Using lightning network on a day-by-day basis has became increasingly easier for bitcoiners thank to the speed of development on this brilliant protocol.
Orange pilling nocoiners is easier than ever because of solutions like Wallet of Satoshi, LightningTipBot in Telegram, Phoenix, Moon, Alby etc. These wallets give really a magic-like experience to users most of the time, for 99% of the transactions, and this is really pushing ahead the ecosystem and the development on top of Bitcoin.
After several months of self-hosted lightning payments with my node I feel that I can provide a reasonable feedback about my own experience being my own bank. Overall the experience has been great, Zeus wallet is the perfect terminal for my node management when I'm messin' around; while at home I really enjoy Ride The Lightning. Both LND and c-lightning are good implementations, with both their pros and cons.
But here comes the troubles. I try to not connect directly with huge hubs like Acinq or WoS but I found out that is nearly impossible to have a good LN experience without having connections with those hubs. I often find myself struggling with routing errors and payments that constantly fail. Liquidity is always a huge issue if you want to try the 'decentralized' LN experience (therefore only connecting with community nodes etc).
I hear your thoughts though:
Man, if you don't connect to big nodes you'r never going to make through with +100'000 sat payments!
I'm aware of it, but in a certain sense I wish I could...I'm in love with the idea of individuals being their own bank and I hope that liquidity issues will be solved as the adoption progress.
I'm really curious about your own experience with self-hosted lightning nodes:
  • Which LN implementation do you prefer?
  • Have you ever experienced troubles while trying to pay with your LN node?
  • What do you think about the idea of avoiding direct channels with big hubs? Is it a good idea from a 'sovreignty' standpoint?
Ps. Stacker News is brilliant!
You are right sadly. I came to the same conclusion as you, without channels with these big hubs it's almost impossible to have a good LN experience for amount above 100/200 k SATs, or you have to lay the price... And even if you can by connecting to multiple middle sized nodes, it's mostly because they are themselves connected to the massive node of ACINQ or WoS.
My hope is that the liquidity will maybe increase if the value of Bitcoin goes up. For example I don't plan to depriciate my own node if Bitcoin goes 10x in price so liquidity in USD term would be 10 times higher on my node. So even if the principal driver for liquidity increasing is demand and usage obviously, maybe an increase in price could substantially increase the liquidity, allowing for bigger payments to be routed by medium or even small nodes (0,5 to 10 BTC) but it's only speculation here
reply
My hope is that the liquidity will maybe increase if the value of Bitcoin goes up.
This.
reply
reply
I've never heard of these, thanks for the suggestion! I'll took into both.
reply
  1. If you're willing ensure your payments wont fail, get blixt wallet (non custodial), open a channel with bfx-lnd1 (the node), it will almost never fail and it will be extremely cheap (on LN fees) here's why:
  • almost every node has a big channel with this node and as this node is a "sink" node, it sucks liquidity real fast then 95% of all nodes connected to this has its liquidity on bfx's side, therefore when you try to route a payment you are pushing liquidity over to bfx and bfx pushes liquidity to the destination real easily because the majority of the liquidity is on its side, ergo your payment will almost never fail even if it's couple of million satoshis (0.01btc+)
  1. But the situation reverses if you're willing to receive bitcoin you can't use bfx since it's gonna be expensive once few nodes have the liquidity to process and those which have will charge you for the effort. In this case, you should search for hosted solution or have an LSP to open a channel with you (at a cost)
So here's my tip:
To send payments, use the solution 1, if you need to receive BTC, use a custodial solution like Alby, Wos, coinos etc.. otherwise inbound liquidity upfront will be sorta expensive if you use non custodial solutions
If you're a merchant and need to receive a lot of liquidity, I suggest you purchase inbound liquidity with deezy.io or lnbig and purchase like 1btc inbound or a little less depending on your I/O.
reply
A lot to digest here, thanks a lot!
reply
Payments can split. That’s why hubs will be less important in the future. Network will evolve
reply
Hope so!
I just understood how little I know about the Lightning and that I can really twikle a lot to improve my node's efficiency, thanks!
reply
I have 28 channels and dont have a problem sending or receiving payments above 100k, using LND btw
reply
People still don't know how to use properly LN or wallet apps. MPP must be by default ON.
reply
Yepp, for sure there is a lot of room for improvement XD
reply
I liked OBW but it looks like they have stopped working on it.
Electrum is pretty cool for unannounced channels.
Switched from alby to mutiny on nostr but still some issues (still early). Thinking to try the new Zeus wallet for this.
Phoenix is one that I use to make sure tge payment doesn't fail.
Also have used WoS, blixt, & umbrel wallets too.
reply
Wow, what a diverse setup!
OBW was one of my favourite bitcoin & LN wallets, unfortunately FOSS projects sometimes stop, same for Zap (even though now BitBanana inherited the project). Elecrum is my way to go for onchain funds (desktop), I've never tried the LN function nor the mobile client, I'll put that on my watch.
Overally I'm glad that there are a ton of possible solutions, Blixt is a fantastic solutions for private channels (thanks @DarthCoin for you blog posts!!)
reply
Hopefully ZEUS becomes my last Bitcoin wallet after having separate on-chain and Lightning wallets until yesterday (shoutout to @ACINQ for making Phoenix, awesome folk).
The node is on the phone, lol.
reply
You should still be having seperate on-chain and lightning wallets tho. Your savings vault has different security needs than your spending wallet.
reply
It's done mainly because of my paycheck recipience through Bitwage; it's easier (less secure admittedly, however) to have one wallet for Lightning and on-chain assets I can then add to my LN channel liquidity if need be.
Plus having unique address makes it so that receiving my checks is far more secure than having Phoenix covert on-chain sats to Lightning ones.
reply
Okay but checking account vs savings account you know?
reply
Great suggestion. Containerization of funds is the best approach to self custody
reply
What do you think about the idea of avoiding direct channels with big hubs?
I try to open direct channels with entities I'm transacting with. And, as I have both WoS and Phoenix on my phone, it makes sense to open direct channels to WoS and ACINQ from my home node.
reply
Makes sense
reply
I find fun to send bitcoin over LN just to see how fast it is...
reply
Yeppp, you're right. Again, WoS is the best tool for orange pilling just for the reason you outlined: it's damn fast
reply
Well Phoenix and muun are fucking fast
reply
Yes, they are
reply
0 sats \ 1 reply \ @nyan 3 Jun
What's the problem with connecting to ACINQ or WoS? You still custody your funds and they don't know who you are. In case they become compromised you can always go back and take on the hassle of connecting yourself to smaller nodes elsewhere in the network.
reply
No technical issues with that, actually.
My point is that if we want a resilient LN that can eventually gracefully decentralise back again in the event of a big player being shut down, we need to make the LN network graph less sparse.
Just to not be trapped into small isolated components when some big players like WOS will be shut down.
Also, my gut feeling is that KYC procedures can be enforced easily on big LN nodes. For example, ACINQ could be forced to KYC anyone in order to open a channel with them. This would mean that eventually KYC-ed nodes will create a net of nodes that are only connected among each other. Since businesses are easily caught, I bet on them being the target for governments. And since businesses nodes are the biggest nodes in LN, if that happens than we're cut out of the LN where real liquidity is.
I might be wrong, I hope I'm. But in the eventuality of that happening, opening channels with moderately big nodes but not giant business nodes like WOS shall be a priority.
And yes, that forces you to sacrifice a bit of routing earnings. That's worth it imo.
reply