LUNA ‘Hardfork’ Review By The Guy Who Forked Ethereum
A couple of words about myself
You can find a bit of info about me on this article on cointelegraph. Long story short — I witnessed The DAO hack and I was among those who drove the creation of Ethereum CLassic. After that I ‘forked-off’ from ETC and founded Callisto Network (launched via the airdrop).
Since I have unique experience in “forking chains” I would like to share my own experience that could be potentially helpful for the LUNA.
LUNA crisis in a nutshell
Terra is a blockchain platform that enables its users to create stablecoins. Terra has its own native crypto currency traded at $LUNA ticker. Terra dev team launched their own (stable)coin $UST on their platform.
Typically, stablecoins like USDT are “stable” because users give US dollars to Tether company in order to receive USDT tokens. At any time USDT tokens could be converted back at 1:1 ration to US dollars by the same Tether company. That’s why they are stable.
$UST was pegged 1:1 to dollar by a complex algorithm that should have minted / burned LUNA and UST in order to keep the price valid.
And this algorithm failed which resulted in huge amounts of LUNA being printed to the market which resulted in its price dropping dramatically.
To fork or not to fork
Terra CEO proposed to “fork” the blockchain and scrap UST altogether.
- Track the status of the proposal live
- Article describing the status of the process
- Original announcement on twitter
- Technical details of the proposal
Notes regarding the “forking plan”
Technically the proposed solution is not a “fork” (read the explanation of the “fork” term here). Chain Split or Hardfork is a term that is used for the chain upgrade process where one chain with its history of transactions is splitted into two chains (with different consensus rules since then) at certain block. Before that block the chains have the same history of transactions. Starting from the “fork block” the two chains have different histories of transactions.
Terra CEO proposes to relaunch the chain from scratch (erasing the history of transactions) so it is not a fork of the existing chain.
We already have precedents for a similar situation. TheDAO hardfork. Ethereum Foundation proposed an emergency hardfork to redeem TheDAO funds that were withdrawn by the “hacker”. They gave community extremely short time frame to engage a pseudo-voting process without a proper description and discussion around the topic. Alternative solutions were not discussed widely and the consensus was not built in the community.
As the result, a large group of the community (including myself) disagreed with the actions of Ethereum Foundation and supported the original chain refusing the hardfork. The key point here is that in the case of the Ethereum Foundation, this was a time-limited decision. DAO funds were time-locked and the “hacker” could not use them. Therefore, the decision was rushed as Ethereum Foundation should have developed a solution before the “hacker” could dump all the funds onto the open market.
Mistakes that must be learned:
- The lack of consensus within the community & dev team is what caused the split in the project — not the proposed solution itself. It is important to investigate alternative solutions and handle a proper PUBLIC discussion of different options.
- A clear process of PROPOSING, INVESTIGATING and EVALUATING of the solutions must be defined and used.
- It is important to adhere to the policy of transparency when making such critical decisions. The processes must be observable and well-documented. Financial operations must be commented and documented as well.
- The consequence of the hardfork (and the chain split) was the split of value between the two “projects” and a so-called ETH-ETC race. At this time, it was possible to clearly observe the inverse correlation of prices of $ETC and $ETH at the market and at the peak point ETC was worth 0.4 ETH. Trading community was expecting “ETH-ETC flippening” which didn’t happen though. The same happened to the hashrate wars between ETH and ETC. It is clear enough that the split in a project governance does not just create the value out of thin air — instead the “value” of one initial project is split between it’s two parts.
Although it can be concluded that ETH emerged as the winner of this drama. Indeed, a project supported by the original team will win in the long run as long as the team proves competent (and the competitor suffers from their internal problems).
However this approach is justified by the time limiting factor that pushed Ethereum Foundation to make a rushed decision. In case of LUNA there is no such time limit and there is a good reason to stick to a more thoughtful solution development process taking into account the above described mistakes.
It is extremely important to note that alternative proposals are presented within the community, most notably this one: https://agora.terra.money/t/say-no-to-the-fork-notothefork/19518
There is another difference. The problem of TheDAO was not the hack. The problem of TheDAO was that it had too much money and it was a significant part of ETH supply locked in the hands of one person with no incentives to be a bagholder. So the solution was to withdraw his stake. In fact, TheDAO bailout was not solving the problem of TheDAO — it was solving the problem of Ethereum ecosystem and the hardfork WAS the solution to this problem.
In case of LUNA the relaunch / rebranding of the project is NOT a solution of the problem. LUNA is a system of stablecoins. The root of the problem was the algorithm that could lead to such a collapse and the loss of USD-peg of the core stable coin. The removal of the stablecoin does not solve the problem of this stablecoin platform however. Instead it simply removes its core part.
- The algorithm must be re-investigated and the stablecoin must be reworked. There are time-proven examples of working stablecoins (such as USDT). Possibly it is worth to stick to a more traditional approach.
- There were funds allocated to “insure” the collapse like this. Take it and use it on purpose.
- There is absolutely no need to split the value of the network between it’s parts (if there will be $LUNA and $LUNC as per Do Kwon’s proposal) — this will only lead to further decrease of the overall value of any of the system parts.
Request to LUNA Team
First of all, I would like to maintain financial transparency. There was a BTC fund that needs to be used to deal with exactly this type of situations. So where did the money go?
I would recommend the team to provide a proof of the funds being spent on the UST compensation. I would encourage the team to comment each significant transfer publicly.
The transfer of 25012 BTC on May 9, 2022, with a value of 842 million USD
Then a serie of transfer on May 9, 2022, 70752 BTC and valued at 2,380 billion USD.
Second, I would propose to delay the emergency hardfork and take the time to investigate the proposed solutions as well as review the root of the problem — the stablecoin algorithm that didn’t work as intended. A postmortem report must be published and the lessons learned must be highlighted to the whole Crypto Community as the accident can be a good experience for further projects.
Callisto Team is investigating the issue
Callisto Network is a security organization dedicated to improving the safety of the blockchain industry.
We are currently analysing the accident and we will publish our report to wrap up (1) what happened, (2) what was “officially said” happened, (3) what was done to solve the problem and (4) what could have been done instead.