Blockchain week in Shanghai

Vincent Serpoul
InnovateForward
Published in
6 min readSep 30, 2016

The last week was eventful for blockchain with the Blockchain Summit held in Shanghai.

The first three days were dedicated to Ethereum devs and the following three days to blockchain for enterprise. You can find a detailed description of the conference in English here, and in French here and there.

ETHEREUM DEVCON2

It was a difficult week for Ethereum, as it started with a critical bug on the Golang client and was followed by a DoS attack in the middle of the week.
Nonetheless, there was much enthusiasm of the community throughout the three days: bugs were fixed, crowd cheered.

We can split the flow of announcements into five categories: security, privacy/identity, scalability/performance, connection to the real world and ecosystem.

Security improvements

Security was one of the main topics of the conference. There was at least one talk each day on the subject of improving, testing and thinking about new ways to deal with security.

The DAO issue made everyone realise that when a system involves financial value, it needs to be perfect. Unfortunately, Ethereum and solidity are still far from perfect.

the infamous DAO issue

Smart contracts need to be treated as critical systems that are not easily updatable. It means that you need to analyze, proof test and secure your code beforehand. A few researchers have therefore started to think about ways to use formal verification for the Ethereum contracts. Essentially, formal verification is a mathematical proof that your code is doing only what it is expected to do. A few solutions to check your contracts already exist, such as oyente (open source) or Imandra contracts.

Even once the formal verification is completed, “escape hatches” are needed to prepare for the worst case scenario: how to stop the contract, how to move out all of the money, etc.

Privacy, identity

identity on the blockchain

Regardless whether for convenience or security purposes, privacy and identity have to be at the center of the implementation of blockchain. Several companies are trying to address the identification problem on the public Ethereum chain and one of them eventually won on demo day: Uport.

Uport allows users to identify themselves on the chain and — to put it simply — acts as a proxy for their multiple private keys. The primary benefit of this identification function is that users own and control their data, preventing it from falling into the hands of third-parties.

The public Ethereum blockchain allows everybody to see all transactions on the chain. Clearly, we do not want that to apply for certain sensitive information such as financial transactions and health. Zcash, a company established by a group of crypto researchers, created a way to use public blockchains without displaying all transaction details, using a zero-knowledge proof algorithm (zk-SNARKS). Zcash will enable “private” transactions on Ethereum.

Scalability and performance (cost, throughput, latency)

Ethereum needs to demonstrate great improvements if it wants to become the “world computer”.

2016 was the year of the Proof of Concepts (PoC), 2017 will have to be the year of production and scaling. Vitalik and the Ethereum foundation are working hard on it:

  • Scalability with Casper through sharding and replacement of Proof of Work with Proof of Stake.
  • Performance through clients (geth, parity), EVM optimisation and off-chain transactions (raiden is one implementation). Off-chain transactions could dramatically increase the number of transactions on Ethereum, without compromising its security and integrity properties.

Connection to the real world

The first topic of the conference was about regulations for a reason. Developers must not forget that regulations already exist and will be enforced on Initial Coin Offering (ICOs), transactions and many types of smart contracts.

Legal issues aside, some initiatives are starting to link real world data on the blockchain.

Oracles or also Cryptlets (as Microsoft named it) allow people to bring external data on the blockchain in a secured way. Such data can later be used to trigger smart contracts. These oracles will enable “proofs”: proof of location, proof of address, proof of identity, etc. which will be available for third parties to leverage on.

Digix, stabl, paxos are bringing world commodities, including gold and other currencies, to the Ethereum blockchain. The tokenisation of these commodities will enable greater trading flexibility.

Currently, the Ethereum blockchain is not user-friendly. Transferring money, accessing services and other basic functions are tedious to use, an issue some companies have started tackling. For example, Metamask helps users to simplify the process of account creation by letting them create and access an Ethereum wallet directly within their regular browsers. Another good example is the “Ethereum name service” which will allow users to access contracts and services, as simply as they do today with URL addresses. The ecosystem is growing strong, little by little…

Ecosystem

The main strength of Ethereum is its attractiveness for developers. The development environment is very similar to what web developers already know. They can use their favorite frontend framework or library (angular, react, meteor…) to do ÐAPPs (distributed apps) in a very short time.

A huge number of PoCs, demos and new services showed developers are already comfortable with the development environment, mixing different technologies such as mobile, javascript frontend, augmented reality, virtual reality, etc.

IPFS heterogenous networking

Services and protocols related to the ecosystem are also growing. One good example is IPFS (InterPlanetary File System). IPFS showed significant improvements as well as new functions such as p2p communication, static hosting, dynamic hosting, etc.

“Web 3.0” was mentioned many times, maybe it is eventually coming!

Blockchain summit

The second part of the week was about enterprise applications of the blockchain.

Microsoft is embracing blockchain technology, especially Ethereum. The company already provides infrastructure to spin up a blockchain environment to develop on (blockapps) and plans to expand its Ethereum related offerings (cryptlets…).

Hyperledger is another well-supported project, providing an open source fabric to spin up a blockchain. A number of big companies are participating in the initiative. It could potentially be a huge success if the support of all its members continues.

A wide range of different blockchain implementations surfaced during the talks such as bloq or dfinity. Those different implementations can be split into three big categories: private blockchains, public blockchains and consortium blockchains (cf Vitalik Buterin blog post on the subject).

Taking into account regulations, laws and compliance to which financial companies have to comply, it’s understandable why they prefer private blockchains. These private blockchains are also easier to scale and control, but compromise transparency, decentralisation and security.

Some other companies are pushing for public blockchains, emphasizing transparency, security, availability and decentralisation. Moreover, some cryptographic technologies (e.g. zk-SNARKS ) allow privacy over public blockchain.

The last category is consortium implementations, which a lot of companies seem to go for. A few nodes have authority on the chain and consensus is reached among selected actors (settlement among banks is a good use case).

Overall, the obvious takeaway from the last three days of the conference is that there will be more than one blockchain, to fulfill different needs.

--

--

Vincent Serpoul
InnovateForward

innovation fellow @LOFT - the Lab of Forward Thinking