TrustCoin - follow blocks in blockchain that are approved by known signatures

[timeofmind]

There is a lot of discussion lately about how to prevent Bitcoin from being taken over by big players and how to prevent it from becoming centralized. What if each miner kept a list of signatures of people who they know are real and trust, and drop any blocks that are not signed by known signatures, then rather than following the block-chain with most "proof-of-work", the client follows the block chain of blocks that have been signed (ie. validated) by the most signatures that match signatures in the previous block? ie. the blocks that have been approved by the most amount of "known" miners. Ie. we come up with a scheme in which each block can be given a stamp of approval by "miners" (except it wouldn't really be mining anymore, just a form of consensus). In this way, a community of trusted miners could be built up, through their communication with one another on chat/forums, in-person, etc. and only block that has been approved by the majority of miners who approved the previous block will make it onto the chain.

[timeofmind]

All other parameters that are important for block creation can be achieved by consensus among "miners" (ie. miners take turns creating blocks, and each one can mint themselves a certain number of coins based on an agreed algorithm, and frequency of new blocks can be agreed upon by network, etc)

[timeofmind]

The "stamps of approval" could be just a list of random strings that are signed by various trusted "miners" at the bottom of the block.

[timeofmind]

Instead of proof-of-work, we use proof-of-origin. Basically, all miners have to prove that they got into the system through the approved channels (ie. by meeting and gaining approval of other miners)

[timeofmind]

In this way, the system could be ensured to be more distributed and democratic, because there would be an equal amount of miners, all with equal power in determining the legitimacy of blocks.

[timeofmind]

We could incorporate into this system, every algorithm for block approval that all miners agree on to maintain integrity and performance of the system. System performance benefits all.

[timeofmind]

Compare this to how these two methods are both used to valid origin of spam...  PGP vs. Hashcash. Hashcash was invented so that you could receive email from an unkown sender (though you could use trusted signature databases to accept mail from a sender you don't personally know)... this is why bitcoin adopted hashcash (ie. proof-of-work), so that any anonymous person could participate. But is this really the best way to go about it? This means that any entity with substantial resources could eventually take over the system. With signatures, that would never be a problem and you also have the added benefit of keeping "mining" perfectly distributed between all members.

[timeofmind]

As with bitcoin, early adopters of this currency would get the benefit of being the first to start accumulating the newly minted coins.... but it would also be perfectly distributed based on the length of time you've been mining for, rather than based on the computing power you have to put towards the task of randomly searching data. So the distribution of wealth would be more fair imo.

[timeofmind]

I guess the only barrier I see to this working is how new "miners" come to gain the trust of the entire network of "miners", but I'm sure a model for trust adoption can be worked out somehow...

[timeofmind]

Ultimately, you end up with a trusted, decentralized group of entities, all with equal power in validating blocks, on which the entire payment network depends for maintaining the blockchain. It seems to me that is what we'd all like to achieve with bitcoin. Also, it is not like this group of "miners" (or whatever you want to call them), would have to reveal their identity completely to each other. The only real important factor is that they prove that they are individuals actually. Beyond that, it is through the power of the majority that the reliability of the blockchain is maintained. Right now, the problem we have is that control of the blockchain could end up in the hands of few individuals. If we can come up with a way to distribute control between many individuals, then we have that problem solved.

* In fact, that, I think was the whole point of HashCash/proof-of-work (which bitcoin uses), was an attempt to verify that the origin of the email was coming from an individual, and in the case of bitcoin, to ensure control of blockchain is spread evenly between individuals. Surely we can do a better job by building up a trust network of signatures.

[HuuHachu]

Bitcoin and all subsequent alts were built on the idea that you in fact can't trust individuals ...

A trusted miner can be hacked or turn evil without notice.
Someone could create many miners by slowly making them go through your "acceptance channel" and then make them all act synchronously to disrupt / cheat the system.

Proof-of-work is a huge waste of energy, but has the advantage to not require trust in any particular individual.
(you kind of trust the network as a whole though)

[timeofmind]

Sorry, each block would only have the signature of the "miner" who added it. The system would work by following the chain with the most consecutive trusted blocks.

[timeofmind]

Bitcoin and all subsequent alts were built on the idea that you in fact can't trust individuals ...

A trusted miner can be hacked or turn evil without notice.
Someone could create many miners by slowly making them go through your "acceptance channel" and then make them all act synchronously to disrupt / cheat the system.

Proof-of-work is a huge waste of energy, but has the advantage to not require trust in any particular individual.
(you kind of trust the network as a whole though)

Hey. I think you misunderstood. This system does not rely on the trust of individuals. It is a method of ensuring mining power is distributed most evenly among the most individuals; thereby ensuring that no one individual needs to be trusted. You seem to have the concept all backwards.

[Fuserleer]

Heh, I chuckled when I saw this, as I just posted about our currency that takes a similar approach to this.

From what we have tested so far with our algos, is that this kinda collaborative approach to peer verification is pretty solid indeed!

[timeofmind]

Bitcoin and all subsequent alts were built on the idea that you in fact can't trust individuals ...

A trusted miner can be hacked or turn evil without notice.
Someone could create many miners by slowly making them go through your "acceptance channel" and then make them all act synchronously to disrupt / cheat the system.

Proof-of-work is a huge waste of energy, but has the advantage to not require trust in any particular individual.
(you kind of trust the network as a whole though)

The problem with proof-of-work, is that the "proof-of-work" can be potentially owned by few entities. This model is to ensure that doesn't happen. Yes, someone could *try* to build up a network of many accepted signatures... my idea is to conceive of a system for adding signatures that would make that extremely difficult to accomplish. Right now, the centralization of proof-of-work is a major concern of many.

[timeofmind]

Heh, I chuckled when I saw this, as I just posted about our currency that takes a similar approach to this.

From what we have tested so far with our algos, is that this kinda collaborative approach to peer verification is pretty solid indeed!

Pleas give me a link to your post! I am very interested!

[HuuHachu]

you still have to prevent people from creating / controlling many signatures, this seems hard to me.
maybe feasible, maybe not ... especially as long as you want to stay fully decentralized.

[bg002h]

How do you validate new users/miners? Consensus is what occurs after you decide on validation parameters...what stops Corporation X from hiring 25,000 stooges?

[timeofmind]

Of course, another benefit of this system is that nobody needs to waste CPU power looking for random strings of data...

[timeofmind]

How do you validate new users/miners? Consensus is what occurs after you decide on validation parameters...what stops Corporation X from hiring 25,000 stooges?

Ya. I started this thread at the same time the idea came to me, so I have not yet had much time to think about how to overcome such obstacles. I'm sure there must be a way..