A critique on the Lightning Network from a regular poster. Long.

[Wind_FURY]

I received this private message from someone who does not want to be identified. He gave me permission to post a new thread about it and start an unbiased discussion, and to avoid the post to be called FUD just because it came from "him".

achow? I believe you have the first right to reply. But if someone else can make a good discussion, go ahead, be my guest.

I did not want to meander the coinbase revoke wikileaks thread. but i seen your post and it made me want to respond.

Bitcoin was developed to eliminate the middleman, to transact directly with each other over the internet. If you use Coinbase for storage then you are doing it wrong.

They want Paypal 2.0. We want a P2P, censorship resistant, decentralized cryptocurrency.

Now you might start to see my point about LN's flaws.  LN Hubs is banking/paypal2.0
also.
 
Core is also the centralised regulator of the rules. they set the policy. change the rules and nodes(merchants) are suppose to follow cores set rules and reject(ban hammer) users(nodes) that send dodgy transactions/blocks

This is why i detest what bitcores network has turned into.

It used to be the case that anyone can alter their node and then promote their own new feature. and if enough users liked it they would add it to thier node and then the network would evolve. now its core making that choice and anything not core gets sidelined as a altcoin.
core have become the regulators. and now nodes are not the regulators but the regulated 'be your own bank' wallet that has to ban/reject payments based on cores policy

Anyway back to LN

LN is not a direct pay the recipiant(P2P). its in laymans terms opening a channel(account) with a well routed(bank branch) hub. depositing funds into a channel(account) that requires co-signing(bank authorisation)

If you imagine a hub of say 8000 users(channels) or a spoke of one user routing 8000 users 'payments' to one merchant. then that is the technicals of a local bank branch. all with the very same banking features of 3-5 business day settlemnt (CLTV) and wire transfer chargebacks(CSV)

Many people think banks are one server one location one entity. but banks are local bank branches each with just 8-20,000 customers per branch
(take the UK, 8000 bank branches.. equals ~8125 people per bank branch (65m population))

This is where the propaganda myth of 'visa transaction capacity' hides the truth. visa does not store everyones balance on one server. visa communicates to local branchs using the customers account numbers (channel address) and bank branch routing/sort code(node URI)

If you still cannot see that LN is just banking 2.0 where hubs that control authorisation will control the network. then you need to take a deeper look beyond the glossy advertising of LN

There are only ~1500 LN nodes. but already there are hubs being the middlemen for ~10% of the network


The argument then becomes.

Do you let there be 10 middle men having the authorisation power of the whole network. so that most people can make a payment only needing consent from maybe 2-10 middle men 'managers'(banks/hubs) to route
or break it down so that hubs can only have 1% of total channels. meaning more middle men are needed between you and the end destination where it would need the consent/authorisation of 2-100 middle men managers(banks/hubs)

Either way LN is not a be your own bank, LN is set up a joint account with someone
LN is not P2P its bank branches and middle men

[1714128852]

1714128852

[1714128852]

1714128852

[1714128852]

1714128852

[1714128852]

1714128852

[1714128852]

1714128852

[TheQuin]

It appears that someone is obsessed with calling things banks when what they do bears no resemblance to a bank.

LN is not a direct pay the recipiant(P2P). its in laymans terms opening a channel(account) with a well routed(bank branch) hub. depositing funds into a channel(account) that requires co-signing(bank authorisation)

As defined by The Oxford English Dictionary a bank is "A financial establishment that uses money deposited by customers for investment, pays it out when required, makes loans at interest, and exchanges currency."

The other important factor that defines a bank is that it must be licensed to operate by a government.

Anyone can set up a Lightning Node and if enough people freely choose to open channels to it then it could be considered to be a hub. LN is entirely in keeping with the philosophy of allowing ordinary people to make financial transactions without the need for a trusted 3rd party. There are third parties involved but you are still in control of your coins. Nobody can use your coins for investment and they are not making any loans with them.

Edit.
Having reread the PM you published it is not so much FUD as misinformed paranoia. It is obvious the author has very little understanding of LN, the difference between a bank and a payment processor or how the dynamics of competition for fees will prevent a small number of hubs dominating.

[aleksej996]

I don't think it is fair to call it paranoia, as it is very reasonable to be suspicious of something like LN.
I also have a bit of a problem in the way Core handled this. Users should have a choice to either use LN or not, if we don't do onchain scaling, they will not have a choice. I have a problem with this.

LN however is far more decentralized than banks, as TheQuin says.
I think it is unreasonable to assume that without hubs you will need 100s of relay nodes to route your transaction. If people simply open channels with their friends, I think around 3 hops on average would be enough, according to a study done by Facebook https://research.fb.com/three-and-a-half-degrees-of-separation/

This would mean that everyone would need to open hundreds of channels with their friends, which is a problem.
We would need to make running a LN node very easy and cheap, like a simple mobile app and make sure people are usually online to route payments.
Technology can work, the question is in adoption, kinda like Bitcoin in general.

It is still reasonable to worry however. This does introduce some scary new things that we simply can't know how will work out in a long run.
LN does offer more good features than just scaling though, privacy of transactions, revenue for nodes which might increase decentralization, etc.
It is definitely something that needs to be discussed and chasing people away with different opinion is a really unreasonable and immoral idea.

[TheQuin]

I don't think it is fair to call it paranoia, as it is very reasonable to be suspicious of something like LN.

It's reasonable to be suspicious but the quote provided in the OP goes well beyond that IMO.

I also have a bit of a problem in the way Core handled this. Users should have a choice to either use LN or not, if we don't do onchain scaling, they will not have a choice. I have a problem with this.

While I agree that I would have liked to have seen a compromise approach with some on chain scaling as well, I do think that framing it as LN or onchain misses the bigger picture. LN is being developed but the blockchain has been opened up to layer 2 applications not limited to LN. If there are limitations in LN then there is nothing to stop people developing better alternatives.

LN, however, is far more decentralized than banks, as TheQuin says.
I think it is unreasonable to assume that without hubs you will need 100s of relay nodes to route your transaction. If people simply open channels with their friends, I think around 3 hops on average would be enough, according to a study done by Facebook https://research.fb.com/three-and-a-half-degrees-of-separation/

This would mean that everyone would need to open hundreds of channels with their friends, which is a problem.
We would need to make running a LN node very easy and cheap, like a simple mobile app and make sure people are usually online to route payments.
Technology can work, the question is in adoption, kinda like Bitcoin in general.

That again is overlooking the competition in the fee market. If it is profitable to run a hub then more people will run hubs and there will always be cheap paths. It's not just more decentralised than banks it is open market economics.

It is still reasonable to worry however. This does introduce some scary new things that we simply can't know how will work out in a long run.
LN does offer more good features than just scaling though, privacy of transactions, revenue for nodes which might increase decentralization, etc.
It is definitely something that needs to be discussed and chasing people away with different opinion is a really unreasonable and immoral idea.

Nobody is being chased away but pointing out the use of the pejoritive term banks when it is completely inappropriate is necessary, otherwise people will be worried and scared for no reason. Some might even find that an unreasonable and immoral idea.

[Samarkand]

The major argument against the Lightning Network in my opinion is still the fact that
nobody really needs a Lightning Network when Bitcoin´s main use case is being a
store of value. If you want to preserve your wealth from inflation, confiscation and other
comparable threats you can buy Bitcoin. However, you probably won´t touch your Bitcoins
for years if that is your reason for buying Bitcoin and therefore you are not the target group for the
LN.  

Besides, nearly everyone, who bought something using his Bitcoins over the years has regretted
that after the fact, because BTC only goes up in the long-term. If you believe this will be the case
going forward it is pretty stupid to buy something using a Lightning Network transaction when
the opportunity costs of buying anything with your precious Bitcoins are so tremendous.

I´m a strong believer that Bitcoin´s biggest advantages are rather in its unique characteristics
that make it a form of "sound money" than in its use as a payment system for small value
purchases or transactions (which is the major use case of the Lightning Network).

The great thing about the Lightning Network is that it is non-mandatory and anyone can still
decide to use the main layer for his transactions even if the LN becomes a giant success (which
it will not become in my opinion). This is the beauty of 2nd layer solutions in general,
they can offer new features while maintaining the trust and integrity of the main layer.

[achow101]

Core is also the centralised regulator of the rules. they set the policy. change the rules and nodes(merchants) are suppose to follow cores set rules and reject(ban hammer) users(nodes) that send dodgy transactions/blocks

Except Core does not have any sort of centralized authority to force any rules to go through. Otherwise you would see all of the forks that the Core developers propose to activate within a few weeks of the software supporting the fork being available. Case in point, segwit. Segwit took several months to activate, way longer than any previous soft fork. It took people threatening to perform a user activated soft fork before miners finally activated it. The UASF was not something that was officially supported by Core (i.e. no Bitcoin Core release containing UASF logic). If the Core developers had centralized control over the consensus rules, then segwit would have activated quickly and without contention, which is clearly not what happenend.

It used to be the case that anyone can alter their node and then promote their own new feature. and if enough users liked it they would add it to thier node and then the network would evolve.

This is still the case. You can do that if you want. No one is preventing you from creating the next Bitcoin XT or Bitcoin Classic or Bitcoin Unlimited. No one can stop you from writing the software and running your own node. Just because the vast majority of users decided to stick with Core does not mean that Core can prevent you from doing this or that Core is forcing you do follow them.

LN is not a direct pay the recipiant(P2P).

It certainly is. You can open a payment channel directly with your recipient and avoid having to route your payment through anyone else.

its in laymans terms opening a channel(account) with a well routed(bank branch) hub. depositing funds into a channel(account) that requires co-signing(bank authorisation)

It can be used in that way, but that is not the only way to use LN.

all with the very same banking features of 3-5 business day settlemnt (CLTV

That is completely false. There is no 3-5 business day settlement. That is not what CLTV is used for. It is a timeout for if a transaction fails. Not a "transaction goes through after X time".

) and wire transfer chargebacks(CSV)

That is completely false. That is not what CSV is used for. It is used as a timeout during which time a punishment can occur which is not the same as a chargeback. Once an HTLC is resolved or once new commitment transactions are put in place, the payment is final. There is not charging back. Trying to perform a chargeback would mean that you are broadcasting an old commitment transaction which means that you LOSE ALL of your money, not you gaining back whatever money that you sent.

There are only ~1500 LN nodes.

It's still in beta, what do you expect?

but already there are hubs being the middlemen for ~10% of the network

You seem to be ignoring the criss-cross of gray lines for all of the other channels.

The "hubs" currently on mainnet are due to people opening direct channels with merchants (i.e. performing direct P2P payments).

Do you let there be 10 middle men having the authorisation power of the whole network. so that most people can make a payment only needing consent from maybe 2-10 middle men 'managers'(banks/hubs) to route
or break it down so that hubs can only have 1% of total channels. meaning more middle men are needed between you and the end destination where it would need the consent/authorisation of 2-100 middle men managers(banks/hubs)

None of the above, because nothing stops you from closing your own channel unilaterally and then opening a channel with someone else directly to avoid these middlemen.

You don't seem to understand how LN works. These hubs can't force you to do anything. You have full control over your money and you can close a channel any time you want. If you want to avoid hubs, then open a channel directly with the person you are paying. No one is forcing you to have channels with hubs nor is anyone forcing you to use LN anyways.

[RGBKey]

With the banking system:

• Your payments must be routed through a bank
• Your bank requires you to identify yourself fully before making an account

Neither of these apply to the LN. It will probably be more convenient to route payments through a well-connected node. That does not make it a bank. You can't send an ACH transfer through your buddy Mike, but you could route a lightning payment through him.

[franky1]

Core is also the centralised regulator of the rules. they set the policy. change the rules and nodes(merchants) are suppose to follow cores set rules and reject(ban hammer) users(nodes) that send dodgy transactions/blocks

Except Core does not have any sort of centralized authority to force any rules to go through. Otherwise you would see all of the forks that the Core developers propose to activate within a few weeks of the software supporting the fork being available. Case in point, segwit. Segwit took several months to activate, way longer than any previous soft fork. It took people threatening to perform a user activated soft fork before miners finally activated it. The UASF was not something that was officially supported by Core (i.e. no Bitcoin Core release containing UASF logic). If the Core developers had centralized control over the consensus rules, then segwit would have activated quickly and without contention, which is clearly not what happenend.

It used to be the case that anyone can alter their node and then promote their own new feature. and if enough users liked it they would add it to thier node and then the network would evolve.

This is still the case. You can do that if you want. No one is preventing you from creating the next Bitcoin XT or Bitcoin Classic or Bitcoin Unlimited. No one can stop you from writing the software and running your own node. Just because the vast majority of users decided to stick with Core does not mean that Core can prevent you from doing this or that Core is forcing you do follow them.

lets just clear up achows101 core defense

1. segwit only had 35%.. based on bitcoins consensus.
core team got mad.. the main devs wouldnt have got paid their next tranche of investor funding if segwit got opposed.. so they employed samson mow(UASF) and bloq:-jGarzic(bitcoinABC) to do something that goes totally against consensus. which is called a bilateral split. as proposed by gmaxwell.

firstly. making core able to bilaterally split frrom its opposition by deception.. getting the propaganda rumours of everyone to upgrade their nodes to either core or bitcoin ABC (false election as it was rigged to avoid consensus)
problem is .. bloq is actually team core
(same investor as blockstream)(https://dcg.co/portfolio/#b)

then all those that opposed core, got pushed over to an altcoin. making cores bip jump from 35% to 95%, it reminds me of apartheid(race segregation) of the opposition to rig elections
apartheid:
core(white people) own a bus company. pretend there are seats for everyone..
"its ok, this bus is for everyone".. but
if you dont like core(white people), sit at the back of the bus.. core(white) people dont want you upfront anyway
....... then
shoot the people sat in the back of the bus or push them off the bus
 
now the front of the bus with the engine is only core(white people) so only white people get to the destination(polling station) and the back of the bus doesnt count..

so yes core does have control and now core can slide in as many soft forks as they like.. take the newest addition.. bc1q addresses..
this new address type did not require consensus or even bilateral split to win by default to add these new addresses. they just added it because now they have a back door to do what they like without consensus
Luke JR's backdoor

even things like Knots is another core defending node client. thus the community do not have a free choice.
if you check through history. XT classic and the others that opposed core all got REKT due to cores control.

anything oposing cores roadmap is always treated like an attack, not fair competition.. the core team do not take it as an honourable competition of consensus but scream out its an attack against "bitcoin" because they feel that core own "bitcoin". thus anything opposing cores roadmap should be thrown into an altcoin

even R.ver's reverse psychology is proving that. core absolutely hate of anyone saying that cores network is not bitcoin.. yet NO ONE should own "bitcoin".. and thats the point of roger vers subtle reverse psychology. by making the core defenders literally scream that core own bitcoin and anything else is just an altcoin.

fiat analogy:
dollar is not owned by anyone... australia can use it, canada can use it. america can use it
r.vers subtlety "i australia own real dollars"
core defenders "we america own dollars, anything else is a fake/fraud/misleading people"

thus its like america saying australia does not exist, australia is a fraud and should not use dollars

end result no one should own dollar and when people ask 'i want dollar' the reply is 'U.S, canadian or australian?'
end result no one should own bitcoin and when people ask 'i want bitcoin' the reply is 'core or cash?'

remember its not
_______________________ 1
                   \__________  2
(2 unilaterally created an alt)

its not a
                    __________ 1
____________/__________ 2
(1 unilaterally created an alt)

its a
                   ___________ 1
___________/
                  \___________ 2
(1&2 bilateral split)..

yes a fork. both side go in separat directions. where no side is a straight blade
no side is the same rules as 2009-2014
no side is "the bitcoin"

also
by core PRETENDING its not centralised by saying they are a team of individuals rather than one person. is like saying apple cant be a centralised because its more than just steve jobs.. EG apple isnt centralised because of wozniac's involvement and sometimes disagreement with jobs in the past aswell as there being hundred other people work who on apple products.. also yes apple had unpaid interns(volunteers) too

its still a team that has a roadmap(5 year business plan)

go on deny samson mow wasnt employed by blockstream
deny segwit wouldnt have activated unless the opposition were not moved away via ABC
deny ABC wasnt created by the guys that got paid by the same investors that pay gmax, wuiile and samson(blockstream).

the BScartel (Barry Silbert/BlockStream) really does have the puppet master strings.

(i expect may harsh cries of defense about this post from the core lovers.. but while they spend most of their time defending core.. they are not
concentrating on th big picture of dcentralised "bitcoin")
(i expect those core defenders who dont care about decentralsiation to tr pigeon holing me into another team. purely because they lack the understanding of the concept of true decentralisation and independance.. to them its only core loyalist or opposition)

[franky1]

LN is not a direct pay the recipiant(P2P).

It certainly is. You can open a payment channel directly with your recipient and avoid having to route your payment through anyone else.

its in laymans terms opening a channel(account) with a well routed(bank branch) hub. depositing funds into a channel(account) that requires co-signing(bank authorisation)

It can be used in that way, but that is not the only way to use LN.

LN is not the BANK..
LN is the swift network banks use.
hubs. one entity with authorisation power over multiple contracts of value store are the banks

yes you can set up a joint bank account with your wife(p2p) and never need to use the swift(LN routing).. but if you are just going to set up a payment system with someone you trust enough to be a counter-signer. you can just do that as an ONCHAIN multisig  and just use a piece of paper to tally who owes who what and just sign a final cheque/tx to finally settle up.

the purpose of LN is about routing/hubs and hops not about p2p

[franky1]

all with the very same banking features of 3-5 business day settlemnt (CLTV

That is completely false. There is no 3-5 business day settlement. That is not what CLTV is used for. It is a timeout for if a transaction fails. Not a "transaction goes through after X time".

its a timeout to not allow a CLTV output to be spent until X time (usually days to allow time for someone to realise a tx got confirmed but have time to still send a new tx with a revoke code to chargeback
hense CLTV concentrates on the 3-5 day before truly settled

github: CLTV " until that block height or block time has been reached the transaction output remains unspendable. "

) and wire transfer chargebacks(CSV)

That is completely false. That is not what CSV is used for. It is used as a timeout during which time a punishment can occur which is not the same as a chargeback. Once an HTLC is resolved or once new commitment transactions are put in place, the payment is final. There is not charging back. Trying to perform a chargeback would mean that you are broadcasting an old commitment transaction which means that you LOSE ALL of your money, not you gaining back whatever money that you sent.
[/quote]

CSV  REVOKES aka CHARGEBACK = punishment
you can call it a "punishment" all you like but all your screaming is tom8o.. the reality of average joe is actualy tom@o(chargeback)
if your waiting to receive funds and it shows its destined to be yours in a couple days.. but then someone else takes it from you. its a chargeback

EG
today 27tth April    multisig btween franky1 and paypal2.0 bank
tx1 CSV allow paypal to spend else 5 days pay franky 1btc..  signed franky1 signed paypal2.0
tx2 CSV allow paypal to spend else 5 days pay franky 1.2btc..  paypal gets tx1 Revoke .. signed franky1 paypal2.0 refuses to sign tx2
because paypal is not being co-operative franky1 sends tx1 as its the only tx franky has fully signed.
its broadcast, gets accepted into block. but franky1 cannot SPEND tx1 until 2nd may as he doesnt have tx1 revoke key

paypal does. so before 2nd of May paypal spends frankys tx1 funds back to paypals pot of funds. franky1 lost out

yep paypal done a chargeback

because its open source and all handshaking is done in private without the community decentralised auditing...
paypal node can tinker with its sourcecode to be the one that requests franky1 is the first to give up the old key first. thus franky1 loses out

if LN was so flawless and trustless. the blockchain(2009 invention) would not ever needed to be invented. because people could just smart contract
in private. and the whole many decades of trying to find a solution to decentralsied money would not be blockchain but LN

come on achow.. are you really saying that decentralised blockchain technology is a dead technology and not required. because private smart contracts are flawless. come on. atleast admit LN has limitations and flaws

as said
its not about YOU making a payment and YOU getting a refund.. its about YOU wanting to withdraw funds from paypal2.0 and paypal taking the money. leaving YOU out of pocket

[franky1]

There are only ~1500 LN nodes.

It's still in beta, what do you expect?

but already there are hubs being the middlemen for ~10% of the network

You seem to be ignoring the criss-cross of gray lines for all of the other channels.

The "hubs" currently on mainnet are due to people opening direct channels with merchants (i.e. performing direct P2P payments).

yes the merchant becomes the banker2.0
put all your funds into a richard branson channel so you can buy train, plane and space travel tickets. by the way you need richard bransons signature every time you want to buy something.. oh look richard branson is now a bank

as for other posts
some banks dont use customers funds for internal investing
some banks dont do mortgages
some banks dont do credit cards or debt

these hubs are not best friends you conect to p2p, they are centralised services(HUBS) that the LN dns seeds can send a certain preferential list of hubs to users.
yes you can manually request to join a friend.
but that does not negate the point that hubs do exist and are centralised authorisers of hundreds of peoples funds


a bank is a storage vessel of value that needs authorisation. a central bank is where a central authoriser has powers to authorise or reject payment of other peoples funds. if you wish to object. then i guess another of satoshi's ethos of bitcoin has died "be your own bank"

LN is not a "be your own" concept because it requires secondary authorisation but it is a "bank" because it stores value and requires authorisation

you can play symantics all you like
"dont call it a chargback, use 'punishment' buzzword instead to hide average joes understanding"

but to generalise it
LN is not fit for purpose as a scaling solution everyone should use.
LN is not a 'unlimited use' concept that never needs to close channels.
LN is not a service where it is guaranteed to work if you only connect to one friend(you cant pay anyone by only connecting to a friend(DDos/out of funds issues))
LN is not a service where you can pay anyone in the (imaginable future of 7billion LN users) for just 1-3 hop fee's and only needing 1-3 channels to do that.

HUBS is the concept of getting millions of users 'well connected'

100 nodes with 100 connections and EVERY connection had 100 connections..
then you can be lucky to pay anyone in 3.5 hops(lets call it 4 just for sanity sake)
...screw it 95 nodes 95 connections and every connection had 95 connections = more realistic 4
or
300 nodes with 300 connections and EVERY connection had 300 connections..
then you can be lucky to pay anyone in 3 hops

but not all nodes will be 95-300 channels to pay within 3-4 hops, meaning much more than 4 hops will be required(fee costs increase)
or required to connect to hub(keeps costs down but then use centralised authoriser)

there will be some users with 3 connections (accounts with 3 banks)
and there will be hubs(banks) with 300(or more) customers that the hub(bank) will need to sign for

now imagine each node/hub equally agreed to charge the minimal 1millisat(1MS)
would you prefer a DECENTRALISED network of users with only 3 connections each which would cost upto 22MS but no guarantee of routing
where sometimes route fail due to DDos and funds unavailable in the route of channels

or would you end up using a CENTRALISED system of connecting to 3 hubs(banks) where it could cost you upto 4MS

because some how achow101 has this DIS-belief that LN is a utopian dream of perfect limitless connectiveness of only needing to connect to a couple friends and having guaranteed payment system that costs literally nothing to pay anyone in the world.

before you answer. take the mindset of average joe that can only predict their spending habits for a couple weeks ahead so will only want to risk $60~
would they want to split their funds into 300 channels of $0.20 or have maby 1 or to channels with hubs of $30~ each

[Kakmakr]

I think where these people gets bogged down in these arguments, they miss one of the most crucial aspects of the technology. Nobody is forcing you to use a specific hub, you can still open your own payment channels. This is not something that you can do with Banks. The Banks will not allow you to open your own direct payment channel to someone in another country.

Second layer technologies like this will only be centralized, if it was controlled by centralized entities, but the LN is open to everyone. Big difference, right. 

[Carlton Banks]

Second layer technologies like this will only be centralized, if it was controlled by centralized entities, but the LN is open to everyone. Big difference, right.  

Lightning devs are also working on a routing algorithm that causes the network graph to be arranged like a grid. It's gonna be pretty difficult for hub-like nodes to monopolise routing when that techniques makes it into the protocol, trying to predict which nodes to open channels with could become frustrating experience if the hub operator was trying to make money. It'll make more sense to just open up lots of independent grid nodes, in the hope that you can somehow predict where money will flow on the network, and put nodes in strategic parts of the overall network graph. And the network graph is going to be changing shape (and it's payment flow dynamics) constantly, so the algorithm that places your (supposedly profit making) nodes needs to be able to predict the randomness of payment demands, in essence. There cannot be 100% success rate, let's just put it that way, and it wouldn't surprise me if profit making Lighning nodes is a non-viable business in that scenario.

[DooMAD]

Probably the most relevant argument one can make at this point is that it doesn't matter in the slightest if some people think it's going to be centralised or it won't scale, because it's still happening anyway.  The people who believe it's a good idea aren't going to stop working on it just because some other people don't think it's a good idea.  So everyone might as well just wait and see. 

If it doesn't meet your exacting standards, feel free not to use it.  If you think it's so bad that you can do better, go right ahead and make something better.  No one's stopping you.  The various Lightning dev teams don't need permission from anyone to build their projects, neither do you.  Put your plans into action if you think LN is such a terrible concept.  See who takes an interest.

[mu_enrico]

Oh, perfect timing on this discussion. I'm currently researching about this topic.
From what i read, the major flaw about LN is the route will break if one node doesn't have enough BTC amount as seen in this video
https://youtu.be/UYHFrf5ci_g

That said if you have great amount of BTC, most likely many transaction will be routed to you (aka become Hub -> centralization)

However Antonopulos argue that the higher amount BTC in that "hub" the higher risk of attack. And he also mention about re-balancing channel which I have no idea about how that works.
https://youtu.be/D-nKuInDq6g?t=6m40s

What do you think?

[Theb]

If the Bitcoin Core Team is acting as a centralized authority of Bitcoin then BTC is not counted as a decentralized currency in the first place. I think the writer thought that LN is the way of the Core on having some kind of banking authority with regards to transaction, which he is wrong about. The concept of having a middleman in LN transactions are totally avoidable, the short solution is opening up your own channel and pay directly to who you are sending your money with. So technically you are still in total control of your money. Also I guess that some hubs involving a "middleman" is good as you know some transactions here in the forum uses Escrows, and by having middlemen involve it would surely eliminate the use of these escrows.

[TheQuin]

That said if you have great amount of BTC, most likely many transaction will be routed to you (aka become Hub -> centralization)

You're only going to become a hub if people choose to open channels with you. They are unlikely to know or care how much BTC you have. Merchants, exchanges and other large sites will most likely become hubs simply because they are the businesses people want to transact with.

Centralisation means under control of one body which the LN can never be. I find that most people that accuse LN of being centralised have no idea what a decentralised network is.

https://i.ytimg.com/vi/Dettsr71gKg/maxresdefault.jpg

[franky1]

the lightning network DNS will have a hotlist of prefered well conncted nodes to hand out to new users.

now who do you think these DNS managers are going to decide are top of the list... themselves


hint blue nodes are all one person/entity

both my edited image and the one provided by quin do not prove decentralisation, but prove distribution.. there is a difference.
think of the blue dots as a bank branch of a bank corporation. each blue node has a separate bank branch manager that
authorises payments of its customers of that bank branch. but its all part of a bigger corporation... oops i mean "factory" cough cough

also carlton banks hints at the "grid" concept.. which is done via the DNS list choosing best nodes(to be the corners) to promote to fresh users.
put short.. you cant have a grid unless your controlling who connects to what.
in a decentralised network there would be no control of who connects to what. thus impossible to create a grid in a true open decentralised randomly selected network

yet LN has a autopilot feature, a DNS seed,
and remember if you imagine a grid of squares.. who do you think will be the (hubs) well conected corners


P.S just wait until the explanation of 'channel factories' gets the ELI-5 treatment. then it all becomes clear to average joe

[Wind_FURY]

P.S just wait until the explanation of 'channel factories' gets the ELI-5 treatment. then it all becomes clear to average joe

I cannot wait for that. I want to see your point of view. Thanks.

But depsite your argument, I believe the Lightning Network will not be used for the sake of "decentralized offchain transactions". People will use it because transactions confirm instantly and because it will be more private than onchain transactions.

Plus the fees will be cheaper.

[cellard]

The major argument against the Lightning Network in my opinion is still the fact that
nobody really needs a Lightning Network when Bitcoin´s main use case is being a
store of value. If you want to preserve your wealth from inflation, confiscation and other
comparable threats you can buy Bitcoin. However, you probably won´t touch your Bitcoins
for years if that is your reason for buying Bitcoin and therefore you are not the target group for the
LN.  

Besides, nearly everyone, who bought something using his Bitcoins over the years has regretted
that after the fact, because BTC only goes up in the long-term. If you believe this will be the case
going forward it is pretty stupid to buy something using a Lightning Network transaction when
the opportunity costs of buying anything with your precious Bitcoins are so tremendous.

I´m a strong believer that Bitcoin´s biggest advantages are rather in its unique characteristics
that make it a form of "sound money" than in its use as a payment system for small value
purchases or transactions (which is the major use case of the Lightning Network).

The great thing about the Lightning Network is that it is non-mandatory and anyone can still
decide to use the main layer for his transactions even if the LN becomes a giant success (which
it will not become in my opinion). This is the beauty of 2nd layer solutions in general,
they can offer new features while maintaining the trust and integrity of the main layer.

Agree. People stress about scaling Bitcoin, when it's just a secondary thing. Why would I do groceries with Bitcoin when I have fiat? Why would I buy online with Bitcoin when I have fiat? etc etc. As long as Bitcoin keeps going up, there's no point in spending it, unless completely needed.

But earth is big and big chunks of population don't have fiat, or a functional one at least. In this case, they could go directly to a Bitcoin based economy in which they bypass the traditional fiat model and just use LN for payments. It all depends on the context. What's clear is: Fucking up the blockchain so these poor people can transact on-chain is insane, and that is why BCash and all these other forks lead to eventually.