Charles Hoskinson Surprise AMA 12-1-2020
(written by @Eric_Czuleger IOHK)
On January 12th 2020, Charles Hoskinson sat down for a surprise AMA. There he discussed the preparations being made for the Haskell Shelley Testnet. He also talked about the methodology of building on two different codebases and looked forward to increased development velocity in 2020. Furthermore, he touched on expansion initiatives by the Cardano Foundation and Emurgo.
SUMMARY
- The Incentivized Testnet team is once again fully operational and picking up development cadence to solve difficulties on the testnet.
- The incentivized testnet served the purpose of understanding user experience and testing parameter expectations.
- As we prepare for Goguen, we are focusing on identity, oracles, multi-asset standards, and establishing rules for data and metadata.
- We have encouraged Emurgo and the Cardano Foundation to release information on developing dapps for the Cardano ecosystem, as well as expanding geographically in Africa and Japan.
- The Byron reboot, the OBFT hardfork and the results of our security audit are all upcoming within the near future.
Developing the ITN
We launched the Incentivized Testnet in the middle of December. Because of this we ran into challenges due to the fact that many of our key team members went on vacation for the holidays. This put us in a situation where known bugs and difficulties could not be addressed immediately. Last week the team returned and we are now performing at full strength.
This means that we should be able to issue a patch for network stability next week. We will then continue to release updates at the cadence users have come to expect from the Rust team. During the ITN we have been able to collect an enormous amount of data about the use and misuse of the protocol. This includes information a good deal of information about saturation metrics, if they’re necessary and how they should be adjusted. We also examined the possibility of stakepools operating on multiple chains at the same time. We have anticipated all of these eventualities and the ITN is invaluable for assessing our previous assumptions.
Goals of the Testnet
The goal of the incentivized testnet was twofold. The first was to get an understanding of the user experience as an operator of a stakepool and a delegator. We wanted to allow users to create pools, build relationships with one another, and to demonstrate that there is a group within the platform that is ready and capable of running its infrastructure. The software must continue to evolve in order to make it a perfect user experience. However, UX is always a moving target. Given that this is the first time that we have released the ITN to the general public, we have already generated a good deal of actionable information on how user experience is best adjusted.
Secondly, we were curious at the outset of the ITN about how we designed our parameters. We wanted to know if the protocol would achieve decentralization, stability, and performance in a live network. We have the capability to install measures proactively and retroactively. This allows us to adjust network conditions, activities and behaviors which we discover to be counterproductive. For example, we have a pledging mechanism. We can install punitive measures in case the mechanism is abused by a malicious actor. We now know that it is possible for us to take such measures.
Gathering crucial data
At the moment, we are collecting data about why people are doing the things that they are doing, and how they are doing these things. We are finally making decisions about how the protocol will evolve prior to mainnet. This is another crucial point of the evolution of the ITN other than the user experience and the business of delegation.
The ability to tune the ITN live in conversation with the users has been crucial to our development. The point of the testnet has been for things to break, change, and update. New additions can be tested even if they will not go into the mainstream. The point of launching an incentivized testnet, was to encourage users, and the company to work together to find the most successful path forward. With all that said, it seems more than likely that the code deployed in production will be the Haskell code.
Looking towards Haskell
The Haskell code has a different temper, design and quality level than the Rust codebase. This is because it was built from formal specifications. We did not launch with the Haskell code because it takes longer to bring to market. The current ITN allows us to test user experience, parameterization, and the business that will be required for the protocol to get where it needs to go. When we launch the Haskell code this will give users high assurance that everything will work properly.
The Rust and Haskell testnets are different, and they were created for different purposes. Many of the bugs, and connection issues that users see in the ITN will not be there with the Haskell code. Again, this is because it was constructed with formal methods, rigor and different specifications. However, the Haskell testnet will benefit from the large community of stake pool operators which has already developed. Furthermore, it has the benefit of the understanding that we have created around parameterization and if we need to put punitive measures in or not.
We understand that there are currently some issues with the ITN but we believe that this is how good engineering is done. It is about dividing and conquering identified problems. By breaking things down into simpler difficulties and working closely with the community we have been able to come up with good mitigations. As we now return to our normal development cadence current issues will begin to be resolved. We are optimistic that the month of January will resolve most, if not all of the issues people are having with the ITN. This will also give us the ability to finalize some experiments we have performed in the run up to the mainnet.
Agenda 2020
We recently looked forward to the 2020 agenda with our chief scientist Aggelos Kiayias. The agenda includes finishing the Ouroboros research started in 2015. We are also looking forward to our collaboration with Peter Shwabe to produce a post-quantum signature scheme for Cardano. This will allow us to improve our network stack design, include usage of recursive snarks, and of course, harden ourselves for the rise of quantum computing. Our 2020 budget also includes a donation to the University of Wyoming for our upcoming partnership.
Our development philosophy is also continuing to express in every product of our portfolio. This includes Daedalus, the wallet backend, Cardano node, and Jormungandr. We are currently looking at weekly or biweekly releases for this software. We are not only looking at bugs and issues, but feature requests. We are now far more commercial oriented and far more disciplined. Because of this we are optimistic that we will satisfy the requirements of Shelley and Goguen within 2020. We also believe that we will implement Voltaire and part, if not all of Basho within this year. This of course, depends on our ambition in scalability.
The Hydra protocol should be ready for its first paper sometime in February. This is a very good protocol which is accelerating to becoming much better. At a conservative estimate, Ouroboros is 5 times faster than Ethereum at its peak efficiency. This is without optimization. Already Ouroboros has proven itself to be an enormously valuable stream or research.
Furthermore, we have figured out a good deal about the theory of proof-of-stake. This allows us to account for problems like spikes of dishonest majorities. The current iteration of Ouroboros does not account for this. We are confident that with our research and development velocity we will be able to account for upcoming challenges.
Preparing for Goguen
In terms of Goguen, we believe that for Cardano to be a successful platform for smart contracts we need four fundamental properties. They are as follows:
First, we must have a grid-spanning, multi asset standard. That means that we need to be able to generate assets which are fungible and non-fungible as well as regulated and non-regulated. This would generate competitive assets to the Ethereum ecosystem. This will also attract individuals and businesses which want to develop on Cardano into the platform.
Second, we must be able to bring external information into the network. These are commonly called oracles or data feeds. We’ve seen the success of Chainlink and other successful companies in the decentralized finance space and we believe that we need to operate in a similar fashion. In essence, Cardano must be able to interact with the outside world. Thus we are pursuing an oracle standard with Emurgo and the Cardano Foundation.
Third, we consider identity to be a first-class citizen in the development of Cardano. Bitcoin started out with only an asset which had no understanding of identity. Since then, we have seen the rise of the DID standard and the self-sovereign identity movement. As a consequence we are well positioned to make identity a first-class citizen within our ledger. We want to make identity for multi-sig, pubsub, aliasing, and embedding contracts.
Currently, we have a project called Prism within the Atala team. They are building a ledger agnostic identity system based on the DID standard and Hyperledger Indy. This will be ported to Cardano within this year. This will help us improve the user experience as well as helping the development of decentralized applications by outside parties. This is one of our highest priorities for our developers.
Finally, there must be a good story about metadata and data embedded within the transaction. It is becoming an increasingly common requirement to maintain regulatory credibility in the developed world. In order to maintain compliance with the law, we must develop capabilities within the ledger and transactions for users to decide what to do with the transactions they broadcast.
Commercializing the ecosystem
Of course, there are other important concepts to work on this year as well. These include development tooling, defi infrastructure, and interoperability with Ethereum to name a few. We are currently working with the Cardano Foundation to find the minimum viable investment necessary to get a bootstrap as quickly as possible for this type of infrastructure to be built out.
Plutus and Marlowe are transitioning from being in the laboratory and emerging as commercial projects. Now, we are having aggressive conversations about how we build great things that change the world. In the coming months as we close out Shelley and move into Goguen, you will see a great deal more investment in hackathons, tooling, and building pedagogical tools.
Working with our partners
We hope that we will see Emurgo release a formal investment thesis for those who wish to build in the Cardano ecosystem. This will also include a financial commitment of how much Emurgo intends on investing in each portfolio company. It’s important that we let the market know what we are looking for, what we are willing to commit, and which geographies we are looking to target.
We have also begun speaking with the Cardano Foundation about two geographies. They are specifically Africa and Japan. Currently, we are building out our network to gain exposure within every country in Africa. This will lay the groundwork for millions of people in Africa to engage with Cardano as a platform. The Cardano Foundation is also working on developing a dedicated strategy within Japan. There has been mutual support between Japan and Cardano since the very beginning and we would like to see that expand into meetup groups, pilots, projects, and dapps coming out of many locations in Japan. We have asked the CF to draft a strategy around this along with working with their partners at Emurgo and IOHK.
We are moving very quickly in the near future. We will soon have the Byron reboot shipping with the Haskell wallet backend integrated. This will hit either in January or February. We will also have the OBFT hardfork. These are the last updates required prior to the Shelley mainnet upgrade for the Haskell code. It just remains to be seen when the Haskell shelley testnet will launch. Cardano is currently undergoing a security audit from out partner Root9B. They will finish their security audit soon and remediations will follow. We will publish the results of this report so that the public has an understanding of the security audit.
We believe that 2020 is the year of Cardano. There is a lot of hard work ahead of us and we’re looking forward to building a platform that changes the world.