Oddbean new post about login | logout Except for Drivechain, what can be done to enable seamless Bitcoin payments at scale?
If BitVM can enable a trustless bridge then a ZK rollup on bitcoin could be the ideal way to get there.
A better question might be: is there really that much demand for Bitcoin payments? If yes, why not make it a custodial protocol to start with to remove all the complexities of a truly decentralized system? The reasons I ask is because: 1) payments is a highly competitive industry with very low margins, LN will be constantly fighting all the world's banks and payment systems (and musk soon), and it's going to be an uphill battle for LN because it's a decentralized system and fiat solutions are actually reliable and pretty neat UX wise. 2) Non-custodial low-value payments do not have quite the same ring as non-custodial storage of large amounts of value, I fail to see why any kind of censorship resistance is even needed in majority of LN payments.
After all these years, IMMORTAN lite wallets still give me the best non-custodial user experience (success rate, speed, fees). And for onboarding, ppl could still use the hosted channels (which i don't) and swap from HC to own channel once they got enough inbound. They just all look a bit shit.. 🤷
The demand isn't so much for Bitcoin payments as it is for low cost, fast and reliable payments. The fiat card payment networks and alternative fintech solutions that rely on them are all too expensive for small merchants. A café I was at recently (that doesn't accept sats) wouldn't accept a card payment for a drink either. The proprietor explained that he loses money on the soda since the card network 2%+25 cents is bigger than their margin. He'd said he'd happily accept sats if it was they didn't have these fees and it was technically simple. Driven by this need, it seems to me that centralized LSPs (LN service provider), like Wallet of Satoshi, will end up being the PSPs (payment service provider) for LN payments. They can match the UX of most fiat payment systems if they wanted to and they can do this with much lower transaction fees than fiat networks. Of course, this isn't the LN vision, but it will be the reality. However, it is also nice that LN is open and can be semi-decentralized as well, so while a few large LSPs will handle most payments, there is still room to hook in niche channels and nodes for other use cases. Finally, the centralized LSP as a PSP is not such a bad thing as most of the value add from the card networks can be replicated as well, such as purchase protection.
There are numbers: 5 payments per year and your channel becomes economically reliable. nostr:nevent1qqs0kl6x47jpel07kerqyzdpzut8kzs3wzxpc69jxs3q388fd7mnnsqpz3mhxue69uhhyetvv9ujuerpd46hxtnfdupzp3wd64ehu3l4gfkfm63yxqfpztkk97a8k5683p5pvphhnadtj6p2qvzqqqqqqyyxeenj
I assume you mean economically justified to make that channel opening tx. Compared to what, then? It will lose to zero fee bank payments I have now. For my peer, it will lose to lending btc at 2+% return instead of putting them in a channel which does 5 small payments per year. What is the probability channel will even live that long and not get force-closed?
Compared to onchain fees. That's theoretical numbers derived from papers about payment systems from 1960s. On such timescale LN if compared to legacy payment systems is looking like a frog leaping forward like crazy. Force closings are problem indeed and it is worrysome that everybody is busy with something not including FC as KPI. But this is free market activity entirely. So we'll, cest la vie. Dashed line is for hosted channels on this plot. Horizontal axis is payment frequency. Vertical is for economic effect which is difference between onchain and lightning network operational costs. https://i.nostrimg.com/601eb8a02cc2382d180748c35fffc76cd0269a24edb5b816e5c78a4ea75df215/file.png
It is important that you can't compare lending bitcoins, normally, since bitcoins are the best collateral. Lending will always lead to additional risk but in lightning it appears to be measurable while in other cases it is subject to third-party evaluations or collateral asset depreciation against bitcoins.
off the top of my head, in LN you should also account for hardware failure, 0days, fee spikes which will destroy in-flight htlc resolution in broken channels, all of these not very predictable.
There are many technical ways to loose bitcoins. All of them are different to custodial and third party risks. You could say that one may use unreliable random nber generator and leak keys without any LN.
You bring up a very important point, and I think this is where the majority of Bitcoiners are split. The plebs want fast, cheap, and private payments. Even LN without the vulnerabilities cannot compete with cash. Hopefully, the people on this side will realize drivechain is the only decentralized option. The other side are the Michael Saylors and BlackRocks of the world. They do not want privacy or scalability. They want rich people to have a new rock to invest in, so they can trade it for CBDC tokens. What do you want your Bitcoin to do?
I think you’re underestimating the risks with this. Sure would love cheap payments but not at the increased risk drive chains might be (I don’t know how much more risk they are, and I don’t think anyone is really capable of knowing that) I’d argue the smart play is to prioritize security and add in the cheap payments later when ideally you can do it without changing the layer 1
It's a bit biased way to put it but my answer is: I'd rather be closer to Saylor et al and their billions than to plebs who enjoy running stuff on RPi, or whatever they do these days.
Thank you for your honesty. I hope that side loses though 😆 I know it seems like only the nerds want privacy, but the entire world NEEDS privacy. A $5 wrench attack hits much harder when governments can print wrenches out of thin air. A lot of people have hope that Bitcoin will liberate us from fiat authoritarianism. Do you think this dream is dead?
