ICO Preview: Taraxa [Work In Progress May 21st, 2018]


What is Taraxa

Unfortunately this isn’t going to be a full review. I don’t have a sneak peek at their GitHub repo and their full whitepaper. So for now I am going to report what “I think I know” about this project. Much of it is borrowed from my understanding of the Nano project from which Taraxa is based. I also list out my outstanding questions down below so you can ask them yourself or simply follow my updates. Until final, the title will carry “Work-in-Progress Update [Date]” .

Screen Shot 2018-05-19 at 8.58.10 PM

Taraxa aims to “democratize Internet-of-Things (IoT)” data. On their website the project will achieve this with infinitely-scalable concurrent contracts, trustless light nodes, and encrypted data markets on top of block lattice. These words are directly from their site. The technology will be based on Nano (formerly known as RaiBlocks) and its lattice structure. If you are unfamiliar with Nano let’s do a recap since Taraxa’s foundation starts there. But first, let’s look at the team.

The Team

Borrowed from their website. At first glance, very strong backgrounds and tons of doctorates. Great to see.

  • Steven Pu, Founder: IoT and healthcare entrepreneur, strategy consultant; LinkSens, EviPath, Monitor Deloitte; Stanford University BS & MS in EE
  • Vikram Saraph, Concurrency: Distributed computing, software concurrency, Ethereum hacker; Facebook; Brown University PhD candidate in CS
  • Dr. Ilja Honkonen, Ledger: Distributed computing, space weather modeling, Ethereum & IOTA hacker; NASA, Finnish Meteorological Institute; University of Helsinki PhD in Physics.
  • Professor Paul Gazzillo, Virtual Machine: Programming languages, security, systems; Google, Bloomberg, University of Central Florida; New York University PhD in CS.
  • Willem Wyndham, Virtual Machine, Storage, Reputation: Programming languages, compilers, security, WebAssembly and Bitcoin hacker; IBM; University of Maryland PhD candidate in CS.
  • Dr. Justin Snapp, Hardware: Analog IC design, microfabricated vacuum electronics, IoT enthusiast; Intersil, Qualcomm; Stanford University BS & MS & PhD in EE.
  • Chris Dai, Business Development and Partnerships: Entrepreneur and blockchain investor; Leland Capital, Longhash Japan, blockchain incubator; Stanford University BS in MS&E.

Incorporated in the Cayman Islands with designs to incorporate in Switzerland, Japan, and other countries as necessary.


Token Metrics

  • What stage is Taraxa in now: Currently in private sale but currently oversubscribed
  • No pools: Will not be working with pools (no surprise there!)
  • Symbol: Not announced yet
  • Token standard: TBD. Considering ERC-20 before swapping into their native mainnet.
  • Total supply: There will eventually be 2⁵³ -1 tokens, or 9,007,199,254,740,991 tokens. Seems ridiculous but considering the expected number of IoT devices over the next decade plus the bandwidth and hardware constraints an incredibly high number of pre-mined tokens may make sense.

A Refresher on Nano

A network offering instant fee-less and miner-less transactions. Instead of a blockchain it uses a lattice. Compared to Bitcoin, Nano’s transactions are not stored in the mempool (queue) waiting for a miner to pick up the transaction and add it to a block. Nano differs in that each address/account has its own blockchain. The lattice structure comes from the interconnected blockchains.

When an account creates a transaction that record is then stored on a block in that account’s blockchain. Sending funds from one account to another would record a send transaction on a new block in the sender’s blockchain. Likewise, the receiver would have to create a receiver transaction on a new block in their blockchain. The criss-crossing lines of send and receive transactions between chains create a latticed mesh pattern. This structure differs from Bitcoin in that Nano’s transactions are asynchronous and handled between the two parties transacting. Transactions on Bitcoin must wait for the entire network to validate the transaction. I mentioned above that Nano is miner-less but that is not exactly true. There is a simpler (than Bitcoin) Proof-of-Work consensus algorithm that the network miners use to validate blocks. Solving this puzzle is much faster than in Bitcoin and gives the illusion of instant transactions.

Nano prevents double-spend attacks with its Delegated Proof-of-Stake (DPOS) system its “representative system”. In this system a representative is a large token holder with the responsibility to represent a tie breaker between blocks with a double spent transaction.

So if Nano is around, why Taraxa?

Smart Contracts on Taraxa

Taraxa claims their smart contract achieves high throughput because of two primary reasons: 1) single nodes are able to process hundreds of smart contract calls simultaneously, and 2) the ability to break up smart contract processing tasks to reduce wasted work. The second point is vital in comparison to Proof-of-Work protocols especially. With Bitcoin and Ethereum, all miners are working to produce blocks but ultimately one wins and confirms the newest block. All the other miners (n-1) generated work that was redundant and wasted. With Taraxa, once a miner runs a smart contract that node is the sole block producer and thus reduces waste and makes for high speed output.


Smart contract parallel calls

Virtual machine (VM)

Differentiating itself from Ethereum, Taraxa chooses to leverage existing, battle tested technologies. Creating new languages such as Solidity leave the network vulnerable to a wide range of unknown attacks at the time of the technology’s development. Taraxa’s VM stack is based on WebAssembly (WASM) to enable executing code nearly as fast as running native machine code. WASM is supported by a wide host of mature languages incuding C and C++. Other projects I have covered such as Ontology have used Wagon for WASM since their platform is written in Go. WASM is built on mature technologies such as asm.js and supported by blue chip firms such as Google and Microsoft.

Enabling light nodes

Taraxa will run on light nodes such as Smart Phones and IoT devices. Each node will own a few chains for its transactions (account balance and history) and the smart contracts they have deployed. To remain part of the network independently it will need to generate its own blocks (send or receive transactions as mentioned earlier). Taraxa’s goal is to maintain trustless nodes with independence while minimizing the computation and storage burden. Rather than nodes syncing the entire blockchain they will be able to quickly participate in the network by latching onto the latest snapshot of the Representative’s voting history (a vote counting mechanism). This snapshot is similar to Arweave’s plan (which I detail here) to continuously monitor the network to allow for quick node bootstrapping.

Developer Support

Nano set the bar fairly high for Taraxa to borrow and innovate from. On Nano’s developer tools website there is a wide range of libraries for developers to build with. There are RPC libraries written in PHP, JavaScript, Python and Elixir to make it easier for developers to get started.

Ongoing business developements

Taraxa is currently in serious conversations to become the primary public ledger for CarBlock.io. Additionally, they are progressing through conversations with agriculture businesses in Western China to create a public ledger to capture farming equipment sensor data. Please note I use the word conversation since the protocol is still in active development. Once the team launches their testnet they an start rolling up their sleeves for the two conversations mentioned.

No verdict, still a lot of questions

By 2025 Statista forecasts 75 billion IoT enabled devices will be installed. If cryptocurrencies were to be part of the IoT landscape it would need to account for the device’s hardware capability and bandwidth capacity. The total supply of Taraxa might be outrageous but so is Nano’s. The fact there will be ample enough tokens for each device is a nod to the likely trend of IoT devices proliferating throughout the world. Nano used a 128 bit integer size to represent transactions. They did so to reduce the space needed to store transactions on each IoT device node. Over the coming weeks I will update this “Preview” or eventually turn it into a “Review”. Below are questions I have for the team. As soon as I get answers I will update with you all.

Open Questions

  1. Does Taraxa have a built-in mechanism for data transfer on layer 1?
  2. Or will it rely on a layer 2 solution?
  3. What is the expected ledger size if Taraxa’s transaction volume rivals IOTA and/or Nano? Will the global ledger capture all transactions or simply the total balance in each of the block lattice’s blockchains to compress ledger size?
  4. Will Taraxa also have a DPOS system like Nano? If so, will the Representative system be at risk of failure as a point of failure either through duress or collusion?


  1. Taraxa FAQ
  2. An explanation on Block Lattice on Nano
  3. Nano developer tools
  4. 75 billion deices by 2025



Thank you for coming to the site. Quantalysus publishes blockchain research and analysis for the crypto community. Please follow on Twitter, Steem (please follow and upvote if you can – thanks!)Telegram channel (New!), and Medium to stay up to date.

If you want to earn Aelf (ELF) tokens for just using Twitter and Reddit, sign up for their candy / bounty program.

If you learned something:

Other posts:

3 thoughts on “ICO Preview: Taraxa [Work In Progress May 21st, 2018]”

Leave a Reply