How about placing miners into groups where each group is similar to a bank?

[Peter11kz]

From what I've read about "block chains", how about changing the Bitcoin protocol to the following:

1. Placing maybe every 10,000 to 20,000 miners into individual groups (each group acts like a bank) with their own address so that groups could communicate with each other (similar to two banks transferring money between each other).

2. A limited number of wallets say 10 million customers are assigned to each group, each time all groups are full of customers then a new group is created (with a new address).

3. Each group has their own block chain different from another group's block chain (yet still secure because say 20,000 miners all have the same block chain) so that when a transfer occurs either within the same group or between only two groups then each block chain would only need to retain the wallets (and transfers) of its own customers.

Say there are 10,000 groups and only 1  (1 block chain) to 2 groups ( and 2 block chains) are used per transaction then this would free the other groups and therefore provide the faster transfer time and smaller transfer fees that everyone is looking for.

The downside would be that a wallet is stored in a block chain of each group which is say 20,000 miners but this would be sufficiently secure same as storing your money in one bank instead of all the banks in the world?

[Charles-Tim]

This can not be possible, having a single blockchain is better, having multiple different blockchains can make 51% attack possible. Also, you can not compare bitcoin and its blockchain to banks, they are completely and entirely different, which is another reason it can not work. Bitcoin is not closed as banks but open to public and transparent with open source codes and decentralized network.

[ranochigo]

As we have explained in the previous thread, miners are not entities that can be regulated on a protocol level. There is no reason why you should partition them into groups; they are solely the ones finding blocks with sufficient PoW and building the blockchain. Schemes like that would overhaul the entire Bitcoin and it would be better for anything like that to be implemented on an altcoin instead, though I don't really understand how that would differ from a traditional bank.

You could try to understand how Bitcoin actually works first before suggesting ideas. It would be far easier and more conducive for discussions that way.

[Peter11kz]

From what I've read about "block chains", how about changing the Bitcoin protocol to the following:

1. Placing maybe every 10,000 to 20,000 miners into individual groups (each group acts like a bank) with their own address so that groups could communicate with each other (similar to two banks transferring money between each other).

2. A limited number of wallets say 10 million customers are assigned to each group, each time all groups are full of customers then a new group is created (with a new address).

3. Each group has their own block chain different from another group's block chain (yet still secure because say 20,000 miners all have the same block chain) so that when a transfer occurs either within the same group or between only two groups then each block chain would only need to retain the wallets (and transfers) of its own customers.

Say there are 10,000 groups and only 1  (1 block chain) to 2 groups ( and 2 block chains) are used per transaction then this would free the other groups and therefore provide the faster transfer time and smaller transfer fees that everyone is looking for.

The downside would be that a wallet is stored in a block chain of each group which is say 20,000 miners but this would be sufficiently secure same as storing your money in one bank instead of all the banks in the world?

Same as trying to have Bitcoins created on planet Earth1 transferred to Bitcoins on planet Earth2:

Since they are 2 independent entities (2 groups), the Bitcoins from Earth1 is converted to US dollars then from US dollars is converted to Earth2 Bitcoins and stored in a wallet on Earth2. An intermediate conversion to US dollars is pointless therefore the only limiting factor is that Earth1 Bitcoins were not created on Earth2....but you would have the same level of security as when every single computer on Earth gets automatic windows software update every day from a THRUST worthy address such as Microsoft.com (instead of NorthKorea.com). Otherwise all of Earth's computers would presently be infected by viruses but they are not.

The only security breach that I see is when picking random 20,000 (or more) miners into 1 group is to make sure their addresses (IP addresses) are not fake and instead are all coming from one corrupt source such as North Korea....?

[ranochigo]

but you would have the same level of security as when every single computer on Earth gets automatic windows software update every day from a THRUST worthy address such as Microsoft.com (instead of NorthKorea.com). Otherwise all of Earth's computers would presently be infected by viruses but they are not.

That alone would defeat the purpose of Bitcoin won't it? Making all of the computers ping certain centralized server is not good. There is really no such thing as "computer security", there are tons of zero day exploits which could be used by any adversary before it gets patched.

From what I can tell, you're basically describing something like a centralized banking system but instead of bank officials, miners are the ones that regulate this. It would be good if you could read up more about Bitcoin first and its derivative protocols. I think Lightning network would interest you with this particular aspect that you're talking about.

[o_e_l_e_o]

the only limiting factor is that Earth1 Bitcoins were not created on Earth2....but you would have the same level of security as when every single computer on Earth gets automatic windows software update every day from a THRUST worthy address such as Microsoft.com (instead of NorthKorea.com).

Who says Microsoft are trustworthy? I don't trust them.

What you are describing involves everyone placing their complete trust in a centralized third party. This is exactly the thing bitcoin was designed to combat.

The only security breach that I see is when picking random 20,000 (or more) miners into 1 group is to make sure their addresses (IP addresses) are not fake and instead are all coming from one corrupt source such as North Korea....?

How are you going to "pick" the miners in a decentralized manner? How are you going to prevent other miners from mining in a specific group? What's to stop all the miners from one group moving to another group and 51% attacking them?

[HCP]

May I ask what you think would actually be achieved by doing this?

Are you thinking that by having multiple blockchains, we're going to have multiple blocks being solved on multiple chains and therefore be able to have more transactions? If so, your reasoning is flawed... as how do you deal with "intergroup" transactions? You're going to be doubling up transactions... ie. PersonA-Group1 -> PersonB-Group2... you somehow need a transaction on Group1 blockchain AND on Group2 blockchain. What happens if a Group does rouge? What happens to all the users in that group? Can users choose which group they want to be in?

How do you deal with HD wallets and address creation to ensure that GroupX wallet private keys/addresses are only used on GroupX blockchain? Are you going to only use miner issued private keys/addresses?

Additionally, you make it impossible for users to actually see for themselves the data from the other blockchains... they have no way of knowing if something is accurate with trusting their particular group of miners etc.


Honestly, this sounds like a terrible idea. All the downsides of centralisation, with none of the benefits of Bitcoin

[NotATether]

Lets look at things at a different angle. I'm not trying to disagree with you I'm just showing where your proposal is flawed.

1. Placing maybe every 10,000 to 20,000 miners into individual groups (each group acts like a bank) with their own address so that groups could communicate with each other (similar to two banks transferring money between each other).

You say that each 10,000 or so miners will function as a bank, but what exactly does a bank do?

A bank moves transactions in and out of your account[1] and charges you a fixed fee each time you make a transaction. Whereas a wallet user has the ability to charge a variable fee to miners.

- Bitcoin users cannot control which miner receives the fee, however a bank account holder can only choose one entity to take the fee (the bank). Also if someone makes a money transaction to somebody else in a bank, both banks charge a transaction fee. Do you see the issue? We don't want to have two different miners levying a tax on each transaction.

Also banks process transactions by order of date, and then by smallest amount, while bitcoin miners take highest-fee transactions first regardless of when they were made.

2. A limited number of wallets say 10 million customers are assigned to each group, each time all groups are full of customers then a new group is created (with a new address).

Customers cannot be represented as a bunch of accounts on a bank record, because there is no link between people and the wallets they create.

We cannot assign wallets to different miners because the miners themselves will not be able to detect all the addresses in your wallet without giving them something called the master public key (a big privacy violation equivalent to Facebook selling your data without your consent).

3. Each group has their own block chain different from another group's block chain (yet still secure because say 20,000 miners all have the same block chain) so that when a transfer occurs either within the same group or between only two groups then each block chain would only need to retain the wallets (and transfers) of its own customers.

It is not the existence of only a single chain that slows transactions down, it is the variable time to find a block. So proposals to make the network faster have to adjust (or work around) the time it takes to find a block.

The bolded part poses a special problem. If only transactions within a single blockchain are recorded on it, then where will inter-chain transactions be recorded, and how will they be verified? (In the bitcoin network, all transactions are verified when they are downloaded by a wallet to make sure they are correct) Presumably, different wallets won't know there are other chains that exist.

[1]: https://www3.mtb.com/personal-banking/checking/understanding-checking-account-transactions

[aliashraf]

OP,

What you are describing from scratch is known as "sharding" a hot topic in cryptocurrency. You know, bitcoin is based on Proof of Work, sharding for PoW networks is somewhat harder to implement, but unlike what some people have argued here, it is very easy to partition Bitcoin in as many "shards" as you wish, and in any topological form of interest, and it can be done even without any essential upgrade to Bitcoin!

Give me a majority of hash power in bitcoin, and I'll do it almost instantly without a glitch, no hard forks or soft forks, no disputes with user wallets who insist on following the classical one god, one chain Bitcoin.

Don't listen to these guys, naysayers, too much. They got points, I know, but still, don't get distracted.  
Most people do not appreciate their own talent and are inherently attracted to boxes, they like being trapped in boxes. You, keep thinking out of the box.