Topic: LN : Question on scarcity of coins (Read 262 times)

You can, but I suspect that the typical user won't bother.  Only time will tell.


It's not that LN REQUIRES the exchanges to be hubs.  It's just that when I look at how it works, that currently feels like the most likely growth pattern.

If you receive your purchased bitcoins over a lightning channel that the exchange set up with you when you open your account with them, then you can spend those bitcoins through that channel for free (or extremely cheap).  If you want to open a new channel with someone else, then you'll need a new on-chain transaction. This will require you to pay a on-chain transaction fee to open the channel.  Why would the typical user pay fees of multiple dollars to open a channel with anyone else when they can just send them bitcoins already very cheap?

If the exchange goes down, then the user can pay an on-chain transaction fee to close the channel, then pay another on-chain transaction fee to open a new channel elsewhere, and then they'll be able to spend those bitcoins again, but there's a good chance they may only bother doing that once.  Again, why pay transaction fees to open multiple channels when a single channel to a well connected and well funded entity will be sufficient 99.99% of the time?

There will be some users that will open multiple channels with non-hub partners.  Those will probably be the same people that currently run FULL nodes.  But I figure there will be 3 major types of users:

1. Knowledgeable, and technically capable users that want to maintain a decentralized system and connect to many non-hub channel partners.

2. Those that hear they shouldn't leave their bitcoins in an "account" with a third-party service providers, and set up a lightning enabled wallet. They make withdrawals from exchanges and then they make payments. They may not even realize that all their payments are routing through the single exchange channel.

3. The vast majortiy of users that don't understand the difference between a "wallet" and an "account".  They just leave their bitcoins in their account at the exchange (or some other web service) and spend them directly from there.

Ah, so the exchanges will function as payment hubs and customers will link to the exchanges and not to the merchants directly.

Interesting observation, but I see a problem here, when these exchanges goes down. We will have a single point of failure, if the backbone of this system is built on a network of centralized hubs. 

The exchanges will in a way function as centralized payment processors and if we have to rely on centralized authorities to "channel" our payments, then we are moving away from the original idea behind Bitcoin being P2P.

I am on the wire with this kind of implementation of LN, because I am a strong believer of not having to rely on centralized authorities, to be able to do transactions. You should still be able to open up "private" channels, without having to go through centralized third party services to make transactions.

That type of user won't likely be sitting on any blockchain bitcoins at all.

They will get an account at an exchange to purchase bitcoins.  The exchange will open a channel with them, with the exchange funding the channel with some amount of bitcoins and the user funding the channel with 0 bitcoins.

When the user requests to "withdraw" funds from the exchange, the "withdrawal will just be a lightning channel transaction to the user.  Those funds will be in the refund transaction INSTEAD of being on the blockchain.  This makes the funds no more or less "out of circulation" than if the exchange had sent the funds to the user with an on chain transaction.

The merchant wants to be able to sell bitcoins when they receive them.  Therefore they will open an account at an exchange. The exchange will open a channel with them, with the exchange funding the channel with some amount of bitcoins and the merchant funding the channel with 0 bitcoins.

If the exchange used by the merchant and the exchange used by the user are two different exchanges, then the exchanges will all most likely have channels open between each other funded on both sides with amounts that are calculated based on the net daily transfers between them.

Now when the user wants to pay the merchant, a route is established from the user back to the exchange where  they got the coins and from an exchange out to the merchant.  The user doesn't need to open a channel directly with the merchant. That would be a hassle for the merchant, opening individual channels with EVERY user that wants to buy something from them.

Now the merchant wants to sell those bitcoins, they don't need to close ANY channels. They just send the bitcoins that they JUST received form the merchant channel and send them back to the merchant over the same channel.

Yeah, there's still a lot I don't understand about the risks I've been reading, particularly the case as you've mentioned about a party being able to take control of channel funds. Like you, my plan right now is to find someone(s), likely from this forum, to start trying out channels with. I have a rough idea of how it works - very curious to see if it really is as easy as it is supposed to be in practice.


This makes me think, if a good number of people from this forum signs up to LN and we open channels to them/each other, we'd already have a small but significant network with probably a healthy balance of coins between themselves to probably stay off-chain for a very long time. Of course, since I expect exchanges to be the slowest to adopt, we'd probably see channels closing and re-opening to liquidate some coins, but they'd likely be enough left in the network.

Do you guys think there'll be a scenario where "trusted" payment hubs will start charging a fee to open channels? I was thinking again of a scenario where there might be individuals like Tom only wishing/needing to open 1 channel with coffeeshop, and coffeeshop will be looking at demands from perhaps scores of Toms who will only maintain singular channels with coffeeshop. Seems a situation that might be taken advantage of.

What I am getting at is this. Currently people hoard coins because they have "speculative" value and they increase the price, because they are taken out of circulation. The Lightning Network will encourage this, but in a different scenario, because people will fund channels to reduce fees on micro payments between them and a merchant that they use regularly.

I was curious to know, when the merchants can take "ownership" of the coins in a channel that are being kept open indefinitely. At some stage, the balance must be validated on the Blockchain for the change of ownership to take place. 

The point of LN is that there are multiple payment channels over which payments are routed. Sure the "actual Bitcoin" (whatever that may be) itself is not moved, by the value represented by it is. So the coffee shop can, if it has a channel open with an exchange, effectively sell the Bitcoin you paid him.

The coffee shop could route payments through Tom if the channel has enough balance. He's a node on the network, thus if he has other open channels, then payments can be routed through him.

I think there is an edge case that adds a small amount of risk...

If you open a channel with someone that you don't trust, and then you are not monitoring that channel, there is the potential for them to broadcast a stale state. They would need to be very certain that you are not monitoring the channel, because if you notice what they've done then their broadcast of the stale state is supposed to allow you to take FULL CONTROL OF ALL FUNDS in the channel.

Note that this also means that if you can be tricked into broadcasting a stale state yourself, then the channel partner can take all the funds in the channel.

You don't have to be physically monitoring the channel with your eyes, you can have software that monitors it for you, and I think there is a wait time after the stale state is broadcast, so you you have hours or days, not seconds or minutes.

I suspect that software and best practice processes have already been worked out for all of this, but since I don't understand them yet my plan is to only put VERY small amounts into channels with individuals that I don't already have a trust relationship with.

Same here, and I have a feeling most people have the same problem, which makes it hard to find reliable information.

I'm nowhere near having it installed.

I've been putting off getting a solid understanding of lightning network because I haven't felt that I had the time available to read and absorb everything I want to know about it.

I'm currently planning on spending some time this weekend learning and figuring this stuff out.  Then, I hope to get a node up and open some channels.  I'd be happy to establish a channel with you once I'm sure that I understand what that means (hopefully early next week sometime).

As far as I know (I feel like I need this "disclaimer" on anything I post about LN) Lightning Network is trustless. The only thing someone can do, is broadcast transactions for on-chain settlement. I wonder (or worry) about the fee-implications that can have on a hub though.

I envision LN to be for payments what Bittorrent is for downloads: one client program that does whatever you tell it to do, connecting on it's own to as many hubs as needed, without the user having to worry (or understand) how it does this.
If this can be made into a simple GUI, I'll be very happy!

Oh, of course... that didn't occur to me. Channels connecting to channels, and anyone could propose to be a channel (and I guess, this means anyone could "appoint" payment hubs). So merchants could all sign up so a hub, perhaps set up by someone like Blockstream, through their Lighting Chain, I think I definitely could see this. In fact, if it succeeds even by a little, I've no doubt this will catch on quickly.


I kinda want to look for people doing this, and connect to them - as soon as I figure out how! I foresee a new way for people to build trust and reputation (and possibly lose it) through LN.

How is this different than coins in my wallet? I and only I decide when I spend them, so of course they're not accessible to anyone else.

It's still a bit of a mystery to me how this will work in all details. I can imagine the coffeeshop spending the coins again on another channel, keeping some balance and acting as a hub. Alternatively, the coffee shop can have it's customers connect through a hub. And if that hub is part of an exchange, the coffee shop can sell their revenue instantly.

Interesting for me to know as well, as I am looking at a few LN applications myself... but in your example and in my scenario, payment only goes one direction, so LN might not seem beneficial to one party (the recipient in your scenario and the sender in mine). Apologies in advance if my understanding of LN is flawed. Scarcity I believe is only an issue for as long as the channel is kept open - and open forever channels is unlikely, unless either party doesn't need the Bitcoin committed... in which case, they wouldn't be seeking extra Bitcoin outside the channel anyway.

1. If Tom only opens that channel between himself and the coffeeshop, presumably, the coffeeshop will want him to close the channel every week or every month, wouldn't they? Because only when the channel is closed can the final version of txs be broadcast to the blockchain, and only then can the coffeeshop "realise" the coins in their wallet. I don't see many small businesses willing to open channels with each customer - so it might not work very well in that direction (of the customer opening the channel).

Rather, a merchant might offer to open channels for customers for a small fee, perhaps like a form of subscription, i.e. pay for one-month, two-month, or even one year subscriptions, allowing instant settlement for small transactions.

Also, in my own scenario where I receive regular small payments from a fixed number of employers (say, in a freelance gig), I thought to offer them to open a channel with me to settle the regular payments and broadcast to finalise every month. Benefit for them: they save on tx fees, benefit for me, I get one input a month. But downside for them: they must estimate an amount to hold in the channel, and forces them almost to "pay upfront". It affects cashflow for them, and they are also small businesses that almost rely exclusively on cashflow.

Cashflow issue also works with Tom's scenario, since the coffeeshop might rely on daily cashflow to run their business (inventory maintenance, credit repayment etc.).

Be keeping a close eye on Blockstream now, curious to see how retailers are working out with Lightning Charge.

Let's say Tom funds a channel between himself and his favorite coffee shop with 1 BTC to buy coffee daily, but this coffee shop is local and has few customers using Bitcoin. That 1 BTC will decrease over years, but at the same time not really accessible to anyone else. So in a way, Tom is still hoarding that coin, because it is not for sale on exchanges.

If enough people buy coins to fund these "private" channels, then it will eventually decrease the overall supply of coins in the market, right?

If this is true, the price of bitcoins will go through the roof, because there will be less coins available and hopefully a increase in the demand.

This Lightning Network is going to be the cherry on the cake for this experimental technology. We just reached 100 nodes and 225 channels today!