12-07, 13:30–14:00 (UTC), General Track
The Julia programming language has proven to be a solution to the two-language problem, especially in the area of scientific computing. However, being both easy and fast is just the "syntactic" feature and benefit. An extension or superset of Julia can unleash its "semantic" potential to provide value to every company going through digital transformation. We will discuss in more details with examples in the context of quantitative trading and hedge fund. We will also mention Julia's potential in combination with technology such as blockchain. We will release a new package as the first step towards an extension or superset of Julia for building decentralized systems.
The feature of being both easy and fast is crucial for Julia's success in the numerical and scientific computing areas. But an extension or superset of Julia could become the first decentralized programming language, to provide great value to every company, just like division of labor did for the economy.
Julia's Multiple Dispatch decouples attributes with methods in Object-Oriented Programming, such that struct and function can be written by anyone, anywhere, and at any time, but can run together all independently in any application under any company.
Julia has features that if extended can enable decentralized control and ownership. Total transparency can be achieved, unlike traditional open-source projects, because the cost of change is minimum in Julia-based systems.
Julia also has the feature of accuracy, because it was designed with mathematics in mind, unlike some other languages purely for computer science projects, e.g. with the purpose of information sharing, rather than function compositions in mathematics. This feature is key to ensure the correctness of composition in a decentralized-controlled system.
Julia has been used in the quantitative trading and hedge fund industry, especially for high-frequency trading, which are traditionally very centralized and highly secretive. However, we see trends from top quantitative hedge fund and sovereign wealth fund using tournaments to crowdsource their research. Similar type of activities have been adopted long time ago by companies such as P&G and Novartis in various non-financial industries. Using Julia-based systems for decentralization is a practically-viable way for companies to enhance their R&D capabilities digitally in near future.
Julia potentially offers the lowest cost for any company to explore innovative ways to expand its reach digitally (through software system). It also enables collaboration far beyond the data-sharing seen in the crowdsource projects. Together with the blockchain technology, and some extension or superset of Julia, we see a great opportunity of a decentralized system that is correct, scalable, efficient, consistent, decentralized-control, and with better economics. At the end of the talk, we will announce a new package as the first step towards an extension or superset of Julia for building decentralized systems.
No previous knowledge expected
Martin did PhD in Computer Science (inter-discipline with Economics) at Oxford University, and studied in Singapore and MIT for his undergraduate and master degrees. Martin published paper on parallel and distributed computing, supercomputing, Grid computing etc, won best paper award, and served as reviewer and session chairs of internationl conferences in related area.
After PhD, Martin worked as a quant in quantitative hedge fund in London for about 3 year, and started a FinTech startup in China focusing on quant and blockchain-related technology/trading as well (including high-frequency trading). Martin moved to Dubai UAE at the begining of Covid pandemic, and co-founded a technology-based general trading company focusing on live-streaming. Martin is a passionate and true believer of decentralized collaboration and partnership using technology.