Can you call blockchain trust-less or more like decentralized trust?


#1

One of the main features that attracted me to blockchain technology was the trust-less system. You don’t need to trust a third party, like a bank, to do your transactions between peers. I really dislike banks, their extra fees, long wait queues, mountains of paperwork etc. I am sure you feel that way. That is why to me blockchain was amazing, you get a transferring system where the validation of your transactions are already embedded within it. No need of third parties.

However, recently I came across to this article https://hackernoon.com/a-christmas-ethereum-mystery-4bbbe910d22b. And let me to ask this question. Is blockchain trustless or is more of a decentralized trust?

I am starting to think that instead of a trust-less system. You get a decentralized trust system. You still need to trust the miners they will move your transactions, and you still need to trust the chain is the correct one and not a malicious one (51% attack). And this is just for PoW system. What if you get other less robust ones like PoS or dPoS where the decentralized trust falls to fewer validators?

Finally, this is more an open question, and I would like to see what do you guys think. It is your choice, do you think we still need to trust others?


#2

More like decentralized trust, in the sense that we can trust some protocol because we have reason to believe it to be incentive-compatible (that by acting truthfully in one’s own interest, one can achieve the best
outcome to oneself).

With something like Bitcoin, based on our belief in cryptographic protocols and the lack of adversaries that can crack them (e.g. Shor’s algorithm against our public key cryptography), and our belief in the ability of the blockchain to continue being secure given some level of decentralization (this is a very fuzzy notion), and our trust in the Sybil control mechanism, we can avoid the need to trust any one party in particular, and just
trust the protocol.

Somewhat tangential discussion:


#3

This is such a fantastic question. I actually wrote a blog post about this a while back (here) and this was exactly what we talked about at our meetup last night (recording).

Overall, any technology (blockchain or otherwise) can only partially take away some risk of trust in other institutions or technology. There is no such thing as a fully trustless institution or technology. This is even true in technologies like encryption, where we might think it’s absolutely failure-proof but it’s not. This is because we can never 100% account for all possible behaviors of all participants. For example, if you send an encrypted email to a friend, even though you don’t have to trust third parties not to eavesdrop on your message, you still need to trust your friend not to forward that message to third parties.

In short, trust is never eliminated in blockchains. You’re instead distributing the trust to developers, miners, internet providers, exchanges, etc. Hence, it can be minimized and distributed, but never eliminated.


#4

Yes. This was a misconception of mine too. That blockchains allow for 0 trust. But what they really do is minimize trust. Everything you do requires some form of trust. With blockchain-based software or money, instead of relying entirely on a 3rd party (100% trust to 1 person/entity), you’re distributing trust among thousands of people.

Not trust-less, but less trust.


#5

In the ACH-model, the number of nodes is very low so the required trust-per-node is quite high. On a blockchain like Bitcoin or Ethereum, the number of nodes is very high and the trust-per-node is much lower. So really what you’re trusting is the consensus mechanism. It’s this reason that my mind is always blown at how little interest most blockchain engineers/entrepreneurs put into consensus mechanisms and the underlying infrastructure of their tech stack. That’s what you’re trusting.


#6

100% I don’t know why there isn’t a bigger discussion around consensus mechanisms. I feel like even though we’ll have 1,000 new crypto projects in the next 10 years, they’ll be using variations of only a few consensus mechanisms and there should be more research done there.

Professor Emin Sirer gives some clarity on consensus mechanisms here.

This post by Kyle Samani does a good job explaining the layers of the web 3 stack.
https://multicoin.capital/2018/07/10/the-web3-stack/


#7

There is a LOT of discussion around Consensus mechanisms. They just happen in private Telegram channels :wink:

One of my goals is to fix this and make this knowledge more public. soon :wink:


#8

“Trust less” is nearly impossible to pull off. With incredibly high probability the consensus algorithm will maintain the most accurate ledger of transactions. There are quite a few ways you could create a scenario where the blockchain is functioning as intended but the ledger is able to be manipulated such as a big majority of the hash power is located with one entity. A big liability blockchains aim to solve is the counterparty risk. I believe blockchains use decentralized trust because it is a robust way to remove bias. Decentralized trust is less of an absolute and more a sliding scale. At the worst case with one entity controlling consensus you have too much power consolidated in one location and no enough checks on this power. At the best case there is no large clustering of hash power no matter how you filter the nodes into groups. (ie. geography, political affiliation, social status etc).


#9

I would say that Blockchains try to minimize trust as much as possible. Most older systems have guardians and custodians who can wipe history if they feel like it. Many blockchains are designed to be immutable, uncensorable, publicly auditable and accessible. This is a huge leap forward from what we have had in the past.

I think the Byzantine General’s problem outlines the space pretty well.

Byzantine Generals’ Problem, an agreement problem in which a group of generals, each commanding a portion of the Byzantine army, encircle a city. These generals wish to formulate a plan for attacking the city. In its simplest form, the generals must decide only whether to attack or retreat. Some generals may prefer to attack, while others prefer to retreat. The important thing is that every general agree on a common decision, for a halfhearted attack by a few generals would become a rout, and would be worse than either a coordinated attack or a coordinated retreat.

The problem is complicated by the presence of treacherous generals who may not only cast a vote for a suboptimal strategy, they may do so selectively. For instance, if nine generals are voting, four of whom support attacking while four others are in favor of retreat, the ninth general may send a vote of retreat to those generals in favor of retreat, and a vote of attack to the rest. Those who received a retreat vote from the ninth general will retreat, while the rest will attack (which may not go well for the attackers). The problem is complicated further by the generals being physically separated and having to send their votes via messengers who may fail to deliver votes or may forge false votes.

Byzantine fault tolerance can be achieved if the loyal (non-faulty) generals have a majority agreement on their strategy.

Consensus mechanisms are a good example of various ways people have tried to solve this. IMO there is a good amount of work being done in this space since Bitcoin came out.