Hence a solution that "works great 90% of the time" is absolutely unthinkable for any real-world application.
Meanwhile, there are other products on the market--native fast stablecoins--that work great every single time.
So Lightning is a product that has security problems, does not provide any of the benefits of Bitcoin, and is not consistently low-latency--and competes with products that are a thousand times better. That's why it hasn't taken off despite being around for so long and pushed so hard by the Bitcoin community.
The 90% number is generous honestly. Try onboarding someone new to self-custodial Lightning and count the steps before their first payment: open a channel (on-chain fee, 10 minute wait right there), then discover they can't receive anything because a fresh channel has zero inbound liquidity, so now you're explaining what inbound liquidity is to someone who just wanted to buy a coffee. There's a reason Wallet of Satoshi got huge and it's not because people love custodians. It's because it skips all of that. The network's own users keep voting custodial, which is the quiet part.
But "work great every single time" is doing a lot of work in that stablecoin comparison, and it doesn't survive the Chivo point from earlier. What is Tether if not Chivo with better marketing? It froze over a billion dollars of USDT in the first half of this year alone, and more than half of what it froze last year it permanently destroyed. Circle froze wallets too. Works great every single time, until your address is the one on the list. So it's not Lightning failed and stablecoins won. It's that people chose the convenient IOU over hard money twice in a row, and one of the IOUs was wearing a Bitcoin costume.
Tether at least never claimed to be Bitcoin. Which is kind of the whole thread.
