Pages:
Author

Topic: Is the Lightning Network centralized? - page 8. (Read 1981 times)

jr. member
Activity: 154
Merit: 3
July 10, 2018, 01:57:09 AM
Log into the Lightning network through the opening of the payment channel. This requires a transaction in the blockroom (the one in which transactions are always written). The payment channel allows you to send bitcoins between two parties (this is not a new development). To put it even easier, imagine that you use Yandex.Money or Paypal. You can send money to another account within the system quickly and without commissions. The limit is only how much money you have in your account. To enter or enter the payment service, you also need to make a banking transaction (from your card or ATM enter money to your account). Similarly, for Lightning - to open or close the payment channel, you need to perform a bitcoin transaction.
legendary
Activity: 2898
Merit: 1823
July 10, 2018, 01:41:02 AM
Yes, as long as the block size is "regulated" to the most necessary size, the network will scale out. The Bitcoin Cash supporters say it does not matter because "Moore's Law". But what they did not consider is not all node operators are willing to do the hardware upgrades.

i actually bothered to read the bitcoin cash codebase.. and actually. if you read the code. the 32mb is known as a consensus limit.. but they also have a "regulated" limit of 2mb called the policy limit
https://github.com/Bitcoin-ABC/bitcoin-abc/blob/master/src/policy/policy.h#L19

this is the same as what bitcoin core had  before 2013.. 1mb consensus limit but 0.5mb "regulated limit" (in policy.h)which then went up to 0.75mb
and then in around 2015 went up to 1mb.. all without needing to fork or have big discussions because the policy was a flexcap limit

check it out consensus.h vs policy..h

so if you think that spammers can just spam their way all the way to 32mb you will be wrong. just like spammers couldnt spam thier way to 1mb in 2013 on bitcoin core..
try reading code and not reddit propaganda next time

That is not the argument at all. The argument is if Bitcoin Cash does scale to billions of users and that it was necessary to use the full 32mb everyday then it would centralize the network with users losing the ability to run their own full nodes. Or is becoming a widely used global network for decentralized payments not in Bitcoin Cash's long term roadmap?

Quote
p.S im not an advocate for bch i just like to wash away the bull crap that core and utopian propaganda sheep spout out..
funy point is core devs themselves have admitted core have issues and core is restricted on purpose to incentivise use off offchain services... but ass kissers love to brush that under the carpet

What are you? An advocate of Roger Ver? You have the same FUD and misinformation styles. Haha.

Quote
the idea of policy.h is that a limit exists, but can be moved without 2 years of debate and without needing forks.
gotta love it how even 2 years on the same scripts of 32mb by midnight, 1 gb by midnight screams are still shouted . but the real code shows something different.

it was requested by the community at satoshi roundtable 2013 that bitcoin cores consensus.h go to 8mb. and then allow policy to increment up at 0.25mb amounts as and when needed.
luke JR said 8mb is bad.. even with the enphasis of using policy.h to flexcap control the growth without neding constant debate and fork warnings
then at a later conference, called consensus december 2015 another compromise, that bitcoin cores consensus.h go to 2mb. and then allow policy to increment up at 0.25mb amounts as and when needed.
luke JR said ok....   but then a few months later.. 2mb is bad

now luke JR says 4mb weight is good but lets bring policy.h down to 0.5mb (facepalm)

I believe what was being avoided was the risks of a hard fork? There are people out there who want to kick out the Core developers. The "2x" hard fork was one supposed to be one of their traps. But post quotes and links of the facts. I would like to read them.
sr. member
Activity: 635
Merit: 251
July 09, 2018, 09:28:16 PM
this network of lightning is very good for things that are important and only done by professional people and can make us better understand about it.

Yeah..right...
I like on-chain transactions a lot better though.
full member
Activity: 476
Merit: 100
July 09, 2018, 09:20:34 PM
this network of lightning is very good for things that are important and only done by professional people and can make us better understand about it.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 09:11:29 PM
frankly I do not see what's wrong with the swap atoms used to move the funds to another blockchain if the transaction is for a smaller amount and a lower fess like that. In many ways it is better than not chain. I think that scaling can take that form if it is the way people are free to choose to use LN.

imagine you made a highway. you made it free from expensive toll roads. you made it maintainable by the community rather than government..
then a few years later, some corporation decided to remove the fee controls and suddenly costs to use the highway become more expensive. the community requested extra lanes be added. to allow more traffic and also alleviate the risk of high fee's.. but this new corporation decided not to allow it. instead they made a bus lane(segwit). that is only usable for buses where parents could sit, while the children had to use the limited highway lane (parent=witness/ proof(signature) child is real).
this results in only a 2.1x~ extra capacity. even though the one lane highway is now a 1 lane highway with 3 extra buslanes for parents.

then they decided to re advertise that the highway was broken and said it wont scale.
yes the core devs themselves called satoshis invention broken and obsolete...(facepalm)

(much like kodak said you cant store images on removable digital media in the 90's because it was only 1.44mb of space on rmovable media back then(floppy disks)... look what happened to kodak

so although everyone, including devs knew that it could scale(yet they were screaming it cant scale) and that they knew 2-4mb was not such a big ask and was even safe even as far back as 2009.. the devs kept the highway at a legacy limit and started developing a new road network that was not even a strong immutable network that the whole blockchain revolution/invention was based on.

basically taking away all the features and benefits of blockchain. and reverting back to how banks work where assets need to be locked up and where a user no longer has 100% control to randomly give assets to anyone.
they now need not only the signature of somone else and play around with unconfirmed paper receipts..which are co-managed in a channel(bank account) but also now the funds have to be 'hopped' through multiple accounts where each account holder co-party need to agree. just to get to a destination.

then to add to that devs realised they cant trust users to press agree. so they made it automated. so now although its not a human agreement. an agreement is there.. but now a human cant disagree and stop their funds being used on behalf of others. (yes people have run scenarios and stolen funds by emptying peoples reserves via routing. not allowing users to use all their funds themselves)

then imagine devs envision the horror of people wanting to withdraw. much like a bank run. so they create factories(fort knox) where funds dont broadcast to the network.. but sit back in fort knox and then redistributed back into the unaudited unconfirmed network. thus never needing to be audited by the community(use the blockchain)

while still not increasing scaling bitcoins mainnet to have buffer capacity to cope with any bottlenecks if people really did want to get back on bitcoins mainnet.
thus strangling bitcoins utility in the hopse people hand over their bitcoin and grab an altcoin via offchain swaps becasue devs have made it.. yes they made it undesirable to want to return funds back to the mainnet. (imagine it like gold.. throw the gold into a bank, get handed a bank note signed by th bank.. then get told by the bank. 'nah you dont want to lug around the real gold, its heavy and the fee's to withdraw your gold is too high... take back this bank note and put it into anothr account and if you really want to. go find someone to swap it for silver(litecoin)"

anyway. LN is not a sole solution for bitcoin. its a roadmap to move people away from bitcoin.
and thee whole segwit stuff was not to add transaction capacity. but to allow all them multisigs which would have bloated the legacy limit. to sit outside the lgacy limit thus in the end not hlp capacity growth. but just delay capacity reduction

if people actually cared about bitcoins ethos, utility and function.. and not just see it as a system to run back to fiat ASAP richer. you would see th issues. but too many people are blind and only wantt o se extra fiat in thier pockets. thus missing the whole purpose of bitcoins invention/revolution in the first place

but hey.. let me guess all ill get is insults and get told to screw off to an altcoin... (facepalm) simply because bitcoin is no longer a revolution but a fiat get rich quick scheme.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 08:11:32 PM
and if you think devs should get to do whatever they want and not listen/answer to the community... then that is totalitarian because you think no one should tell them what they should be doing.. you sir have just handed control to the devs...(facepalm)

If Bitcoin were closed source and we had no other choice but to run their code, you might possibly have a point.  But since that isn't even close to how Bitcoin works, you just sound like a raving lunatic who has no idea what freedom means.

freedom
.. augusts MANDATORY bilateral split.. mandatory does not sound like a freedom to me.


open source
yea you can read the code.
i have a open window... doesnt mean that it means anyone is allowed to come in without asking.

but have you seen the process and moderation of getting code added. did you forget how all the proposals that were NOT core roadmap related got treated as an attack rather than a difference of opinion.. or rathr than a different option.. do you even know who the gate keepers are that you have to go through just to get a bip added. or even get code acknowledged.

firstly you discuss it on this forum or IRC. and look gmax, achowe moderated. the IRC again moderated. then you have to go to the mailing list.. moderated by rusty russell.. then you have to make it a bip, again moderated by lukeJr.. and then you need it 'ackd'  by certain people..

many have tried to do their own implementations because the core roadmap is a one way street that has tunnel vision. so because it had been made unsuccessful to divert from the core roadmap, devs have had to make their own implementations..
then those implementations got treated like altcoins and REKT. and told to get off the network if they want changes..

even you yourself have said if you dont like it make your own altcoin..


do you seriously not see that segwit. only had 35% vote for 6 months. thus they knew they didnt have community support for consensus. so instead of them  giving up or changing the code or compromising to something the community would want. they went to the mandatory activation date of august first. and threatened the community and the pools to accept it or get their blocks/nodes rejected/banned from the network.

go check it out. its in the history. its even wrote in the blockchain that it all changed on august first. so you cannot pretend that it did not happen
block 478559

im guessing while reading the previous sentance you now racing to reddit to find some propaganda script that it was cash that instigated it.. so let me pre-empt your attempt and correct you.
cash didnt make a new block until hours AFTER the core network started banning nodes and rjecting blocks that didnt have a certain version number.
core = UTC 13:23
cash = UTC 18:12
 so it was cores network that acted first. oh and there was a special hat people had to wear to show they loved the mandatory split idea.

oh and before you try to say core devs were not involved in it. guess what LukeJr was part of.. Cheesy and a colleague of his (paid by the same boss) was involved too..

you can call me any name you like. but the blockchain data never lies. #478559

member
Activity: 294
Merit: 10
W12 – Blockchain protocol
July 09, 2018, 07:38:35 PM
Lightning network is a hardfork of bitcoin that enables lightning speed transactions between bitcoin miners who participated in the blockchain and it is considered to be the solution to bitcoin's scalability problems. For further readings, you can check it for yourself from this topic here.
newbie
Activity: 210
Merit: 0
July 09, 2018, 07:31:56 PM
frankly I do not see what's wrong with the swap atoms used to move the funds to another blockchain if the transaction is for a smaller amount and a lower fess like that. In many ways it is better than not chain. I think that scaling can take that form if it is the way people are free to choose to use LN.
legendary
Activity: 3948
Merit: 3191
Leave no FUD unchallenged
July 09, 2018, 07:27:17 PM
and if you think devs should get to do whatever they want and not listen/answer to the community... then that is totalitarian because you think no one should tell them what they should be doing.. you sir have just handed control to the devs...(facepalm)

If Bitcoin were closed source and we had no other choice but to run their code, you might possibly have a point.  But since that isn't even close to how Bitcoin works, you just sound like a raving lunatic who has no idea what freedom means.  Their code means nothing unless people choose to run it.  That's why it's fine for them to code what they want.  If the users don't like what they code, no one will run it.  Is there any way that will ever sink in to your impenetrable skull?  It's the people who run the code who make the decisions, not the people who made the code.  The devs are listening to the community and what they're hearing is the community's support for their ideas.

I can quite happily say that all the BCH dev teams are free to code and create whatever they want, as I'm sure you are.  So why are you totally incapable of extending that same common courtesy to developers on the other side of the fork?  Is it just the ideological differences?  Or is it possibly because you're so broken in the head that you can't even see the double-standard anymore?  


heres a few fixes

(...)

2. incorporate a new fee priority mechanism that penalises people that rebroadcast low confirm count tx's (average users dont spend 144 times a day so make it expensive to resend every blocks(spam))
  EG 144sat per byte / confirms) so 1confirm =144spb     144 confirms=1spb
(i very simplyified it but i remember doing code for it and also pseudocode documentation for a fee priorty years ago.)

In fairness, I do recall that one.  We discussed it when I was promoting a proposal for dynamic blockweight adjustments and I actually thought your fee priority mechanism was a decent idea.  However, I also recall that, even then, you were so belligerent that you couldn't accept the part of the dynamic adjustments that involved the maximum blockweight potentially decreasing if there wasn't sufficient demand, even if that meant alienating people who could potentially support the proposal.  You just wanted perpetual increases and weren't remotely open to compromise or reason (like a totalitarian, one might say).  I think that's yet another one of the occasions where I lost some respect for you.  And ever since, our encounters have only lead to more of that respect being eroded away because you became more and more of an extremist.  And now that I think you've attained BoJo-grade levels of shitweaselry, that respect is now long gone.

Remember that I campaigned for larger blocksizes before your so-called "bilateral split" too.  The difference between us is that I'm prepared to back down and accept the community's decision on this.  Consensus is more important than any one person's ideals or ideas.  My proposal clearly lacked support.  Hardly anyone cared.  I have no shame in conceding that fact.  I don't have a problem with that, because we now have things being developed that will scale Bitcoin far more than my proposal ever could.  It was the right call.  And the simple fact is, that however good your fee priority mechanism, or your other ideas, might be, it doesn't even touch the sides compared to Lightning.  None of your minor tweaks and adjustments would create the potential to move the coins within any given channel an unlimited number of times with just two entries in the blockchain.  None of your ideas come close to achieving scaling in the sheer magnitude that Lightning potentially can.  Of course we don't know with absolute certainty how it's going to turn out and there are clearly some things to solve along the way, but everything I can see tells me we're on the right track.  

If you don't think we're on the right course, no one's forcing you to stick around.  That's not me being a totalitarian, it's just you pissing into the wind yet again.  Consensus made this decision, so either follow or don't.  Those are the options now, because that's just how Bitcoin works.  Again:

This.  Is.  How.  It.  Is.  

If that causes you frustration, feel free to take it out on me if you like, but don't blame the devs for merely giving people the choice.  And stop being so manipulative and tory-like with your rhetoric.  
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 03:06:08 PM
And LN can cope with 32,000,000x or probably even more capacity without sudden fear of bottleneck.

32,000,000x..

LOL
im laughing..
oh.. and stats of btc in 2009 had the potential of 7tx/s
even upto 2016 btc never got to 7tx/s... in 2017-2018.. bitcoin still hasnt got 7tx/s even though bitcoin devs have meant to be innovating the main net for upto 9 years.

all i hear the core devs says is that bitcoin cant scale. bitcoin cant do this and that..bitcoin fee's cant be controlled..
why
because they stifled onchain growth. they removed fee controls. they backtracked consensus agreemnts and with one hand pretended they were not core devs to get out of the community agrement. but then went full retard going in a diffrent direction of mandatory bilateral split to get what THEY wanted.
the 4mb weight is not a scaling compromise to shut up the community. its still a 1mb limit for legacy and then extra side space for waste data they can prune off or keep because of heavy multisig and many signature data and even confidential payment waste data to sit in while not really giving the room for more transactions onchain(estimates of 2.1x~ at best if EVERYONE used segwit tx's).

..
and if you think devs should get to do whatever they want and not listen/answer to the community... then that is totalitarian because you think no one should tell them what they should be doing.. you sir have just handed control to the devs...(facepalm)
...
heres a few fixes
1. change things to no longer allow someone to make just 5tx's to fill the limits.. yep core changed a few things to make it so someone can create a just 5 tx's that can fill a block. but done nothing to mitigate it (research signops)
  EG pre segwit argument "it cant scale because linear sigops take time".. solution.. dont allow Xk sigops/tx instead limit a tx to be just 500sigops so it then means more tx's can go through without having clumpy tx's with big sigops to fill blocks and slow verification using just 5 tx's

2. incorporate a new fee priority mechanism that penalises people that rebroadcast low confirm count tx's (average users dont spend 144 times a day so make it expensive to resend every blocks(spam))
  EG 144sat per byte / confirms) so 1confirm =144spb     144 confirms=1spb
(i very simplyified it but i remember doing code for it and also pseudocode documentation for a fee priorty years ago.)
  

3. increase the base block limit(consensus) and then use policy to "regulate" the increments that way it does not require constant fork debates to control the block size but can be done more easily when actually needed without years of debate.
 
4. fix the issues of segwit.. yep some devs and even pools that advocated their love for segwit in 2016.. are still too afraid to use segwit bc1q addresses even now due to some issues.

5. if you want to make bitcoin great again. make LN be a sole feature of bitcoin.. that way it doesnt sway people to use other coins as their exit and it keeps bitcoin ontop as they get the cheap speedy tx if they do want to do 144 tx+ a day.. (once LN solves its flaws ofcourse)

but while theres no fee control onchain. people can spam 144 tx's a day onchain without it costing them extra for the privelidge.
and they can fill a block using just 5 tx's. all without having to pay a huge fee
EG imagine a plane with 2000 seats first of all would you want to hold up all the passengers at the gate because someone bought just 5 tickets but had 1995 carry on luggages, but didnt need to pay for the carry on luggage, but that carry on luggage used up all the seats.
wouldnt you as airport control mwant to implement something that if someone kpt doing that, they will need to start paying more for the privelidge

then... guess what. then you will find your niche market of people who would (if genuine business needs) wanted frequent travel and bottlenecking the gate decide its better for them to want to use LN..(when/if LN is fixed)
but no, the idea devs have is bottleneck the gate and then tell the infrequent passengers (emphasis: that dont frequently travel) that their family(funds) have to split up and go to LN airport where husband has to go to gate 7(one channel) wife needs to go to gate10(another channel) and the kids have to go to different gates and hopefully everyone will arrive at the destination if all the alternative route work in a timely manner.
...
do you know that visa/master card users only do ~40tx a month. and bitcoin users only do 5tx a month..
LN is not in practice economical for people to preplan a month-6 month ahead their spending habits to know how much to fund into LN then have funds locked to save themselves a few sats.

LN's niche would be for the ones that want to move funds multiple times a day to save the sats where they can plan for a few days ahead and make enough savings offchain for it to be viable to lock funds off
try running real scenarios. without reading a promotional reddit post / watching video/cartoon that suggests LN is going to be a solution for everyone
hv_
legendary
Activity: 2534
Merit: 1055
Clean Code and Scale
July 09, 2018, 03:01:13 PM
There is only one proper proposal for a working Bitcoin scaling concept, that also keeps the SEC out of it.

It is real industrial style on-chain scaling. Sorry to say that, but I fear this guy here, Joannes Vermorel is absolutely right.

There is no time any more left for romantic experimentation with Bitcoin and some fancy unsolved SEC layer stuff around.

Listen around 1:25

https://memo.cash/post/f3cba7ea021820b07435eba96c209609b643ca81cf4f9db60b87546f52a9ea19

Repeat if you feel not understanding that.

Feel free to not understand it as well.
legendary
Activity: 3948
Merit: 3191
Leave no FUD unchallenged
July 09, 2018, 01:47:06 PM
I guess you must have missed it, so where's the reply to this part franky1?  
once you admit that developers have stagnated innovation onchain to push these commercial networks. then you can kick the devs asses to stop concentrating on commercial sid services and get back to innovating and scaling the mainnet.

And what "innovation" do you think will solve scalability, exactly?  I hear all the critiques, but absolutely bugger all that sounds like a solution from you.  Come on, let's hear it then...  

Hint:  Blocksize increases are not "innovative".  Look up the definition of the word.

Developers are also working on Schnorr Sigs and MAST, which (like SegWit) also contribute towards on-chain scaling in an innovative way.  But you're too busy bawling like a spoiled brat to acknowledge that either.  Again, grow up.

What's your proposal to fix everything?  

You're clearly all-knowing, so a reply from you must be worth the anticipation we're all feeling in waiting for a response on this question.  We eagerly await your wisdom and enlightenment.    Tongue


If you don't like what the developers are doing, either make something better yourself or use something else.  Developers don't owe you anything.

gotta love and laugh at your mindset, basically our saying
'if you dont like the fact that devs have stiffled bitcoin. make an altcoin or go use another altcoin.'

now that is the most monarchistic totalitarian comment you coulld ever say.

how about stop kissing devs asses and letting them get away with stiffling bitcoin and instead kick thier ass to make it better for everyone.
devs owe the community to be responsible.. and its a damn shame there are so many sheep that just have a mindset of, leave the devs alone, let them do as they please

The totalitarian is the person insisting that developers shouldn't be coding what they want to code.  The totalitarian is the person telling people they shouldn't have a choice to use off-chain if they desire.  The totalitarian is the person who would demand that everyone else be willing to sacrifice the qualities they appreciate in Bitcoin for the sake of scaling.  The totalitarian is the one saying the market was wrong to choose this route and shouldn't have been permitted to do so.

I advocate freedom.  You evidently don't.  

Also, the developers haven't "stifled" Bitcoin.  You're simply too dense to grasp the wisdom of what users and miners have chosen.


in short. 32mb blocks do not mean 32mb of data need to be filled/used befor something occurs. it means there is buffer, space, potential, room to grow,
even if blocks only utilise 1mb now. guess what. if it gets popular/busy tomorrow, it can cope with 32x more capacity without sudden fear of bottleneck.

And LN can cope with 32,000,000x or probably even more capacity without sudden fear of bottleneck.  AND without fear of adding all that extra data to the blockchain, meaning we can still have a decentralised layer 0.


sorry dude but it seems its you that loves alternative payment systems that are not done via bitcoins mainnet..

Good luck trying to make a Lightning payment without making a mainnet transaction first.  Reassure me one more time you actually understand how Lightning works, perhaps?  'Cause I'm pretty sure you don't with the nonsense you keep spouting.  

10 years time
DooMAD's child "my dad told me once that he used bitcoins mainnet, that was a laugh, these days i use unaudited LN transactions that just circle settlement payments through LN factories.. bitcoins blockchain is so old school no one uses that no more.. if i ever want to get out of LN id never settle to bitcoins mainnet. its too slow, cant scale and too expense, i attomically swap it to a cheap altcoin instead. i literally laugh at the fact that my dad use to use bitcoins mainnet.. infact my LN unaudited payments are measured in hyperledger tokeks with the USD1q prfix. i dont know anyone at my school who even bothrs with bc1q prefixed payments anymore"

While it's possible you might be exaggerating just a tiny bit, I honestly don't see what's wrong with atomic swaps being used to move funds to other blockchains if the transaction is for a smaller sum and the fess are lower that way.  In many ways that's preferable to staying off-chain.  I'm of the view that scaling could well take that form if that's how people freely choose to use LN.  It's just one of the many possibilities we can now explore.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 01:22:03 PM
Yes, as long as the block size is "regulated" to the most necessary size, the network will scale out. The Bitcoin Cash supporters say it does not matter because "Moore's Law". But what they did not consider is not all node operators are willing to do the hardware upgrades.

i actually bothered to read the bitcoin cash codebase.. and actually. if you read the code. the 32mb is known as a consensus limit.. but they also have a "regulated" limit of 2mb called the policy limit
https://github.com/Bitcoin-ABC/bitcoin-abc/blob/master/src/policy/policy.h#L19

this is the same as what bitcoin core had  before 2013.. 1mb consensus limit but 0.5mb "regulated limit" (in policy.h)which then went up to 0.75mb
and then in around 2015 went up to 1mb.. all without needing to fork or have big discussions because the policy was a flexcap limit

check it out consensus.h vs policy..h

so if you think that spammers can just spam their way all the way to 32mb you will be wrong. just like spammers couldnt spam thier way to 1mb in 2013 on bitcoin core..
try reading code and not reddit propaganda next time

p.S im not an advocate for bch i just like to wash away the bull crap that core and utopian propaganda sheep spout out..
funy point is core devs themselves have admitted core have issues and core is restricted on purpose to incentivise use off offchain services... but ass kissers love to brush that under the carpet

the idea of policy.h is that a limit exists, but can be moved without 2 years of debate and without needing forks.
gotta love it how even 2 years on the same scripts of 32mb by midnight, 1 gb by midnight screams are still shouted . but the real code shows something different.

it was requested by the community at satoshi roundtable 2013 that bitcoin cores consensus.h go to 8mb. and then allow policy to increment up at 0.25mb amounts as and when needed.
luke JR said 8mb is bad.. even with the enphasis of using policy.h to flexcap control the growth without neding constant debate and fork warnings
then at a later conference, called consensus december 2015 another compromise, that bitcoin cores consensus.h go to 2mb. and then allow policy to increment up at 0.25mb amounts as and when needed.
luke JR said ok....   but then a few months later.. 2mb is bad

now luke JR says 4mb weight is good but lets bring policy.h down to 0.5mb (facepalm)
legendary
Activity: 3906
Merit: 6249
Decentralization Maximalist
July 09, 2018, 01:06:38 PM
@franky1: I get you point. But if you're a mini-hub only because you use LN "as a consumer" and have no interest to exploit your LN node commercially, then you will be able to charge zero fees, as long as your node does not become part of a "unbalanced" route (with funds moving almost only from A to B and not back from B to A). Channel factories should mitigate that problem if it appears.

Just for understanding: You really lose the fees you paid if the route you chose does not "reach out" to the recipient of your transaction? I meant to have read that everything is rolled back in this case.

I also think there will be lazy users that will use mega-hubs (e.g. their exchange), but I think there will be enough "libertarians" trying to use LN the most decentralized way possible, so even "semi-commercial" mini hubs will have a niche.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 12:54:26 PM
Quote
Competing with a large hub , which is making profits means the large hub can continue improving while the small hub continues to do less, until no one uses the hobby hub.
I still don't get it - if there's a small hub that charges a very small, or even non-existent fee and is not profitable (this fee is more of a "tip"), why should you use a large hub instead if he needs to charge you a relatively high fee to be profitable?

And small, non-profitable "mini-hubs" (e.g. "your well-connected techie friend") are not money transmitters, as they are not businesses.

to answer the question for zing zang its probably easier you run some scenarios.

EG imagine you are a mini hub concted to only 14 peers.. the chances that you are a hub thats connected to the recipient zingzang wants is low. so its more chance that the person you are connected to. is then connected to someone else and then connected to someone else.. and so on.. to finally get to the destination.
do the maths. how many mini-hubs are needed to have connections of 1bill plus users
14
196
2,744
38,416
537,824
7,529,536
105,413,504
1,475,789,056

(14 to the power of Cool

so imagine these mini-hubs only charges 1 millisat.  you probably end up ACTUALLY paying upto 8 millisats once you include all the hops
and have the extra risks of 8 points of failure/offlines/unfunded routes.

now imagine a mega hub of 1000.
1,000
1,000,000
1,000,000,000
(1000 to the power of 3)

so imagine these mega-hubs only charges 2 millisat.  you probably end up ACTUALLY paying upto 6 millisats once you include all the hops
and only 3 points of extra risk

so although it appears cheaper to conect to mini hub of 1 millisat.
with the risks and add on costs of the multiple hops of a mini hub model. people end up moving to mega hubs where theres less risks. less end to end cost per payment.. and the mega hubs themselves with 2millisats have 1000 'customers' a day (=2000millisat if everyone does 1 payment through them a day) instead of a minihub whos only income is 14millisats if 1 payment per conectd user a day

thus the minihubs get less incentivisd to reserve thier own funds for use for other users because they dont earn much.. leading to them not bothering after a while

EG - bitcoin mining scenario
an S9i asic costs $950 and can break even in 9 months
as USB asic costs $5 and can break even in 9 months.. but no one bothers USB mining because $5 for 9 months is not worth it they just let the s9i take over control of mining.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 12:31:50 PM
sorry dude but it seems its you that loves alternative payment systems that are not done via bitcoins mainnet..

Good luck trying to make a Lightning payment without making a mainnet transaction first.  Reassure me one more time you actually understand how Lightning works, perhaps?  'Cause I'm pretty sure you don't with the nonsense you keep spouting.  

10 years time
DooMAD's child "my dad told me once that he used bitcoins mainnet, that was a laugh, these days i use unaudited LN transactions that just circle settlement payments through LN factories.. bitcoins blockchain is so old school no one uses that no more.. if i ever want to get out of LN id never settle to bitcoins mainnet. its too slow, cant scale and too expense, i attomically swap it to a cheap altcoin instead. i literally laugh at the fact that my dad use to use bitcoins mainnet.. infact my LN unaudited payments are measured in hyperledger tokeks with the USD1q prfix. i dont know anyone at my school who even bothrs with bc1q prefixed payments anymore"
legendary
Activity: 3906
Merit: 6249
Decentralization Maximalist
July 09, 2018, 12:28:25 PM
Quote from: Zin-Zang

 Smiley
Hmm,  Electricity costs money,  VPS costs money, the very definition of small in reference above is that the small hubs have less resources to draw upon.
A LN node does not use much electricity. It can work also in the background of your PC. It does not even have to be online 24/7 (most people have some device running at least for 10 hours a day), but If LN gets popular, LN software could be even installed on always-online devices like routers.

So no, I don't see why people need to rely on profits from LN.

I would not even qualify it as a "hobby", but simply as a part of managing your personal finances. Instead of dealing with online banking, you deal with your LN node ...

Quote
Competing with a large hub , which is making profits means the large hub can continue improving while the small hub continues to do less, until no one uses the hobby hub.
I still don't get it - if there's a small hub that charges a very small, or even non-existent fee and is not profitable (this fee is more of a "tip"), why should you use a large hub instead if he needs to charge you a relatively high fee to be profitable?

And small, non-profitable "mini-hubs" (e.g. "your well-connected techie friend") are not money transmitters, as they are not businesses.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 12:25:42 PM
If you don't like what the developers are doing, either make something better yourself or use something else.  Developers don't owe you anything.

gotta love and laugh at your mindset, basically our saying
'if you dont like the fact that devs have stiffled bitcoin. make an altcoin or go use another altcoin.'

now that is the most monarchistic totalitarian comment you coulld ever say.

how about stop kissing devs asses and letting them get away with stiffling bitcoin and instead kick thier ass to make it better for everyone.
devs owe the community to be responsible.. and its a damn shame there are so many sheep that just have a mindset of, leave the devs alone, let them do as they please

as for the 32mb and your obsession with BCH
ill say this.
imagine you had a PC and ther was a game you wanted to play that needed 1gb of RAM.
you would be the fool that would buy a PC with 1gb of RAM.

and then scream to the world there is no need for 16gb ram mothrboards because 1gb is enough
..
imagine you had to do some shopping. its winter. you would only buy that days food. you would not bother thinking about tomorrows needs. you will scream you only need to eat one sandwich today so its all you need.

think about it.
..
imagine you own a bus company. you buy a bus with only 4 seats because th average bus stop only has a queue of 4 people.


in short. 32mb blocks do not mean 32mb of data need to be filled/used befor something occurs. it means there is buffer, space, potential, room to grow,
even if blocks only utilise 1mb now. guess what. if it gets popular/busy tomorrow, it can cope with 32x more capacity without sudden fear of bottleneck. without panic of 2 year discussion of should it grow to 8mb.
its like a bus of 32 seats. ya most bus stops might lt on just 1 passanger. but if 7 people were waiting at the next stop. the driver doesnt have to refuse srvice and tell people to wait for another bus.. or spnds years debating if  2 seater or 4 seater bus is better...it just lets them in.
if theres 16 people waiting. no debate, just get onboard.

if you want to play your 1gb ram game, and then twitch stream it and have skype on along with having som music playing. you dont need to dbat geting a ram upgrade.. you just multitask to your hearts content.

research the word buffer
it does not mean the space needs to be filled to function.. it means extra room to cope with variation/chang of size without getting strangled so soon.

but anyway you go back to kissing devs ass and calling other networks the best thing since sliced bread.. because its actually you that does not like bitcoins mainnet to innovate, you are the one that wants people to use other networks.

you said it yourself you want me to go play with an altcoin and you want others to enjoy LN...
legendary
Activity: 3948
Merit: 3191
Leave no FUD unchallenged
July 09, 2018, 11:42:44 AM
The Bitcoin Cash community have their 32mb blocks, which is more than what they need. What is the problem? Why keep attacking Bitcoin?

It could have something to do with the fact that BCH's 32MB blocks are continually empty because there are hardly any transactions, so franky1 feels compelled to drum up support and recruit some new users by attempting to smear the thing people are actually using.  It's basically an aggressive form of marketing.

Consider that if all you're doing is you're listing the "selling points" of a Bitcoin clone with a single value altered (and some of the features missing), the sales pitch doesn't last very long.  So you naturally resort to attacking the clear market leader, because it's literally the only way you'd have anything left to discuss.

BCH apparently doesn't have enough interesting content worth talking about, which is why he has to resort to dramatic fantasy tales, with evil banking hubs and a malevolent Blockstream pulling the strings like a sinister puppeteer behind the scenes.  At least it sounds more intriguing than BCH, even if it's a total fiction.   Roll Eyes

im laughing first of all you sound so desparate that bitcoin is perfect that you are in denial of its issues. yet then you want to promote sidechains and offchains which by their own invention show that bitcoins mainnet has issues for these commrcialised side networks need inventing.

And if there is an issue, the solution is not to get rid of the people who freely offer up their own time and effort to find potential solutions to the scalability problem (rather than just ignoring the fundamental issue and pretending that infinite blocksize increases will magically fix everything without consequence).  The solutions proposed have now been accepted by the majority of the users and miners, so blame them.  There is nothing you can do to change that, so quit your infantile whining already.  

If you don't like what the developers are doing, either make something better yourself or use something else.  Developers don't owe you anything.  Get over yourself.  Life doesn't work like that.  I spent a long time thinking a larger blocksize was the way to go, but I'm mature enough accept that not everyone sees it that way.  So we (that's the users and the miners) went with the compromise of a larger blockweight via softfork, which incidentally does allow for more on-chain transactions (which you strangely keep failing to acknowledge  Roll Eyes ) and the introduction of Lightning.  

This is Bitcoin now.  Lightning happened.  It's not going away.  Not a single thing you can say or do will change that.  And best of all, it's optional.  No one is putting a gun to your head and forcing you to use it.  So it's time for you to grow up, be mature and accept that simple and undeniable fact.  

This.  Is.  How.  It.  Is.


once you admit that developers have stagnated innovation onchain to push these commercial networks. then you can kick the devs asses to stop concentrating on commercial sid services and get back to innovating and scaling the mainnet.

And what "innovation" do you think will solve scalability, exactly?  I hear all the critiques, but absolutely bugger all that sounds like a solution from you.  Come on, let's hear it then...  

Hint:  Blocksize increases are not "innovative".  Look up the definition of the word.

Developers are also working on Schnorr Sigs and MAST, which (like SegWit) also contribute towards on-chain scaling in an innovative way.  But you're too busy bawling like a spoiled brat to acknowledge that either.  Again, grow up.


sorry dude but it seems its you that loves alternative payment systems that are not done via bitcoins mainnet..

Good luck trying to make a Lightning payment without making a mainnet transaction first.  Reassure me one more time you actually understand how Lightning works, perhaps?  'Cause I'm pretty sure you don't with the nonsense you keep spouting.  

Ultimately, this is the road the users have decided we're going down, you can get off anywhere you like.  But then it stands to reason you can't be a loudmouth backseat driver from the pavement, so I'm sure you'll stick around and continue to be an annoyance to everyone else by blaming the developers for everything, even though they aren't the ones behind the wheel.  One of these days you might figure that out.
legendary
Activity: 4410
Merit: 4766
July 09, 2018, 10:33:26 AM
The Bitcoin Cash community have their 32mb blocks, which is more than what they need. What is the problem? Why keep attacking Bitcoin?

It could have something to do with the fact that BCH's 32MB blocks are continually empty because there are hardly any transactions, so franky1 feels compelled to drum up support and recruit some new users by attempting to smear the thing people are actually using.  It's basically an aggressive form of marketing.

Consider that if all you're doing is you're listing the "selling points" of a Bitcoin clone with a single value altered (and some of the features missing), the sales pitch doesn't last very long.  So you naturally resort to attacking the clear market leader, because it's literally the only way you'd have anything left to discuss.

BCH apparently doesn't have enough interesting content worth talking about, which is why he has to resort to dramatic fantasy tales, with evil banking hubs and a malevolent Blockstream pulling the strings like a sinister puppeteer behind the scenes.  At least it sounds more intriguing than BCH, even if it's a total fiction.   Roll Eyes

im laughing first of all you sound so desparate that bitcoin is perfect that you are in denial of its issues. yet then you want to promote sidechains and offchains which by their own invention show that bitcoins mainnet has issues for these commrcialised side networks need inventing.

once you admit that developers have stagnated innovation onchain to push these commercial networks. then you can kick the devs asses to stop concentrating on commercial sid services and get back to innovating and scaling the mainnet.

also i laugh that you think because i do not fear admitting there are issues that somehow makes me a lover of alternative networks.. sorry dude but it seems its you that loves alternative payment systems that are not done via bitcoins mainnet..

first sign of fixing a problem is admitting there is one.. think about it
Pages:
Jump to: