in fractional reserve and breaking the 21 million supply hardcap! No to mention other rent-seeking behavior.

Note: when criticizing Liquid as an 11-of-15 multi-sig, I should have pointed out that most Ethereum L2's are also functioning effectively as multi-sigs today! But the difference is that those systems have a clear, well established, path to complete decentralization — Liquid does not (as far as I know). The L2s have training wheels on them. If those training wheels don't come off then I'll be ripping into them as well.

0:00 - intro
0:34 - state channels
1:09 - why layers
2:28 - how lightning works
3:45 - opening & closing channels
6:21 - routing & capacity
8:31 - hubs
10:44 - liveness & storage requirements
12:59 - L2 is throttled by L1
16:53 - L1 is too slow for adoption
19:16 - custodial services (banks)
20:18 - breaking the 21M hardcap
25:27 - Liquid
26:50 - actual scaling solutions
28:48 - afterword

[danmunch]

I’m a BTC Maxi/cultist myself, and am truly grateful for any thoughtful, well-reasoned criticism of the project. Thank you. Subscribed.

[subliminal_donkey]

This is literally what the gold standard was. It was too expensive to move huge amounts of gold around, so the central banks just kept everyone's gold and a ledger of who owned what, (kind of like the lightning L2) and when someone wanted to pay someone else the bank just adjusted the ledger (big simplification) instead of shipping bars of gold around all the time. However at certain times of course a big shipment would be justified, maybe to settle trade between countries, i.e. changing the L1. It was also easier for people to use bank notes (kind of a layer 3) because they were more divisible and portable than gold coins, so there was an incentive to leave your gold in an institution and use notes instead. Though ironically that system was better because you could, at no cost, simply walk in to a bank and exchange your bank note for gold coins again should you wish. Which clearly the Bitcoin system cannot do. I would, for future reference, stop comparing Bitcoin to the batshit crazy mess of a fiat system we have now but at least compare it to a gold standard system which was at least based on a scarce resource of value. Any system based on Bitcoin, and thus on something limited and of actual value, is better than the fiat system we have now where the currency is literally baseless and entirely corrupt.

[dh]

This is one of the best videos on bitcoin/crypto tech I have ever watched. Clear and paced explanation with great graphs and sources. Thank you.

[Blockchain_guru]

Great video I was surprised someone else also realized that lightning network will lead to centralization of bitcoin which of course we don't want

[cheesyandy]

The important point is that the 'banks' or institutions will be able to provide proof that they own the real BTC. This is much easier to do than auditing fiat.

[kkiller]

As things are currently, Bitcoin LN definitely cannot scale enough for global adoption. However, the LN model, with some tweaks, will be the most scalable solution for a global transaction network because it does not require that every Miami node holds transaction data for cafes in Delhi -- it keeps the channel state data local to where it is relevant, and a properly designed system can further compress txn data using roll-ups of txn IDs and aggregate txn effects.

[kkiller]

Doesn't FediMint solve the fractional reserve problem?

[anthonyriseley]

The difference is that any rehypothecation the banks do will not be bailed out. We get a free banking market with bitcoin. The current system is not free market due to money printing

[AndyHage]

even with lightning factories its not enough scaling. and now this guy is afraid that transaction fees will be too cheap to sustain miners after 2140. that doesn't make any sense at all. like its just meant as fud even though he clearly understood it 10 months ago.(20:42)

[jdoedoenet]

I may be missing something here, but if the majority of transaction volume starts to be offloaded from L1 and begins to concentrate to Lightning (L2)...then this completely alters the supply/demand equation for transaction fees. A lot of what I heard is premised on the assumption of ever-INCREASING L1 transaction fees. Can this really be a hard assumption from which you extrapolate arguments if DECREASING L1 transaction volume counteracts transaction fees? After all, transaction fees are primarily a result of competition for L1 block space. If we see decreased competition for L1 transactions, does that not reduce transaction fees?

[MrCoreyTexas]

At about 5:40 you say you can't add more bitcoin to the lightning channel, that's not correct, there is a thing called a submarine swap, but guess what, it involves another L1 bitcoin mainnet transaction, which kind of defeats the purpose of lightning. Also one thing I noticed on electrum is some of the funds you deposit are unusable, because it has to hold a significant portion for miner fees. And the latest electrum the smallest channel you can open is .002 BTC. I guess you could say that lightning is good for testing your understanding of Bitcoin Script at least, LOL

[Bobby_Strange]

Lightning has not ‘failed’. It provides a viable network for a significant bump to transactions. It doesn’t have to solve the entire world to be worth something and is working pretty well for a significant number of people currently. More scaling solutions are needed, as LN is not the end all be all so thank you for demonstrating.

[cbemerine]

It's interesting, I will give you that. I believe that any short-comings will be mitigated programatically as they crop up, so still optimistic!

[lludol]

Thank you for the video, it's hard to find content about that! I am trying to learn more about lightning and other L2 solutions.

But from what I understand, we are screwed no matter what:
1. We change the BTC L1 (block size or faster block), but then it means more resources will be needed to actually mine/check the blockchain, which means only big institutions will do that (like Ethereum)
2. We change nothing, and we will arrive at institutions managing L1 / L2 and everyone will be forced to use only custodial, which means we use the same current "fiat" system. The only differences are the currency (btc), the speed (faster), and the cost of transactions (no more wire transfer fees, for example, so sending money between countries will be cheaper).
3. We do a fork, but then we will be against big institutions/maxi, who will not want to change anything

What am I missing?

[ishtyleretienne]

Kaspa, digital silver and programmable money has entered the chat😆

[Leinad_noES]

Best explanation of lightning I've ever heard so far.

[ishi]

Liquid IOUs have a verifiable peg. No possibility of fractional reserve

[JeremyFisher]

best video on this I've ever seen. Thank

[scholasticphilosopher]

The thing is there are like ten different projects all of them are going to be together in sync increment the scalability of BTC to billions of users. All of this is simply a question of they all solve a different problem.

They all work together and strengthen each other. Bitcoin can absolutely scale and the higher the number the greater the amount of people who will try to solve the issue.

[kkiller]

Eltoo solves the channel state management problem and makes it safer to run a node. Coinpool further extends L1 scalability to allow Eltoo channels to exist as leaves in a Merkle tree embedded in a UTXO.