Lightning network channels

[rbruggem]

Hi guys,
I've got a doubt that has been puzzling me for while and I hope some of you can shed some light on it.

The consensus is that the Lightning Network is going to solve bitcoin's scaling problem.
I do believe Lightning is an innovative solution and I do like off-chain solutions, but I have one issue that makes me think that Lightning will not be the solution to bitcoin's scaling problem.

Every day I make many transactions to buy goods and services at many different merchants. I usually pay by card. If I'd use Lightning I would have to create a peer-to-peer channel between me and the merchant, which entails funding transactions on the blockchain and settlement transactions on the blockchain. These are two on-chain transactions!
This process would have to be repeated for every purchase I make with any merchant. I therefore conclude that Lightning transactions coincide with onchain transactions.
The only way, from what I understand, that Lightning is going to solve the scaling problem, is that the channels stay open for a longer time. In my case, I can see this happening only when I do recurring payments, for instance Amazon, bills, coffee shop, food shopping. In this case though, transactions are usually in one direction, which means, once the funding transaction has been fully spent, the channel will need to be closed and another re-created.

The above cases lead me to think that even though Lightning is a solution to offload work from the blockchain, in practice though, since we need short lived channels, it won't happen.

Please confirm the validity of the thought or prove me wrong.
I really appreciate your answers.
Thanks
R

[Anti-Cen]

Hi guys,
I've got a doubt that has been puzzling me for while and I hope some of you can shed some light on it.

Hi

I read a paper that says the sticking plaster will not hold and it comes down to waiting for the end of the month to ensure the money is yours
so it will need a type of broker to lend you the money and no one is going to do that for free.

Lets face it the BTC protocol is broken and even with low volumes of trade the database is already at 200gb in size and having to parse
all that data chain to stop double spending needs a massive amount of processing power.

Block-Chain might be good from an accountants perspective but its a house of horror from a developers point of view.

IOTA Tangle might be a better way to go 

[rbruggem]

Please reply with serious answers.

[Anti-Cen]

Please reply with serious answers.

Maybe you need to read this
https://news.bitcoin.com/mathematical-proof-may-show-lightning-network-is-heavily-centralized/

You can find many more too so cut the insults when someone is trying to help you

[HeRetiK]

You don't need to open / close a new lightning channel everytime you make a transaction.

Single bi-directional channels are just one part of the equation. The other is that these bi-directional channels will enable routing between multiple parties, meaning you can use the same channel to pay your local coffeeshop, Amazon or the buddy that you split your lunch tab with. So even if it's just a one off payment, you can still use the same lightning channel you use for regular purchases as long as a payment route is available.

In the end you only need to pay transaction fees for opening and for closing the channel, regardless of the amount that gets transacted for the duration the channel is open. Whether that duration will be one day, one week or one month will be up to you and the lightning client you will be using.

This article series by Bitcoin Magazine goes more into detail, especially the second part about building the network topology should answer some of your questions.

https://bitcoinmagazine.com/articles/understanding-the-lightning-network-part-building-a-bidirectional-payment-channel-1464710791/

[...]

IOTA Tangle might be a better way to go 

The Tangle is an interesting concept, but I'm highly skeptical of IOTA as long as it requires a central authority to protect itself from double spends. As far as I know even IOTA's tokens are centrally issued, but that might be a misconception of mine that I'll gladly have debunked by someone with more insight on IOTAs architecture.

Maybe you need to read this
https://news.bitcoin.com/mathematical-proof-may-show-lightning-network-is-heavily-centralized/

The original post this article refers to is based on faulty assumptions, making the mathematical proof a pointless one.

[cupronickel]

A couple of good higher level videos I came across while researching this may help those who are puzzled by the basic concepts:

https://www.youtube.com/watch?v=rrr_zPmEiME short animation with Bob pays Carol etc

https://www.youtube.com/watch?v=3PcR4HWJnkY talk by Elizabeth Stark, CEO/co-founder LN

[hatshepsut93]

You don't need to open / close a new lightning channel everytime you make a transaction.

Single bi-directional channels are just one part of the equation. The other is that these bi-directional channels will enable routing between multiple parties, meaning you can use the same channel to pay your local coffeeshop, Amazon or the buddy that you split your lunch tab with. So even if it's just a one off payment, you can still use the same lightning channel you use for regular purchases as long as a payment route is available.

In the end you only need to pay transaction fees for opening and for closing the channel, regardless of the amount that gets transacted for the duration the channel is open. Whether that duration will be one day, one week or one month will be up to you and the lightning client you will be using.

This article series by Bitcoin Magazine goes more into detail, especially the second part about building the network topology should answer some of your questions.

https://bitcoinmagazine.com/articles/understanding-the-lightning-network-part-building-a-bidirectional-payment-channel-1464710791/

This is a correct answer, and anyone who is interested in it can already try the Lightning Network on testnet.

https://htlc.me/ is a custodial online wallet, it already comes with opened channels, so you can start sending and receiving payments right away.

ACINQ has released their Eclair wallet for Android which is available on Play Store, or you can get their desktop wallet on github: https://github.com/ACINQ/eclair

They have also released Lightning Explorer which is available at https://explorer.acinq.co/#/

To test payments, you can visit:

https://starblocks.acinq.co/#/

https://yalls.org/

https://lightning.bitrefill.com

[Anti-Cen]

The Tangle is an interesting concept, but I'm highly skeptical of IOTA as long as it requires a central authority to protect itself from double spends. As far as I know even IOTA's tokens are centrally issued, but that might be a misconception of mine that I'll gladly have debunked by someone with more insight on IOTAs architecture.

No you could well be right but we are already more or less forced to trust some 3rd part like Coinbase as it is
so i could trust some type of decentralized central servers to stamp coins somehow without owning them.

I don't mind helping to authenticate two transactions to keep my transaction free but i cannot afford to
keep a copy of a 200gb block-chain on my little PC and to then process all the history for a coin.

Must be a way of using some type of digital stamping of autonomous coins/tokens without then having to be owned
by brokers using a private key pair but i do think we need to separate the value of the coins from the infrastructure
because any currency needs a stable prices regardless of how well the infrastructure is doing.

You can keep adding advancements to a formula one sports car but there comes a time where you have to scrap it and stat again
or end up out of the race is my thinking and we know by now we are heading for trouble anyway.

Just now i think more people would be happier if fees were based on value and not byte size and if the public do arrive en-mass
then they will be dipping toes in the water with $100 and they will not take to $20 transaction fees i don't think
  

[FAB_coin]

 LN and both Plasma are to improve the scalability tho can they really be centralized, unfortunately answer is no. you might wanna read this.

https://medium.com/@jonaldfyookball/mathematical-proof-that-the-lightning-network-cannot-be-a-decentralized-bitcoin-scaling-solution-1b8147650800

[maxxdxx]

I have a question too. Since opening the payment channel requires an on chain transaction does that incur its own set of fees? If I walk into a coffee shop and want to buy a coffee but don't have a payment channel open yet doesn't that mean I could have to wait minutes/hours before the payment channel is ready?

[HeRetiK]

I have a question too. Since opening the payment channel requires an on chain transaction does that incur its own set of fees? If I walk into a coffee shop and want to buy a coffee but don't have a payment channel open yet doesn't that mean I could have to wait minutes/hours before the payment channel is ready?

The on-chain transaction to open a payment channel is treated like any other Bitcoin transaction, so the fees are pretty much the same -- except that it can get used for multiple transactions instead of just a single one. So yes, it would probably make sense to have a payment channel ready before entering the coffee shop.

Taking transactions off-chain is likely to free up blockspace however, so 0-conf transactions may become viable again. However I'm not sure what possible attack vectors would come with accepting 0-conf transactions for opening a payment channel.

[cupronickel]

it would probably make sense to have a payment channel ready before entering the coffee shop.

Worth pointing out you don't need to open a new direct payment channel. If there's enough critical mass in the network and as long as both parties have pre-existing channels (other retailers, friends whatever), then a route between can be found.

[Kprawn]

You also need to consider that on-chain tx's would be cheaper. I presume that you could still do direct payment on-chain for

single once off transactions, if you wanted to. The network will be less congested, because most micro payments will be done

off-chain. This will make single once off transactions much cheaper. {Can someone verify if I understand this correctly?} IF

single on-chain tx's is not possible, then I am with you on that.  

Edit : It is also likely that you can piggy back off an existing payment channel that are open. 

[HeRetiK]

it would probably make sense to have a payment channel ready before entering the coffee shop.

Worth pointing out you don't need to open a new direct payment channel. If there's enough critical mass in the network and as long as both parties have pre-existing channels (other retailers, friends whatever), then a route between can be found.

Yes of course. That's one very important point that often gets overlooked when arguing about LN's viability.

You also need to consider that on-chain tx's would be cheaper. I presume that you could still do direct payment on-chain for

single once off transactions, if you wanted to. The network will be less congested, because most micro payments will be done

off-chain. This will make single once off transactions much cheaper. {Can someone verify if I understand this correctly?} IF

single on-chain tx's is not possible, then I am with you on that. 

Single on-chain transaction will still be possible, just as before. It's not a change of the Bitcoin protocol, just an addition on top using existing Bitcoin scripting capabilities

While transaction throughput will be heavily increased with LN's increasing adoption, single transactions will likely still stay relatively expensive, at least long term in a scenario where a huge part of the global population transacts money using Bitcoin. Short to mid-term I could very well imagine single on-chain transactions becoming cheaper again, just not as cheap as LN transactions, naturally.

Edit : It is also likely that you can piggy back off an existing payment channel that are open. 

As mentioned above, yes.