BIP001, Part 2. Charles Hoskinson and Vlad Zamfir on The DAO, Hard Fork of Ethereum, and Governance

Last week the third Blockchain Incredible Party (#BIP001) conference has been thrown in Odessa. Forklog continues to cover the event, now the story will go on The DAO, Ethereum, and governance issues of open-source projects.

In the first part of the report we talked about the approach of blockchain development for govtech industry and building the Financial Web:

Blockchain Incredible Party, Part 1. Govtech and Financial Infrastructure

On a more specific note, the report touches The DAO hack and the emergence of the governance issues of the existing blockchain-based projects. In his speech, Charles Hoskinson from IOHK tried to address those issues by introducing the concept of Blockchain 3.0, while Vlad Zamfir shared his views on the situation around The DAO suggesting an alternative approach to the community’s hard fork debate.

hoskinson

Below are the key findings from Hoskinson’s speech about the decision-making within the crypto projects’ communities as well as his view on the next generation of blockchain systems.

Redefining the rules. The existing governance problem, as Hoskinson derives, is that when important decisions regarding the changes to the protocol are needed to be done, there are no explicit rules on how it has to be done to reach a community-wide consensus. Currently, the situation is that the developers team, being the “experts”, has to take responsibility for the changes (despite miners making the final blow) to lead the community to some sort of consensus and eventually to decide. However, other problems might arise in this case:

“What happens when the core developers have a conflict of interest? What happens when they’re making decisions not because it’s for the best of the ecosystem, but because they have invested in something and they want to bail it out. Or what happens when they change the philosophy of the system like adding the KYC layer. It can dramatically increase the value in cash [of the tokens] and they can sell it at a higher price. Is that what’s good for the system …or that’s just something for themselves?”

The governance problems. The issues of current blockchain systems’ governance model has become clearly visible in the light of bitcoin scalability discussion and Ethereum’s forking discussion after The DAO hack:

“The fact that the DAO crisis happened has exposed the governance problem with the Ethereum Foundation. There’re some people saying ‘we should fork’ and other people saying ‘we shouldn’t fork’. Well, that’s a decision. Something has to be done. … It’s not explicitly defined in the protocol.”

The whole point of the discussion about the DAO, is not about hard forks or soft forks, nice presentations and technical details, explained Hoskinson:

Who should vote? There are many questions to be raised for almost every cryptocurrency, but the most important is which part of the community makes decisions. According to Hoskinson, the stakeholders are the users of cryptocurrency, so that’s who might vote to make decisions over system’s future.

However, it’s not generally the case, especially when we talk about bitcoin. On the contrary, the politics of decision-making in cryptoworld seems to be overcomplicated: the miners control the processing, the core team has control over the code changes, exchanges have control over the infrastructure, but still only miners can vote. Anyways, none of them has to hold their funds in cryptocurrency to have power over the future of the whole community.

Whether or not the community should be governed by the majority of bitcoin holders or some kind of democracy is rather a dangerous question, so we’d better omit it here. However, as Hoskinson reminded, the discussion within the community is needed and it might eventually lead to a birth of a new kind of blockchain systems:

“The third generation of the blockchain systems is all about saying: ‘let’s explicitly build into the blockchain the mechanism to actually change the blockchain’.”

Ironically, the whole point of these ideas is that in order to solve the governance problems the communities will have to create systems with built-in mechanisms of decentralized autonomous organizations (DAOs, thousands of them!).

zamfir

The next speaker, Vlad Zamfir, was talking specifically about the situation around The DAO hack, hard fork and soft fork misconception, and what challenges it poses to the community.

The attack on the DAO. In order for the audience to understand the whole point of discussion, Vlad quickly reminded about what happened to The DAO:

“The DAO was like a smart contract. A bunch of people pulled their money into that smart contract and that gave them rights to vote on how that money would be spent. … It turns out that this contract is vulnerable. So basically an attacker drained the funds of the DAO.”

The attacker did so via creating a child DAO, using a recursive call vulnerability within the split proposal mechanism, and moving ethers from the mother DAO to a child DAO for many times during the single split. Thus about $50 million was stolen very fast.

“The price of ether went from all-time high of $21 to $11.50 and then $10 in like two days. So there was a huge sell-off due to the hack of this DAO, which was like 40% of the ethers. That lead to a controversy on who to blame? Is it the DAOs code or was it a lack of documentations of Solidity?..”, detailed Zamfir the immediate reaction of public.

Proposed solution. The Ethereum developers (one might even presume the wide majority had stakes in The DAO) lead by Vitalik Buterin himself, responded very quickly on the attack and proposed the solution – to vote for a soft fork, effectively blocking the funds of both mother and child DAOs, and then to vote for a hard fork in order to move funds from the Attacker’s account to accounts of holders of The DAO tokens in a simple refund contract. So that the hard fork (HF) and soft fork (SF) political discussion about what the right thing is to do has started and it continues to the day.

Forklog has thoroughly covered the timeline of the attack, further counter-attack, and some possible consequences of the DAO crisis.

Who is responsible: The DAO only? Despite the polarization within the community, there’s one fact, mentioned by Vlad, to be highlighted:

“70% of contracts that call to untrusted contracts have reentry vulnerabilities”, he noted.

This might mean that the problem is at least not in The DAO only, but that the entire ecosystem was just not ready for this kind of attacks.

The question of whether or not the community should support the hard fork depends on whether or not the community feels responsible for the ecosystem’s functioning, thinks Vlad Zamfir:

“Basically, the discussion that comes up: is Ethereum ‘a CPU’ – general, not caring about what’s running on it, being just kind of mechanical thing; or is it more like ‘my CPU’? When something bad happens on my CPU, I will go and press the reset button or maybe I will change some software that’s running in order to stop that bad thing to happen on my CPU. This is an important difference”, he outlined.

According to the speaker, the consequences of the DAO fail on Ethereum further development and ‘broader ecosystem like fundraising, smart contracts, and relationships with SEC, FBI’ can be harsh. However, the hard fork is something unprecedented before in the Ethereum ecosystem, so some sort of discussion is definitely needed.

Editor’s note: What might be wrong with HF?

First of all, historically the cryptocommunity is generally opposed to any kind of interventions of some central authority into economics. Actually, one of the key milestones of the bitcoin rising popularity was the financial crisis of 2007-2011 when the US government started to bail out banks lead by “too big to fail principle”.

Many people do not want the same thing in crypto, because this might impose the wrong incentives of taking too much individual risks if you’re big enough thus breaking the competition and leading to centralization or, even worse, some kind of socialism.

Another major problem is that Ethereum’s contracts were thought to be immutable, and this is bought as one of the major system’s features. If the community will change the code to punish the attacker (based on the ‘common knowledge’ that the attacker is a malicious agent), one might say the hard fork lays a dangerous precedent, especially for future government interventions. And, well, this might be not the way the railroad should be run.

Some proponents of the HF mostly stress the fact that Ethereum is still a young and rather an unstable project, so interventions of the community are still needed. Others say Ethereum is not a product of crypto-anarchism, as it’s more about the applications of a smart contract mechanism than that of politics.

Why hard fork is the solution (according to Vlad Zamfir ). Speaking further on common SF and HF misconceptions, Zamfir reminded the audience about the recently discovered DDoS-attack vulnerability of the proposed Ethereum soft fork. However, the most important thought was that a hard fork might be better in terms of system’s decentralization than a soft fork in general. While soft forks generally don’t need the involvement of the community, except for the miners – hard forks need everyone to upgrade, so they grab much more attention:

“The hard forks require more consent than soft forks. Because soft forks essentially circumvent the community’s will and go to miners and get them to host change, rather than requiring all the community to [vote]. They’re requiring less consent, they are instituted by a more concentrated group of people and therefore more centralized.”

“Another common misconception is that hard forks will make a platform less trustless. I think this is a misconception because when they cannot trust the state, people want to place trust in software. But I think this is kind of like intentions and expectations, messaging everyone who has invested and [who is] involved in the community economics.”

According to Zamfir, people who invested in The DAO took it as a great experiment and explained why he thinks the community should bail out the project:

“If their money isn’t returned then we might not be able to reinstate these types of systems for a long time. Not just because the trust into software was lost, but the trust in the agents involved in the system also was lost. So there’re two kinds of things: the trust in software and also trust in miners and broader community”, he stated.

“The reason why I think this is so significant is that this is the first real test of the Ethereum community’s ability to govern itself and the first real test of Ethereum community’s willingness to take responsibility for what happens on the Ethereum platform. I think the kinds of damages we see here potentially pale in comparison to the kind of damages we might see in the future due to what happens to decentralized platforms. It’s my feeling that it is the responsibility of the community to take responsibility to make sure that serious damages don’t occur.”

“I can hope that most feel pro hard fork, because I hope that in the future when damages will be much worse we’ll also be able to deal with that”, concluded Zamfir.

We at Forklog are following the course of events around The DAO while not taking sides. Let’s the community decide!

by Eugene Muratov