Achievement unlocked: mainnet release
--
The mainnet release, code-named Auriga, is a major milestone in a long journey which we started a couple of years back. It’s the first time that our network, which has been online one way or another for well over a year, is online for real. The tokens are not test tokens, they exist and can be used to get work done on the network. Using FET as the fuel, autonomous agents, the smart ledger, synergetic and AI-powered smart contracts are now a reality. Now, as they say, you can BUIDL.
It’s an incredibly exciting point to have reached, as those building applications with Fetch.ai are now able to work on a stable platform, one where we won’t rearrange the goalposts, furniture and fridge contents just because we feel like it. Now, there is governance (and this is something we have a lot more to talk about during the start of 2020), a proper, clear process for enhancements and standards, alongside a network that we commit to making great efforts to keep online each and every day. Of course, it is version 1, and it represents a truly vast software effort. Printed out, the source code for the ledger alone, without any of the libraries, would make a stack of printer paper over 50 centimetres high (that’s 19 inches, or a shade over 2 cats on top of each other¹). There will be bugs, there will be issues, there will be things that behave non-optimally, but we’ve hit a point now where the core architecture is good to go. Our foundations, upon which we’re all going to build, are firm enough to take the next phase of work and innovations, whilst being used for real, day-to-day decentralized applications.
Version 1 is our seedling, and during 2020 it’ll be growing up with some pretty incredible things that we have stashed up our sleeves. This’ll all be part of the 2020 vision (see what I did there?) and we’ll be publishing our roadmap for this year and beyond early next year as part of our ongoing commitment to transparency. Before then, let’s look at what our version 1 foundation release includes and some of the key events that we’re currently looking at.
What’s in our seedling?
The key ingredients that allow applications to be built, deployed and maintained on Fetch.ai are present and working in version 1:
- You can build and deploy both synergetic and AI-powered smart contracts. These smart contracts combine on and off-chain computation, enable solutions to large-scale optimization problems (route-finding, for example) and deliver actual on-chain neural networks. This is very cool stuff, and unique to Fetch.ai. You can read all about it in our documentation.
- You can build and deploy autonomous economic agents, and we have a very cool search mechanism that enables active delivery and searching of the network through the first versions of the digital environment capabilities we’ve talked about so often.
- As you would expect, all the key functionality of the ledger is present, and the first fully working version of our proof-of-stake consensus protocol is there, running and doing its magic. We have a block explorer, a web-based wallet, a network status system and much, much more.
- APIs, documentation, tools and guides necessary to develop and deploy applications both for beginners and advanced users alike. This includes, for example, our transpiler to help Ethereum developers and our online Etch smart contract playground.
As followers of our progress will know, we’ve focussed on aiming to deliver a stable, working network that consists of reliable first versions of each of these components rather than trying to bang out performance for performance’s sake. This means that if you download the code yourself, you can deploy your own version of the network with many shards and run the performance up to much higher levels. As the consensus mechanism matures these local deployments, and the mainnet itself, will reach 30,000 transactions per second. But for this first main network release, we’ve chosen to be sensible and — at least for the first month or two — keep the plumbing simple, so we’re running 40 nodes and the performance will be slower, but safer. We’ll put our foot down on the throttle when it is safe to do so for all of the network’s users, and we’ll do this in conjunction with further decentralisation and the efforts of our community.
Sounds great, so what’s happening in 2020?
2020 is the year when we switch focus from laying down the foundations, to a new phase of refinements, performance increases, integration of new innovations, and higher level work that provides deeper, richer application possibilities. It’s also when we purposefully work towards decentralisation: integrating our novel minimal agency proof-of-stake with unpermissioned delegation system fully into the mainnet will be a major delivery in the first half of the year, for example, and it’ll gradually merge the validator staking auction mechanism you may have seen into the main, native network itself.
A larger, more decentralised operation of the network is just part of this process. We will be publishing exciting details of our future plans for network governance. We’re building a technology that ultimately will be owned, maintained and developed by all of the network’s users, and we need a governance framework in place that ensures everyone who utilises our technology gets a say in how it matures in the future.
Usability is a big part of the first half of 2020, and there’s a lot happening. From documentation to tools, we’re going to be working hard to improve everything that we already have, as well as delivering a stack of new things. One of the things we’ll be doing early in 2020 is “closing the loop” on our ERC20 to native FET conversion tool, so you can convert back to ERC20 from native. We’ll also be delivering hardware wallet support and web-based extensions to allow such things to be integrated into web sites. It means that soon, our native network will have the same things you rely on day-to-day on Ethereum, such as the important functionality that tools such as MetaMask, MyEtherWallet, Infura and Etherscan deliver. And as you’d expect, all of our tools will be entirely open source, so others can build on them for their projects and applications.
Then there’s the application side of things. As we switch our focus towards enhancing network performance and adding capabilities to the initial systems, there will be a lot of activity to deliver applications in the multi-agent and smart contract space. There will be developments regarding privacy preserving computing, and indeed, a significant amount of work in privacy and security generally. We will also integrate other aspects of learning and intelligence to ensure more effective intelligent connections, and better exposure of trust and reputation to all users of the network.
We are incredibly proud of what we have delivered, and we can’t wait to have the main network out there, for everyone to use. We have a ton of amazing things to talk about in the coming months, from our governance and network expansion plans, through new partners and applications, to a stack of innovations we’ve yet to reveal to the world. It’s a pleasure and a privilege to have our community and partners with us as we work to make it happen!
-
[1] Lines of code, even if measured in cats, is, as Bill Gates once said, like measuring aircraft construction progress by weight. A better indication is development activity, and Santiment’s measurement of this shows Fetch.ai to be number 5 in the entire world for ERC20 projects during November 2019, and if you include the first ten days of December, we are number 2.
