Topic: Decentralisation is harder than you think - page 3. (Read 740 times)

Sounds like a bit of a leap.  And don't even start with your "meander" catchphrase, since you brought this discussion into the thread.  I can discuss points which you raise.  The above belief is all based on one massive assumption that users actually agreed with it and ran code which supported SegWit2x.  Even if Core had released such code, doesn't automatically mean users would have chosen to run it.  Like it or not, users are actually an important factor and can't be dismissed under the absurd belief that devs make all the decisions.  Also, I thought you said it's "not about blocksizes", so why are we back to talking about SegWit2x for the umpteenth time? 



No one's disputing that you have issues, but businesses are free to do whatever they like.  Some of them chose to support SegWit2x, some of them didn't.  Some businesses are developing for Lightning, some of them aren't.  Some businesses are doing things you or I might or might not support and be funded you people that you or I might or might not approve of, but they're free to do it anyway because permissionless.  Businesses are not the sole factor to consider, though.  The crux of the matter is that you don't like the code that other people and companies are running and that's just something you'll have to find a way to come to terms with.  If things really were as terrible as you keep making them out to be, why are businesses, users, speculators, etc all perfectly content to keep running this code?  It must be a conspiracy, right?  Couldn't possibly be that they're happy with the path we're on.



If Lightning does somehow have a detrimental effect on Bitcoin, users could (believe it or not, since it's just such a difficult scenario to even begin to comprehend    ) simply not use it.  Because if it doesn't benefit users, why would they use it?  At the end of the day, people are going to do whatever they damn well please.  That's the best part about decentralisation.  And that's fine by me.  So by all means keep telling your spooky campfire tales about the Blockstream boogeymen.  Some of us aren't as scared by them as you seem to be.  If Blockstream have plowed money into something that fails, it's not very good business sense on their part.  So perhaps it would make sense for them to make something that is good for users.  Had that thought ever occurred to you?



I can see how someone might come to that rather strange conclusion if they began their reasoning from the standpoint that devs made all the decisions.  But, since that's not even remotely the case, anyone who bases their views in reality will understand that those securing the chain made the choices and ran the code that led us to where we are now.  Try starting from a premise that isn't fundamentally flawed.

"Decentralisation is harder than you think"

It really is.

I don't know what the challenges are for programmers in creating code but they will have to get around
our human instinct to "game the system"... even systems we agree with.
I would like to see blockchain voting---1 person=1 vote but it would need to be resistant to...
us humans ;-)

its not about pumping fee's up. its about controlling more then one hop of a route


imagine users setting up independant broadcasted channels.
to have a network where 1 million users are connected involves
hop:
each node being a full node, paying onchain fe's to open/close. AND have
3 channels(onchain fee's x3) and be possibly 20 hops away (upto 20LN fee's per payment)
5 channels(onchain fee's x5) and be possibly 10 hops away (upto 10LN fee's per payment)
7 channels(onchain fee's x7) and be possibly 8 hops away (upto 8LN fee's per payment)
11 channels(onchain fee's x11) and be possibly 6 hops away (upto 6LN fee's per payment)

would you want to split your funds up into 3-11 allotment and hope that all the nodes on a route are all online just to spend one of those allotments and then still be left with the burden of paying broadcast fee's to move allotments when closing/opening channels

..
play scenarios out in your mind. or use some pen and paper and actually run simulations (thus not having to risk actual spending)
you will shock yourself how many people end up choosing the convenience of just throwing all their funds into a custodian knowing they get convenient service. without the hassles..


also ask yourself why do you think the blockstream devs removed the fee priority mechanism and instigated the fee war and really are pushing onchain fe's to be way over 10cents and hopes that $2 fee's become the norm by saying bitcoin is not fit to pay for coffee($3)
its because of these reasons
1. USERS with more than a few channels and set up as masternodes wont want to get paid just a few cents a day from their few channels. so fee's would be cheaper than onchain but wont remain sub pennies for long. it would end up being USERS that push fe's up for their "independant" routing

2. the commercial side will, due to the 'convenience' affect have many hundreds of thousands/millions of users(check out DCG's portfolio of companies) so add it all up.
if a oligarchy charge 0.1cent per hop with a system of being only 6 hops.=$0.006 per payment * million =$6k a day=$2m a year

now imagine with onchain fee's being $2 and greedy USERS wanting to charge 10cents. as its still 20x cheaper than onchain... meaning the oligarchy can outbid them at 3cents=$180k a day=$65mill a year
and thats just from 1million users making one payment a day..... think about it
imagine if you wanted to trade on coinbase and then arbitrage to circle and then arbitrage to another exchange all via LN.
thats 3 payments a day. which is only going to be cheapest/convenient by using the commercial factory/hub model
thats near on $200m a year DCG get from just 1million users

..
i know what your thinking next.
if a oligarchy can make $2mill a year. so can users, but no
many users are not paying other users. most transactions go into services. and as i shown before. those services are walled up by their own team of hubs and watchtowers.

knowing an independant hop model of users is 6-20 hops. for a user who has a 3cent fee per hop to make minimum wage of america. a user would need to have 200 payment looping through it an hour.
users wont have the funding to hot-potato 200 payments an hour and then pay onchain fee's to shift around the allotments per channel

so while the majority are 'convenience' using the commercial hub model. hardly anyone would be using the independent hop model as their destinations are not close by/cheap to reach/funded 24/7. as the only way to get near a service with low fee's is to use the system the service has set up

you think they can generate hundreds of millions USD from lightning routing fees? at small fractions of a cent each, you must be really bullish on the lightning network's growth!

i only see a hub-and-spoke model developing to the extent that it's not cheaper to route around price gouging hubs. i don't think anyone will have that much success artificially pumping fees up.

'lost'? 12.5
nope
they just didnt gain.
no one expects everyone to win(gain), and then have funds taken off them(lose it). it has always been the case that only 1 person gets the gain.. so those who dont get their block immortalised are not losers. as they were not promised/given anything to lose.. its just a 'better luck next time' thing


what you dont realise is that with factories, hubs and watch towers. thats exactly how things will play out
to even gt within 4 hops of an exchange you would have to be one of their customers which is using them as a factory
imagine there is coinbase exchange they wont want millions of customers direct IP linking to them. so they will set up hubs around them to take the strain. also because millions of users wont want to be masternodes and just want to be mobile wallet apps. they will use watchtower servers to connect their apps to



now imagine people deposit real bitcoin to coinbase. and then coinbase write offchain/unconfirmed 12 decimal IOU transactions with their hubs. and the hubs make (12 decimal unconfirmed) transactions to the watch towers and the watch towers make them for the users.
so now the users have "funds". (12 decimal unconfirmed 3 levels away from blockchain UTXO) which the channel parties can agree on, because of the chain of unconfirmed 12decimals all lead back and can aggregate back to a UTXO with coinbase happily

now imagine a green user wants to pay a blue user.
count how many hops there are.
now imagine red, yellow and purple are all owned by DCG.
when someone makes a payment they are not paying just under 1 cent once. they are paying under 1 cent per hop. and if it takes 6 hops to get to another individual. thats 5x fees for DCG

then because its based on factory funded channels instead of broadcast funded, to open and close channels users pay their watchtower and hub and coinbase a fee to close. all for the convenience of not having to broadcast onchain. and then coinbase, its hubs and watchtower then re-write new offchain 12 decimal transactions to allow users to re-open new channels
so those red/yellow and purple get open/close channel fee's too...(in other words DCG gets paid 5x fee)

...... or users can avoid factories and the inner branches surrounding coinbase and hope they can self broadcast their own independant channel open and connect to someone thats hopefully not more than 10 hops away from coinbase(which even so. they still end up paying DCG more than just 1x fee) once the payment gets to the fringe of the inner branches surrounding coinbase if the particular route has to go through the inner branches

its the same "routing" as banking.(in america the links between banks are called routing numbers)
town banks link to regional banks which link to HQ. that way the HQ doesnt handle millions of customers direct, but each town bank(watchtower) manages ~8000 customers yet the real 'value' is held at the HQ(fortknox)

in the UK. banks dont care much about people swapping banks, because ultimately its all owned by either Bank of england or bank of scotland
in the UK. phone line providers dont care about people swapping providers, because ultimately its all owned by openreach
in the UK. gas providers dont care much about people swapping providers, because ultimately its all owned by transco
in the UK. electric providers dont care much about people swapping providers, because ultimately its all owned by transco

yea a few percent of cases will have independent channels. but when it
costs 10x+ fees due to many hops.
involves praying that the independent channels are online 24/7,
involves knowing it will cost you onchain fee's to closeout,
involves hoping the independent routes have funding to even allow payment hops..

eventually people give in and just play by the oligarchy game for convenience... and thats the point of their game

im not interested in the altcoins.

but also its not that devs get funded. its the actions that occur due to their funding.
it would have been better to not have a single team of devs funded by the same group. especially when that same group then make a roadmap, not just for their implementation, but for the direction the NETWORK should take. thats my issue

if core just wanted to be a level playing field implementation along side other brands. where core would have actually come to some united agreement with the community in 2015 about many different things we would have advanced the bitcoin network alot sooner and alot more diverse people would have seen alot more diverse features implemented.

did you know that at the original late 2015 meetings if core actually held onto a lengthy agreement of then a segwit2x proposition. blockstream would have got segwit implemented sooner AND the rest of the community would have got more utility for legacy transactions. other features would have/could have ben implemented too... but no. it had to be the blockstream paid devs way or no way. which then causes the community divides

w would have also seen many businesses not get segregated off the network, we would have seen less coders making altcoins but instead happily innovating bitcoin under their own full node brands. bitcoin would have been alot stronger and more united by having more diversity

there are literally hundreds of businesses in the crypto industry (thousands, but some are home/hobby business) so its not that hard for talented coders to find employment. but when the main devs are all circled into one group. and the so called 'innovations' of bitcoin do not positively affect bitcoin, but are done to promote alternative networks designed to push people off the bitcoin network. thats where i have issues

LN is a separate network for multiple coins. its not a bitcoin centric network. its only promoted as a bitcoin centric network purely for investment by screaming the word bitcoin in the same sentance as LN.
LN's design is to have blockstreams investors (DCG and digital garage) b the main factories/hubs/watchtowers. where they get the fee's from controlling routes and channel opening/closes. and its all because blockstream got paid over $100m and need to offer a solution to repay their investors. (ever wonder why there was a november segwit deadline and then a sudden price trigger event right at the dates blockstream got paid a tranche of funding the same week in 2017)

if the devs were more diverse where their idea's were more independent and not controlled by some sugar daddy investor. things would be different

@franky1

I like reading your posts even if I often disagree with you but you're too much against Blosckstream. Now how would be Bitcoin without it? A total mess maybe, what do you prefer?
Something I never understood from you is you still own bitcoins, knowing your opinion I wonder what do you expect so? Or you secretly moved to Btrash?

I think we should be talking about at least two types of decentralization: that of mining and of possession. Mining is indeed dangerously centralized in the hands of a small bunch of people whom I don't find trustworthy. Is it bad ideologically? I guess it is. Does it have major impact on cryptos? I don't think so. The thing is that miners can't do much. They don't influence the price directly, and they can just keep mining or stop mining. If they keep mining, nothing changes. If they stop mining, some still will and the difficulty rate will probably have time to adjust (if a half stops mining at once, the rate will most likely adjust in about a month anyway). If we turn to holders and traders - here I'd say we have a pretty positive picture of a seriously decentralized market. If we look here, we can see that while there are some big holders of bitcoin, at least a couple of thousands of bitcoin wallets have to act at once to make a big difference.

It's a sure thing that decentralization is a lot harder than centralization because decentralization involves a distributed system and building such things are harder than building centralized applications.
But anyway, as the technology is improving day by day, decentralized applications are increasing. Programming languages like Solidity has been created to build smart contracts and dapps.

There are definitely incentives against egregious majority miner attacks by pools, but it still represents a nonzero threat.

Let's not forget that in its heyday, the operators of GHash.IO engaged in double spending attacks. Here is one such example. The less hash rate in the hands of any individual operator, the better.

I agree with the opinions of Frank1 and ETFbitcoin, the mining process is not part of a decentralized system. True, what Franky said is that mining only functions as a system supervisor, so that when someone tries to make a duplicate of bitcoin it will be confirmed by the mining system. I think the term decentralization is final that the system is transparent, is not regulated by any authority (CEO) or institution, and without intermediaries.
Bitcoin is able to answer all of that. As for the altcoin case, I feel doubtful. The article mentioned both do not have the right data and logic. I think they make arguments based on price perspective or price manipulation, not on the bitcoin decentralization system.

'Remember the 51% apocalypse in 2014?'

That was some shouting on Reddit and then once it looked like a danger it instantly faded away.

'Bitcoin’s blockchain is constructed by 19 mining entities'

How can someone be asked their opinion on this when they don't even know what a mining pool is?


Bitcoin is the one of the extreme few still attempting to remain decentralised. Its developers, users and miners at least have the gumption to realise how important it is, even when its implementation is far from perfect. Most other projects never had it in the first place or gave it up.

Certainly. Today if we pay attention to the ranking, we will see that some centralized moneys are in the top 10 and what should be understood is that people just concerns is the money that can be used today and more is better. The rest is unnecessary anyway.

Seems like you ignore the point in OP and continue to define "decentralization" in your own terms, which is unclear. I'm not really sure what kind of decentralization we're talking about, is it: the power to change the rules, how new money come into circulation, deciding which transaction is valid, and so on.

Either way, I believe this topic is broader than we think and as pooya said, sometimes people mix imagination and reality to support their stance.

unfortunately the internet nowadays is filled with idiots who keep creating websites and publish a lot of bullshit.
the first article is from someone who is caught up in the Bitmain drama and is just creating a clickbait based on his imagination and is using dramatic statements such as "the 51% apocalypse in 2014" followed by talking about "bitcoin" to make it look legit. a quick search can tell you that the thing in 2014 was literary nothing and it was in fact from GHash.io and there was no 51% attack.
and both articles suffer from the same naivete of not knowing the difference between a miner and a mining pool. they think they are the same which is why they make stupid statements such as "Bitcoin’s blockchain is constructed by 19 mining entities". of course most part of these two are intentional drama feeding on the lack of understanding of the readers.

regarding your title, of course decentralization is hard. in fact it is more like a non stop battle to keep it that way. but when it comes to mining, you can only call it centralized if  a handful of people control a large amount of the hashrate not the pools that only connects the actual miners from all around the world.

Decentralization is harder because we are talking about an autonomous project, in other words, the engine should work alone to have a decentralized project. That's the success key on this kind of projects.

If you think about a business like Microsoft or facebook, with all the control they try to handle is impossible to make from those decentralized projects. But Bitcoin can work that way because it doesn't need a boss.

bitmain dont make the hardware.
comparison.. if bitmain was Asus. then TSMC would be AMD

bitmain dont control 50-80%
bitmain dont operate alot of hardware on behalf of clients
comparison.. if bitmain was a state the mining farms would be private land owner just paying taxes

It's not this simple. Bitmain operates a lot of hardware on behalf of clients, who can pull the plug at any time. The Bitcoin Cash war showed they have far less mining resources than previously thought. Miners have incentive to leave their pools if they're threatening the protocol.

Competition in the ASIC manufacturing sector would be welcome, though. What's the breakdown on those numbers? Bitfury is still relevant, right? I know GMO was a major disappointment and I've heard nothing good about Ebang. Who else is there? Any news on that horizon?

there are 20+ pools
there is one main brand of node

even the top pool which people call bitmain is actually not one team. there are multiple facilities in multiple countries with multiple mindsets. did you know that the blocks tagged as just 'antpool' have half a dozen facilities. this can be fact checked by noticing the half a dozen DIFFERENT coin reward addresses they re-use. and also double fact checked because some of the facilities are accepting of segwit, some are not. which shos diverse management. and then when you geolocate th facilities you will see they are spread out over san fran, georgia, iceland, hongkong, mongolia

yet. when it comes to developers
one team called core. REKT campaigning any diversity off the network, mandating people follow cores roadmap or be found thrown off the network

but hey. if you think losing a couple transactions for an hour until they get reconfirmed by another block is a bigger threat than a dev team that implement changes without consensus.. then let that be your weakness in understanding the security of decentralisation

mining is about collating data and then locking it.

thats it

its then the nodes that have the actual power. they can reject blocks that dont fit the rules.
in short the only power mining pools have is to decide which transactions to include or exclude.

pools dont have the power to change the rules.

the power to change the rules remains with developers that code the rules and distribute their code.
if developers have code that bypass users ability of choice. EG remove the consensus opt-in of new rules by making things mandatory/compatibility bypassed. then that is the bigger threat

many people think bypassing consensus is great as it makes changes easier/faster to administer. but the downside is... guess what. making it easier/faster to make changes.

pools cant make changes
devs can

learn who the real threats are
especially when the devs hate having opposing teams to keep them inline
especially when the devs hate having to play using consensus and prefer by-passes
especially when the devs hate having to work on a lvel playing field with other teams and instead want to be the only core/reference/fullnode/implementation

as for those assuming mining is a threat due to "50% china".. thats just the comedy FUD of racists
firstly the hashrate distribution is actually more diverse then the racists chant