A New Quest for Decentralization

[kushti]

"While most active conversations today in the space are about wider adoption of the blockchain technology (which often means selling out to the Wall St.) and competition with systems like Visa and Mastercard (which often means giving up with decentralization or introducing unclear security assumptions in the name of efficiency), there is the obvious need to revisit the roots of the cryptocurrency movement, which are mostly about decentralization. Many questions to be answered in clear here. Is it okay when 90% of mining power in Bitcoin can gather in one room? Is it okay when 2 or 3 mining pools control majority of hashing power, so can do censorship? Is it okay when almost all the new nodes avoid processing a blockchain from its genesis block? Is it okay when a Proof-of-Work coin developers are doing a hard-fork changing the consensus algorithm to make it GPU-friendly again? Can we summarize all the issues with decentralization? Can we cover most of issues with technical means?"

With this motivation I've made an article on technical tools we can do for better decentralization (of PoW blockchains): https://ergoplatform.org/en/blog/2019_12_12_new_quest/ .

As a core developer and researcher, not a philosopher, I leave room for social aspects to others.

What do you think guys? How can we do better?

[key4co.in]

Is it okay when a Proof-of-Work coin developers are doing a hard-fork changing the consensus algorithm to make it GPU-friendly again?

The change from POW to POS by some devs, for instance Ethereum' might be decided after analyzing the advantages and disadvantages involved. No doubt POW algorithm proffers a more secured and decentralized system, deters cyber attacks and upholds a trustless and distributed concensus, it also has its major disadvantages. For instance the energy requirements for mining is much. Proof of stake on the other hand even though will make consensus mechanism totally virtual, validators get block reward depending on their staked coins and there is no such energy requirements. However, I really like the point raised by OP and will drop my opinion on "how we can do better" after my research. Decentralization still remains a key factor in crypto and shouldn't be thrown under the carpet.

[kushti]

Is it okay when a Proof-of-Work coin developers are doing a hard-fork changing the consensus algorithm to make it GPU-friendly again?

The change from POW to POS by some devs, for instance Ethereum' might be decided after analyzing the advantages and disadvantages involved. No doubt POW algorithm proffers a more secured and decentralized system, deters cyber attacks and upholds a trustless and distributed concensus, it also has its major disadvantages. For instance the energy requirements for mining is much. Proof of stake on the other hand even though will make consensus mechanism totally virtual, validators get block reward depending on their staked coins and there is no such energy requirements. However, I really like the point raised by OP and will drop my opinion on "how we can do better" after my research. Decentralization still remains a key factor in crypto and shouldn't be thrown under the carpet.

Not talking about simplicity and other technical advantages of PoW:

PoW consumes energy but as a result, a crypto-commodity is produced. That commodity nature helps with economic security, not so easy to cheat and control, and there's some minimum price over long-term (defined by production cost).

[shinratensei_]

Is it okay when a Proof-of-Work coin developers are doing a hard-fork changing the consensus algorithm to make it GPU-friendly again?

The change from POW to POS by some devs, for instance Ethereum' might be decided after analyzing the advantages and disadvantages involved. No doubt POW algorithm proffers a more secured and decentralized system, deters cyber attacks and upholds a trustless and distributed concensus, it also has its major disadvantages. For instance the energy requirements for mining is much. Proof of stake on the other hand even though will make consensus mechanism totally virtual, validators get block reward depending on their staked coins and there is no such energy requirements. However, I really like the point raised by OP and will drop my opinion on "how we can do better" after my research. Decentralization still remains a key factor in crypto and shouldn't be thrown under the carpet.

Not talking about simplicity and other technical advantages of PoW:

PoW consumes energy but as a result, a crypto-commodity is produced. That commodity nature helps with economic security, not so easy to cheat and control, and there's some minimum price over long-term (defined by production cost).

How easy POW coin to be cheated and controlled will depend on how much hashrate and active miners in the network. Bitcoingold has not so many miners and it can easily be cheated through 51% attack.

It's okay when 90% of hashrate power has gathered in one room as long as there was no collusion and that will create 51% attack to the bitcoin. I just assume in this case even almost all of hashrate gathered in one room but it was running by the different parties. It's not so good or bad to see 2 or 3 parties are controlling the network. Basically, as long as 51% hashrate were not gathered by a party and 51% attack will not happen. When 51% happen and there was a possibility for validator censorship to happen but that's not a big problem. It's okay for POW dev to do a hard fork to make the coin becomes GPU-friendly again and I think this change will on the algo that used by the coin.

[Mighty_crypt]

Both POS and POW algorithms are good, they both have their dark sides but is there anything wrong if a coin use the two algorithm at the same time? There are few coins like that in crypto space that you can mine with mining rigs and at the same time you can stake too

[kushti]

Both POS and POW algorithms are good, they both have their dark sides but is there anything wrong if a coin use the two algorithm at the same time? There are few coins like that in crypto space that you can mine with mining rigs and at the same time you can stake too

Combination of PoW and PoS is much harder to analyze (and so to be sure in security) that PoW and PoS separately. I'm coauthoring a paper on TwinsCoin, hybrid PoW/PoS protocol https://eprint.iacr.org/2017/232 , but I think that the protocol is not practical (at least, as described in the paper). 

[Ucy]

Is it okay when a Proof-of-Work coin developers are doing a hard-fork changing the consensus algorithm to make it GPU-friendly again?

The change from POW to POS by some devs, for instance Ethereum' might be decided after analyzing the advantages and disadvantages involved. No doubt POW algorithm proffers a more secured and decentralized system, deters cyber attacks and upholds a trustless and distributed concensus, it also has its major disadvantages. For instance the energy requirements for mining is much. Proof of stake on the other hand even though will make consensus mechanism totally virtual, validators get block reward depending on their staked coins and there is no such energy requirements. However, I really like the point raised by OP and will drop my opinion on "how we can do better" after my research. Decentralization still remains a key factor in crypto and shouldn't be thrown under the carpet.

I guess everyone was too focus on competition thereby neglecting the most important things.
I think it's important to have a decentralized system where most members are part of the decision making, else crypto will be consumed by things it was created to avoid/solve.

I don't see why over 100,000 nodes for example, can't elect some representative nodes manned by the most principled, best & anonymous people to scale a Blockchain.     Afterall, the representatives can be quickly/automatically removed by bots and members/many nodes once rules(especially rules based on crypto principles) are violated. Difficult legal issues need to be discussed extensively & transparently by members who are very experienced in the issue (while most members observe) before any decision is taken.

PoW still OK if it doesn't consume too much energy or encourage centralization

[ansi]

I don't care much about decentralization personally, all i want is transparent & profitable blockchain for us the citizen with the lowest Fees possible.
Decentralization is now kind of hard to mass adopt or spread among financial institutions, those guys wanna know where you spent every single penny of yours.