Lightning Network Skepticism

[HostFat]

https://chrispacia.wordpress.com/2015/12/23/lightning-network-skepticism/

I like the idea of the lightning network. I think the developers behind it are very smart and it’s a very clever use of Bitcoin contracts. I’m sure that whatever comes from it will be useful to some parts of the Bitcoin ecosystem at a minimum. What I’m not for, however, is prematurely deprecating critical parts of bitcoin (zeroconf transactions and low transaction fees) when it’s not yet clear that the lightning network will be a viable replacement for them. Yet that’s what the core developers seems bent on doing.

We know that lightning is at least technically feasible. We don’t know if it’s economically feasible or even a desirable alternative. Will it be a decentralized peer-to-peer payment layer or will it end up as a quasi-centralized payment network similar modern banking? We don’t know the answer to this question and probably wont know until we see it in action. Which is why it comes across as irresponsible to go “all-in” on lightning at this point.

My main concern from the beginning was that, as a hub-and-spoke payment layer, there would be very few hubs and the network would be quasi-centralized and a regulatory sitting duck. It seems I wasn’t the only one with this concern as there’s been a fairly recent pivot away from the hub-and-spoke network topology to a more organic, wallet-to-wallet routing. The network is now envisioned as a more pure p2p payment layer without those large scale payment hubs.

Certainly this has to be viewed as a promising development as it begins to address my primary concern. Unfortunately, I find this view of the lightning network overly optimistic. In what follows I will give a lightning network overview and some reasons why I think it’s likely the network will end up with the hub-and-spoke topology anyway.

Continue

Great article.

[1714126790]

1714126790

[1714126790]

1714126790

[1714126790]

1714126790

[1714126790]

1714126790

[1714126790]

1714126790

[1714126790]

1714126790

[franky1]

firstly there is no such thing as quasi centralised... its either centralised (one point of failure) or decentralised.
there is however wildly distributed or confined distribution, which if the distribution of LN nodes is confined, it could possibly lead to collusion..

i personally dont see it as a ripple thing.. where everyone needs to log into a ripple owned server. to then trade with coinbase or bitstamp
i see it as a standalone executable that anyone can have. but where most of the big players such as bitpay, coinbase, etc will be the main channel operators, purely due to popularity of moving funds in and out of their services.

and as long as this executable still uses the same signature method and proof of value to ensure even they cant fiddle with other peoples funds then its all good.
if it turns out to just be a sql database saying | user1=0.1btc | user2=0.09btc | where the channel operator can fiddle.. then it should not be part of the bitcoin ecosystem

[agath]

firstly there is no such thing as quasi centralised... its either centralised (one point of failure) or decentralised.

The same is valid also for bitcoin with larger blocks. And still there is no evidence that larger blocks lead to centralization.

[Cubic Earth]

firstly there is no such thing as quasi centralised... its either centralised (one point of failure) or decentralised.

So what about two points of failure?

[franky1]

firstly there is no such thing as quasi centralised... its either centralised (one point of failure) or decentralised.

So what about two points of failure?

confined distribution

[pedrog]

What's the incentive for someone to run one of these hubs?

[LiteCoinGuy]

What's the incentive for someone to run one of these hubs?

i guess they could recieve a tiny amount of money too?

[chek2fire]

and is this offchain transactions safe. Is impossible someone to fake the system and make a double spent or worst a spent with bitcoins that didnt have?

[Cubic Earth]

and is this offchain transactions safe. Is impossible someone to fake the system and make a double spent or worst a spent with bitcoins that didnt have?

For lighting to be as trustless as running a full node is today you would still have to run full node.  To establish a channel, you would need to wait for the standard 6-confirms or longer to be protected against double spends.  Once the channel is open, you are protected against double spends by funds that held in what amounts to a multi-signature escrow account.  If the other side of the channel signs over some of the escorw funds to you, you can be be sure the funds are available have not been double spent.  How can you be so sure?  Because those funds don't go anywhere without your signature too.

[gentlemand]

Let us wait until it's in a semi operating condition before deciding what's what. Perhaps it'll never reach a working iteration or maybe it'll be what we always dreamed of and more.

[BlindMayorBitcorn]

Let us wait until it's in a semi operating condition before deciding what's what. Perhaps it'll never reach a working iteration or maybe it'll be what we always dreamed of and more.

Fine.

[Klestin]

Let us wait until it's in a semi operating condition before deciding what's what. Perhaps it'll never reach a working iteration or maybe it'll be what we always dreamed of and more.

That's OP's entire point.  It's too soon to start "deprecating critical parts of bitcoin (zeroconf transactions and low transaction fees)".

[gentlemand]

That's OP's entire point.  It's too soon to start "deprecating critical parts of bitcoin (zeroconf transactions and low transaction fees)".

But it doesn't exist yet. We've no real idea what the incentives or disincentives might be. We don't know whether it'll work in the way it was intended or in any other way. It'll be time to start questioning when we can see live examples and then we'll have a concept of its impact.

[chek2fire]

and is this offchain transactions safe. Is impossible someone to fake the system and make a double spent or worst a spent with bitcoins that didnt have?

For lighting to be as trustless as running a full node is today you would still have to run full node.  To establish a channel, you would need to wait for the standard 6-confirms or longer to be protected against double spends.  Once the channel is open, you are protected against double spends by funds that held in what amounts to a multi-signature escrow account.  If the other side of the channel signs over some of the escorw funds to you, you can be be sure the funds are available have not been double spent.  How can you be so sure?  Because those funds don't go anywhere without your signature too.

thx for the info

[knightkon]

That's OP's entire point.  It's too soon to start "deprecating critical parts of bitcoin (zeroconf transactions and low transaction fees)".

But it doesn't exist yet. We've no real idea what the incentives or disincentives might be. We don't know whether it'll work in the way it was intended or in any other way. It'll be time to start questioning when we can see live examples and then we'll have a concept of its impact.

I also agree that there is to much that is not known yet.  As quoted, we want to see some live examples.  Poop can be made to look good on paper, but take some in the real world and you then know that is smells.  Lets see some of this live and see how they hold up. 

[cellard]

I like that things get critisized and the weak points get pointed at because that's how we can improve, but has anyone noticed that everyone criticizing the LN never propose a better idea? (and centralizing nodes with huge blocks isn't a better idea).

[HostFat]

I like that things get critisized and the weak points get pointed at because that's how we can improve, but has anyone noticed that everyone criticizing the LN never propose a better idea? (and centralizing nodes with huge blocks isn't a better idea).

This is a claim without proof, it has been six years without the blocks being full, and even now miners aren't doing it every time because they try to avoid the orphan blocks.

The critics are that Bitcoin Core / Blockstream team is forcing users (by changing the rules of the mempool, RBF and small blocks) to go finding other solution on second layer network -> LN. Solutions still not developed and without years of tests as for the Bitcoin network.

There is anymore free choice. (anyway, it's a free choice if you think that paying $20 fee to make a tx on the Bitcoin network is good)

99.9% of the promoter of bigger blocks thinks that LN and what ever other solutions are great and welcome, but this until the user is free to choose between them and the Bitcoin network as was originally though.

[tokeweed]

Let us wait until it's in a semi operating condition before deciding what's what. Perhaps it'll never reach a working iteration or maybe it'll be what we always dreamed of and more.

This.  And who wrote that article?  

One group I see that could be threatened by LN is the industrial mining cartel.

[HostFat]

And who wrote that article?

He is a developer of the dangerous Open Bazaar cartel