Bitcoin Forum
November 13, 2024, 03:35:51 AM *
News: Check out the artwork 1Dq created to commemorate this forum's 15th anniversary
 
   Home   Help Search Login Register More  
Pages: [1]
  Print  
Author Topic: Are most new blockchains centralized?  (Read 259 times)
OriginTrain (OP)
Full Member
***
Offline Offline

Activity: 406
Merit: 174


View Profile
January 27, 2018, 12:23:24 PM
Merited by mprep (1), Welsh (1), ABCbits (1), Jet Cash (1)
 #1

Most new blockchains like to advertise themselves as decentralized, yet the majority of them rely on centralized actors to maintain their hype, value and sometimes even their functionality. In accordance to the "weakest link" theory, all it takes is one centralized point in the system, and all the "decentralization" becomes garbage hogwash semantics. Let's go through a few examples of the top 10 blockchains/tangle-thingie-whatevers:

1) IOTA. This is not decentralized since it requires the centralized Co-ordinator to prevent certain types of attacks. In addition I believe one of the core devs mentioned putting backdoors in the code to prevent hard forks without the team's permission. Both of these actions show the IOTA tangle is not currently decentralized.

2) ETH. However much you want to argue semantics, ETH succumbed to rule of the mob over "code is law". The original ETH blockchain (Ethereum Classic) is not victim to this accusation, however ECT is not within the top 10 market cap coins and this post is referring to the leading blockchain techs. ETH foundation (centralized) has demonstrated itself to be happy to hard fork when needed, rubbishing the idea of decentralization just like IOTA.

3) Ripple. No explanation needed.

4) EOS/ADA/NEO. In reality, these projects all depend upon their core public figures to maintain their value (block.one/Dan for EOS, academic figures for ADA, some chinese team I guess for NEO). If these teams were to initiate a hard fork like ETH did, then the original immutable blockchains on any of these platforms could easily go the way of Ethereum Classic.

The only blockchain I can think of that has the greatest chance of sticking to its immutability is Bitcoin. Yet it's now plagued by serious issues, such as high transaction fees (I realize LN etc.. but the current reality is what it is), and lack of on-chain upgrade potential.

With all the hope we have for new 2nd and 3rd gen blockchains to become the new leader in the immutable, decentralized future.. how can any of these contenders to BTC be seriously considered? Are they not centralized and dependent upon the actors behind them?

It gets me thinking that if ever there was a replacement for Bitcoin, it needs a central actor like Satoshi Nakamoto - someone who develops the foundational tech - but then gets the hell out of the way and vanishes afterwards, so that key actors can no longer have so much influence over the blockchain, but rather the blockchain itself and its immutability will be considered of the highest importance over these centralized actors.
Jet Cash
Legendary
*
Offline Offline

Activity: 2828
Merit: 2472


https://JetCash.com


View Profile WWW
January 27, 2018, 12:59:32 PM
 #2

I'm having trouble getting my head around the wide variety of "blockchains", and the associated terms. Even the Bitcoin description of a "distributed blockchain" is incorrect in my opinion. I think it is a replicated blockchain. That is a great concept with a slightly misleding description. I see there are attempts to move on from blockchains to mesh networks to create real distibuted ledgers, but I haven't had a chance to research those.

To return to your question though. I can't see how a blockchain can be decentralised without a proof of work to create new blocks. In some cases centralisation is essential - the Royal Mint Gold chain for example. This is based on physical gold stored in a central location. The new FedCoin that will replace the dollar will be interesting, especially if they try to make that into aglobal currency.
stantpro
Full Member
***
Offline Offline

Activity: 280
Merit: 101



View Profile
February 01, 2018, 05:51:01 PM
 #3

If block-chain is centralized, it means the notion of overcoming control
and stringent laws remain elusive.The essence of block-chain technology is
to ensure decentralization of transactions in the platform such that no body
has absolute control or centralized administration to determine the policies
or bureaucracies of the FIAT world.
kaar
Full Member
***
Offline Offline

Activity: 121
Merit: 123


View Profile
February 01, 2018, 11:57:55 PM
Merited by mprep (1)
 #4

I touched this issue a little in another post here in this board. The question that needs to be asked is how do you define a decentralized network? An absolute decentralization like you suggest means that nobody has substantial influence on the network in any aspect. This goal is very hard to achieve as it's almost impossible to prevent people from colluding (just see how centralized bitcoin has become when it comes to the distribution of hash power).

Even if you do manage to create a complete decentralized network, I think it has nore drawbacks than advantages. Take bitcoin for example. It becomes pretty much impossible to implement necessary changes to the code. If you take it to the extreme, in a complete decentralized blockchain no bugs can ever be fixed as nobody holds enough power to change the code.

As long as the code is open source, I don't see a big problem with hard forks. Sure in blockchains like ETH, where the devs have a big influence, hard forks can happen once in a while. If most people abandon the old chain and move to the new fork, that means they think the change is good, or they would have stayed on the old chain. BCH is a good example, people who thought the fork was necessary moved to BCH and the others stayed on BTC.

To conclude, I think a certain level of centralization is invetiable in any network. The important part is in what aspects this centralization actually affect the network. In Ripple's case for example, the influence is absolute and I wouldn't call it a decentralized network. In ETH though, the influence of the devs comes down to potential hard forks in the future, which is something that I can live up with.
OriginTrain (OP)
Full Member
***
Offline Offline

Activity: 406
Merit: 174


View Profile
February 02, 2018, 04:09:15 AM
Merited by mprep (1)
 #5

I touched this issue a little in another post here in this board. The question that needs to be asked is how do you define a decentralized network? An absolute decentralization like you suggest means that nobody has substantial influence on the network in any aspect. This goal is very hard to achieve as it's almost impossible to prevent people from colluding (just see how centralized bitcoin has become when it comes to the distribution of hash power).

Even if you do manage to create a complete decentralized network, I think it has nore drawbacks than advantages. Take bitcoin for example. It becomes pretty much impossible to implement necessary changes to the code. If you take it to the extreme, in a complete decentralized blockchain no bugs can ever be fixed as nobody holds enough power to change the code.

As long as the code is open source, I don't see a big problem with hard forks. Sure in blockchains like ETH, where the devs have a big influence, hard forks can happen once in a while. If most people abandon the old chain and move to the new fork, that means they think the change is good, or they would have stayed on the old chain. BCH is a good example, people who thought the fork was necessary moved to BCH and the others stayed on BTC.

To conclude, I think a certain level of centralization is invetiable in any network. The important part is in what aspects this centralization actually affect the network. In Ripple's case for example, the influence is absolute and I wouldn't call it a decentralized network. In ETH though, the influence of the devs comes down to potential hard forks in the future, which is something that I can live up with.

Interesting points that I agree with mostly. In the ETH example you cited though, the hard fork wasn't to introduce a core code upgrade, but rather to reverse a transaction. In my opinion no matter how bad it is, this is a fatal precedent that means I will never consider ETH decentralized. Imagine one day in an Orwellian society if the "decentralized" network can be reversed like this because the transaction wasn't theft but a thought crime. From a moral perspective of course the thief shouldn't get away with the money, but there's much, much bigger issues at stake here than a heist, the bigger picture here is the need to create a transfer of wealth system that cannot be tampered with by anyone at all.

I agree that hard forks should be possible for code upgrades, but definitely never, ever for a transaction reversal, transaction modification, or updating coin supply. ETH showed itself to be no better than Ripple. I hope future blockchains in the 3.0 realm will learn from this and whatever necessary, centralized hard forks they conduct, will never involve transactions.
kaar
Full Member
***
Offline Offline

Activity: 121
Merit: 123


View Profile
February 02, 2018, 07:02:26 PM
Merited by mprep (1)
 #6

I completely agree that hard forking in order to modify transactions is very bad practice, and ETH were wrong with that move. However, I don't think it has anything to do with centralization. If one person or organization can change the code as he pleases and nobody has any control over it, then yes that means centralization. But if the code is open source and the majority of the community decides to fork to undo a transaction (and in ETH case it was a vast majority), it's another problem.

You are talking about the moral perspective and I was talking about the technical perspective. Technically, the devs of ETH can't do changes in the code without the consent of the majority of the network. That means it's not centralized in this aspect.

ETH actually is a good example that decentralized network doesn't always protect against bad decisions, if they are made by the community.
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!