Lightning Network Channels

[hugeblack]

I have basic information about Lightning Network :
the chains can support the same cryptographic hash function, it is possible to make transactions across blockchains without trust in 3rd party custodians(not broadcast to the blockchain).[1]
people can fund them channels (like fees) which can change constantly.

what will happens if a lot of people fund them channels at the same time?.
Will back to the same problem (high fees)?.
also, creating large payment channels can support decentralization?

[sjefdeklerk]

I have basic information about Lightning Network :
the chains can support the same cryptographic hash function, it is possible to make transactions across blockchains without trust in 3rd party custodians(not broadcast to the blockchain).[1]
people can fund them channels (like fees) which can change constantly.
what will happens if a lot of people fund them channels at the same time?.
Will back to the same problem (high fees)?.
also, creating large payment channels can support decentralization?

I'm pretty sure that every funding transaction is on chain. It's simply a transaction between a user and a shared BTC address (the channel). So if 100 users fund a channel that's 100 on-chain transactions. I don't see any other way this could happen.

[patt0]

I'm pretty sure that every funding transaction is on chain. It's simply a transaction between a user and a shared BTC address (the channel). So if 100 users fund a channel that's 100 on-chain transactions. I don't see any other way this could happen.

I also have just basic knowledge of how the lightning network works, but I have the impression that the transactions are only recorded on-chain when the channel is closed, and not when it's funded, but I'm not sure.

As for the question "creating large payment channels can support decentralization?" I would say it's the opposite. If a large hub is creating, we would almost have a service that would work as a bank. A person or a company would have an incentive to create a large hub that is well funded and well connected since they would be getting the fees, and users would have an incentive to use that channel because they wouldn't need to keep opening channel and paying for it, since they could to a channel that is well connected.

This would not be good, and is a critic for the LN, but it probably wont happen. You can get an explanation on why it wont happen on this video by Andreas
https://www.youtube.com/watch?v=D-nKuInDq6g

Basically he explains that those hubs will be targeted by hacks, so it would not be wise to try to create them. Also, if the network is designed to expand by itself, so after a while there won't probably be any incentive for users to search for these hubs, so they probably wont be created.

[sjefdeklerk]

I also have just basic knowledge of how the lightning network works, but I have the impression that the transactions are only recorded on-chain when the channel is closed, and not when it's funded, but I'm not sure.

No, also the funding is registered.

"In this example, while there was a total of seven transactions, only three of them were recorded on the Litecoin blockchain (the two initial 2.5 LTC deposits (one from each person), and one final withdrawal from the escrow account of 5 LTC to Jane)."
https://litecoinalliance.org/lightning-network-explained/

To help you understand this all you need to realize is that a payment channel is just a (Bitcoin) address. It's a special type though ('multisig') but it's still a BTC address. So if you fund it, the blockchain registers money going from your own BTC wallet to that multisig BTC address. After that you can make offchain transactions and when you're ready, the money gets moved out of that BTC address again into your own address.

It's the easiest to understand when you compare a payment channel to an escrow account where 2 (or more) parties put money in, which gets released after a specific time, but during this time, the parties can agree to change the ratio to whom the money belongs. So if we start with 50/50 ratio, we can agree that after a transaction now 30% of the 'escrow account' belongs to me and 70% belongs to you. That's really all a payment channel is.

[vit05]

  I was still thinking that Lightning Network would be used primarily by exchanges. I have never thought about that as an escrow account.
But it makes sense see that way.
  So it created the possibility of some institutions starts to work for profit providing large hubs. This is interesting.

[Carlton Banks]

So it created the possibility of some institutions starts to work for profit providing large hubs. This is interesting.

The costs of running a Lightning channel are essentially zero. That makes it very difficult to make any kind of profit from Lightining channels. There's more routing potential in multiple independent nodes than there is in one node with lots of channels, so it makes no sense from a business perspective or a technical perspective.

[sjefdeklerk]

The costs of running a Lightning channel are essentially zero.

No, the cost of running a lightning channel is 1x mining fee THE LEAST (for funding the channel). More realistically it's 2x mining fee (opening and closing), or more, depending on the number of funders.

That makes it very difficult to make any kind of profit from Lightining channels.

If no profit could be made, then it would fail for sure. Why would you open your channel for routing if you can't make money on it? Nobody would open his channel for routing, so the project would fail from the start. It all DEPENDS on people making money routing other people's payments.

[pebwindkraft]

If no profit could be made, then it would fail for sure. Why would you open your channel for routing if you can't make money on it? Nobody would open his channel for routing, so the project would fail from the start. It all DEPENDS on people making money routing other people's payments.

This is one of the repeating fallacies in this world (and this forum) - stipulating that success of a topic can only be achieved when you can make money with it. I will get hard to explain then, how this whole eco system developed, and why there is Open Source software... or even help aid or other social activities benefitting the society. It is definitely a severe limitation of imagination, posting such statements.
When it comes to "NOBODY" - well, maybe I'm a nobody  : I AM running a full node, without having benefits, and I know some other people doing the same. And yes, we want to run lightning nodes, of course! Without making money. And yes, I WILL open channels for routing (so the project will not fail, and get a helping hand to start).
What is it, that these statements reappear all the time? Is it this capitalistic paranoia, that communism is already ruling the world, and free speech and free travel and personal privacy is already gone? I would think so... Maybe this is, why we have red stripes in the US flag...

[Anti-Cen]

The costs of running a Lightning channel are essentially zero. That makes it very difficult to make any kind of profit from Lightining channels. There's more routing potential in multiple independent nodes than there is in one node with lots of channels, so it makes no sense from a business perspective or a technical perspective.

Sorry but I cannot agree and you ran off last time when we had this debate

Each channel costs in miners fees and the ledgers needs the money to operate  and the killer is
it works like snooker balls across the route and not as if liquid is being sent down a pipe

I cannot pass your $500 to Alice because you are in credit with me on one channel unless the channel
opened between Alice and myself contains $500 of my money

Money cannot just be moved from one private ledger to another because it could be counterfeit money 

please see white paper and point out if you think my statement is not correct
https://lightning.network/lightning-network-paper.pdf

We have already seen what an open market has turned into with transaction fees here so
"from a business perspective AND a technical perspective" it makes perfect
sense to charge transaction fees and interest on the money for the service you provide.

Some of these transactions take days to be confirmed which only adds to the trouble because
it's not like a end to end TCP connection and is more like UDP packet are being sent

[Carlton Banks]

The costs of running a Lightning channel are essentially zero.

No, the cost of running a lightning channel is 1x mining fee THE LEAST (for funding the channel). More realistically it's 2x mining fee (opening and closing), or more, depending on the number of funders.

That makes it very difficult to make any kind of profit from Lightining channels.

If no profit could be made, then it would fail for sure. Why would you open your channel for routing if you can't make money on it? Nobody would open his channel for routing, so the project would fail from the start. It all DEPENDS on people making money routing other people's payments.

The simple answer is that people are opening Lightning channels to save money, not to make money.

Opening a channel costs very little when you consider that it can be kept open indefinitely. The amount that can be saved in on-chain fees is pretty large, as long as you're both spending & receiving money (i.e. getting a paycheck over Lightning)

[Anti-Cen]

The simple answer is that people are opening Lightning channels to save money, not to make money.

It's far from being live so you must be talking about on testnet here and Bitcoin itself was on that for years

Bankers are always out to make money or did you forget that for all practical purposes we need banking hubs
and you admit it in your next statement anyway with the word "indefinitely" so unless you have a friend that is always
on-line, has the bandwidth and  BTC to forward on your payments then can we just call them banks please.

Opening a channel costs very little when you consider that it can be kept open indefinitely. The amount that can be saved in on-chain fees is pretty large, as long as you're both spending & receiving money (i.e. getting a paycheck over Lightning)

Yes they first talked about days, then weeks and now years just like you keep an account open
for years with a regular bank right

Me and you could open a bidirectional channel and play ping-pong all night long and save $10,000 in fees or more
but when we close the channel in the morning we are left with a hang over of $30 in miners fees.

Now I agree $0.01 transactions should move "Off-Block" and to be honest I don't quite know how this
should be done but $10 transaction just like in other system should remain "on-block"

Remember that old "Virtually free transaction fees" from the original white paper because they
are trying to pull that one again with hub/banking fees so why are you protecting them ?

[sjefdeklerk]

Opening a channel costs very little when you consider that it can be kept open indefinitely. The amount that can be saved in on-chain fees is pretty large, as long as you're both spending & receiving money (i.e. getting a paycheck over Lightning)

That's nonsense. It costs very much because the mining fees are very high. And who wants to lock up money in a channel he's not using, for a long time?

[Carlton Banks]

Opening a channel costs very little when you consider that it can be kept open indefinitely. The amount that can be saved in on-chain fees is pretty large, as long as you're both spending & receiving money (i.e. getting a paycheck over Lightning)

That's nonsense. It costs very much because the mining fees are very high. And who wants to lock up money in a channel he's not using, for a long time?

The idea is that you open a channel to use it, not so you don't use it. Then you can make many transactions with other people for tiny fees. And mining fees aren't high if you're willing to wait. You could set up a channel, using a low fee, wait a while till it confirms. Once it does, you have instant transactions (no more confirmation waits) and you'd be very happy with the overall price of transaction fees you could be paying.

This is not for beginners though, I strongly advise against using Lightning for the moment (as do the developers of Lightning clients & protocol)

[Anti-Cen]

This is not for beginners though, I strongly advise against using Lightning for the moment (as do the developers of Lightning clients & protocol)

Lightning  not live, won't be for estimates of between six and nine months so you must be talking about test-net

See https://github.com/bitcoin/bips/blob/master/bip-0173.mediawiki
Yes i can agree about having to break existing code but not about who pays to convert the wallet

The idea is that you open a channel to use it, not so you don't use it. Then you can make many transactions with other people for tiny fees. And mining fees aren't high if you're willing to wait. You could set up a channel, using a low fee, wait a while till it confirms. Once it does, you have instant transactions (no more confirmation waits) and you'd be very happy with the overall price of transaction fees you could be paying.

So are you going to ever admit that the Alice > Box > Peter and Paul case is theoretical and that we are dealing with major banking hubs here ?

Getting money out as cash to spend on the streets as is the case for "Cam Girls"  mean the poor old girl will need to close the
channel each week so that's $30 gone in miner fees and will force the girls back on the game

You can argue that something that charges fees, locks your money, pays to maintain lots of channel, has counter party
risk is not a bank but can you not at least force yourself to accept that we have centralized hubs ?

[hv_]

So it created the possibility of some institutions starts to work for profit providing large hubs. This is interesting.

The costs of running a Lightning channel are essentially zero. That makes it very difficult to make any kind of profit from Lightining channels. There's more routing potential in multiple independent nodes than there is in one node with lots of channels, so it makes no sense from a business perspective or a technical perspective.

Legal aspects?  You might get regulated as a payment processor....

[Anti-Cen]

Legal aspects?  You might get regulated as a payment processor....

Yes very good point

if the banks take up front fees which I think we can safely assume  they do or else you could be in credit, broadcast to close the channel
and they get hit with $30 transaction fees

if you look at the white paper the process of being in dispute and how this get resolved is quite complicated so was a bank
to go down then this would create so many transactions that it would grid the Bitcoin network to a Holt I would imagine.

I have tried to contact Joseph Poon who is leading the lightning project about major concerns I have and
it looks like he does not want to speak or address any of the points I have raised in the email

Here is the white-paper if your interested in the nuts and bolts of the project
https://lightning.network/lightning-network-paper.pdf

Key here is that money does not move between the many private ledgers held by the bank
and only the entries created from the initial deposit can be changed within the ledger which from
the legality side looks about right but from a clients side means that the bank ends up effectively
lending you your own money when you send it until the channel is closed and settlement happens
on the main block chain

[hv_]

Legal aspects?  You might get regulated as a payment processor....

Yes very good point

if the banks take up front fees which I think we can safely assume  they do or else you could be in credit, broadcast to close the channel
and they get hit with $30 transaction fees

if you look at the white paper the process of being in dispute and how this get resolved is quite complicated so was a bank
to go down then this would create so many transactions that it would grid the Bitcoin network to a Holt I would imagine.

I have tried to contact Joseph Poon who is leading the lightning project about major concerns I have and
it looks like he does not want to speak or address any of the points I have raised in the email

Here is the white-paper if your interested in the nuts and bolts of the project
https://lightning.network/lightning-network-paper.pdf

Key here is that money does not move between the many private ledgers held by the bank
and only the entries created from the initial deposit can be changed within the ledger which from
the legality side looks about right but from a clients side means that the bank ends up effectively
lending you your own money when you send it until the channel is closed and settlement happens
on the main block chain

Thx to CSW twitter:

https://www.law.cornell.edu/uscode/text/18/1960

18 U.S. Code § 1960 - Prohibition of unlicensed money transmitting businesses

[cloud.runner]

Lightning Network maybe centralize, and is opposed to the Bitcoin decentralized design. I hope the bitcoin core team can implement it as decentralized.

[Anti-Cen]

Lightning Network maybe centralize, and is opposed to the Bitcoin decentralized design. I hope the bitcoin core team can implement it as decentralized.

We are being told that wallets can become hubs so if you only open a channel to my wallet/hub and I go off line then as
far as your concerned the network has become centralized and your out of luck even if you can broadcast a message on
the BTC block-chain to get the money back that I owe you.

if big banking hubs cannot charge fees to interconnect the banking hubs then we don't have a network unless
everyone agrees to open three channels each, finance them and keep them open and we can get transactions to
work across routes with 100 nodes or more

The people trying to defend LN won't even accept that it must contain major banking hubs for it to work
so they must be miners hoping to run these banks

[Xynerise]

Lightning Network maybe centralize, and is opposed to the Bitcoin decentralized design. I hope the bitcoin core team can implement it as decentralized.

It is NOT centralised.
But, even if it was, it doesn't matter, as long as the base layer is decentralised.
That's unarguably the most important thing.

If you feel LN is centralised then you DON'T have to use it and can just use the base layer that is decentralised.
But if the base layer is not decentralised then there's pretty much no hope; every transaction you make (on any layer ) will be centralised by default since they take their function from layer 1.