About LN, very interesting posts there, especially
the long one by hisslyness (many thanks to him). I also remember Bob's struggle with LN some years back.
I don't know... I've never set up a LN node or anything, so I can't have an informed opinion, but the feeling I get by reading about it, is that it looks cool, simple, fast, cheap on the outside, but is quite complicated and messy on the inside. Also, the constant need for pampering the node, in order for it to function properly, is not good. These are not good signs, and I fear that they may cause problems for its future/success. I hope I'm totally wrong on this.
The whole thing brings back memories of
Mike Hearn's blog post where he declares he's abandoning the Bitcoin project, and even jbreher's transaction limit posts here in WO... Don't get me wrong, I'm totally in favor of 2nd-layer solutions for micro-transactions and believe this is the future, but I sense that something may not be quite right (yet) in the current state of affairs in this matter. Just thinking out loud...
I think what we are watching is tech finding it's way. It always does.
The fact is Bitcoin encompasses the EXACTLY CORRECT mix of attributes to have been what was inside pandora's box. I am not saying it's perfect. I am saying it is more than perfect enough. What I mean is it's limitations are not big enough to stop it, and tech will find the ways around those limitations. It's amazing to me that Mike Hearn could not see this. Because it is the fallacy of a beginning coder. It goes like this:
1. Make new software.
2. Discover limitations and mistakes in the code or archetecture.
3. Succumb to the temptation to tear it all up and start over.
Even with a project that does NOT need the "immaculate conception" that bitcoin had this rookie mistake is too costly to do much of the time. Other important protocols, and software all have faced this moment, usually multiple times. TCP/IP, Netscape/Mozilla/Firefox, Windows, Debain etc. Those are all protocols and projects that reached breaking points and in ALL cases there was always a contingent of the choir saying that the solution is to just start from scratch. And sometimes that does work. OSX. Chrome... But Firefox has survived and overcome it's many gangly limitations over the years in spite of being the first mover and having to learn a lot of the hard lessons for the competitors to got to avoid "for free". And in some ways BECAUSE of that. It was the first mover. I still use it.
But Bitcoin has a unique property. It is a kind of software that we cannot really just "do over". TCP/IP is the closest in that list, and it is no coincidence that it is a protocol, rather than an application, so to speak. Because Bitcoin is arguably also more a protocol than an application. And there are two reasons we have to "live with it" in spite of the issues that exist.
1. It is "in production". The train is moving and individuals, and companies have parked BILLIONS of USD worth of value in it. It is in use. Well over a decade of use at this point. And value is flowing into it at an increasing pace. The question of whether it has become too big to stop yet is being argued still. But I think the argument is starting to become very hollow. Roger Ver tried to stop it 4-5 years ago, and it crushed him. He may not be ready to admit it yet, but he has lost at LEAST hudreds of millions of USD in value by betting that an "improved" version of Bitcoin would be able to overtake the production network. Too late. Too late 4 years ago. Too late now.
2. It is becoming money. AND it is doing so as a decentralized (unto distributed) network that no single player fully controls. Not even the developers. (BCH proves that as well.) And because of this if we try to start over it will be inferior in this most important ways EVERY TIME. Pre-mined, gamed, centralized etc. And the world is waking up to this. The smartest folks can see at least intuitively why Bitcoin is unique.
So... The lightning network? And all it's problems? I will not go through the argument here... but the need for on chain transactions is one. I believe this will be what it will be. Bitcoin will become much more expensive to use as it is used more and more, and the users will figure out all kinds of novel ways to optimize this without tearing down the protocol. Lightning is one. One clue to this actually working is El Salvador. For them bitcoin is simply what is inside Chivo, or WoS, or Muun or whatever. They don't care about "on chain" or "base layer". They have not only MOVED past that argument... they STARTED past it. And they are using the Bitcoin network nearly for free. I can imagine there will be people who NEVER DO a base layer transaction in their lives, and yet will live on Bitcoin. AND it is possible for them to do so in a self-custodial manner.Yet not required.
What about the difficulties of running a lightning node. I am not sure how long I have been running a node... But the current one is about 2 years old. and the other one was something like #197 on the network. So three years? I have worried about channel balancing, and I have not. I have added and subtracted channels along the way.
Crazy thing is along the way it is getting easier and easier to use. Some of my channel partners keep their channels meticulously balanced. Personally? I do not see the benefit in this. It really only matters if you have a very small number of baldy connected channels. If you have several well connected channels then you can make and receive payments all over the place. And your NODE stays balanced even if your channels do not. I assume if I were living out of it I would have to continue to inject value into the outgoing liquidity side to keep the system working, but that is now easily done using all KINDS of tricks. Opening and closing a channel, yes, but also looping in and out, leasing liquidity etc.
I believe as the LN evolves we will see it begin to manage itself as far as balancing goes. MPP for example will make stagnant small channels useful, and I think as pathfinding evolves nodes can be smart enough to manage their fees in a way to drive traffic into and out of the channels that need a little push. Software can do ALL OF THIS.
I will leave you with one prediction...
Some minders, in the future, will monetize their businesses, not only by SELLING the bitcoin they mine, but also by locking it into the LN (and other layer two) as they competitively become the new payment network of the world. Why would they SELL it all? The only reason now is they need it for OpEx. Well, there is another way coming that lets them HOLD it, and make money as well.
Mike Hearn is undoubtedly a smart dude. But he was wrong. Very, very wrong.
