Image source: Leilah Schubert, UTS
Nathan Langford is a member of the JT-001-14 Quantum Computing committee. He is an ARC Future Fellow and Associate Professor in the Faculty of Science at UTS, where he leads a research group, Circuit Quantum Science UTS, and the Millikelvin Quantum Science lab, a cryogenic facility for studying superconducting quantum processors. He is a member of the UTS Centre for Quantum Software and Information and on the Technical Advisory Committee of the Sydney Quantum Academy.
When and why did you become involved with standards development?
At the beginning of 2022, I joined a relatively new standards working group on quantum computing.
Quantum computing is at a very interesting stage of development as an industry. On one hand, it is rapidly expanding, including in terms of industrial efforts to build and design applications for an industry-scale quantum computer that can tackle problems of commercial interest for economic and societal benefit. On the other hand, it is still a very new, fundamental science, and there are a lot of unknown questions about what quantum computers will ultimately be good for.
Designing applications for quantum computers is a complete paradigm shift from traditional computing methods. First, you have to find applications that quantum computers are good at solving, and then have to find examples where traditional computers are bad at solving them. Because if traditional computers are already good at solving those problems, there’s no way you’re going to want to build a quantum computer to do it. But that raises an even bigger challenge - if we are only interested in problems that traditional computers can’t solve, then how do we design the quantum computers and algorithms to solve them?
How do standards impact and interact with your industry?
Currently, the answer is simply not very much at all. Hardware developments in quantum computing rely indirectly but implicitly on the standards that support all the so-called “classical” (non-quantum) adjoining technologies (like computer control, microwave electronics, optics and photonics, etc) that provide the basic control hardware that runs our quantum computers. There is a natural impetus for quantum software development to happen in a way which is at least compatible with the machinery that drives traditional software development, but there is also a potential need to break away from these existing models in order to capture the fundamentally different character of how quantum computing works.
At a more ad hoc level, the quantum computing community has slowly been “self-organising” some sort of de facto informal standards creation, especially in the context of system benchmarking, but formal standards don’t yet play a very direct role in quantum computing development. In contrast, quantum communication technologies are perhaps slightly further down this industry development pathway, and arguably riper for the influence of standards.
Why is access to standards important?
There are a few reasons I think that the development and availability of standards is important to our broader community and specifically to our quantum computing community.
1. Quantum architectures are rapidly evolving and very heterogeneous: making useful and meaningful comparisons between different architectures and platforms is hard, especially in the context of benchmarking, and especially when they can operate so fundamentally differently. We have already seen examples where the community has, in time and sometimes bumpily, converged to commonly accepted definitions, but we have also seen examples where different approaches and parties still just advocate for an approach that casts their own niche in the best light. Standards can help bridge that gap.
2. Quantum is interdisciplinary: Quantum technologies, especially quantum computing technologies, are predicted to have impact across many, many diverse sectors of society and the economy. People working in quantum come from very diverse backgrounds as a result, and we need that to diversify even further. But communicating between different fields is very hard and requires a lot of work. Good standards can help mediate communication between disparate communities and help to facilitate better and more efficient collaborations.
3. Quantum is deep tech: Arthur C Clarke’s so-called third law is that "any sufficiently advanced technology is indistinguishable from magic”. Quantum tech falls into this category. Because it is based on the physics of extreme scales (very small, very cold, very few), its building blocks are concepts that we do not tend to develop good intuition for naturally. Building up an intuition for quantum physics takes work and study, not to mention a willingness to try to understand the maths. And making advances in quantum tech requires people to operate simultaneously at the forefront of both science and engineering. In this sort of environment, it can be very hard to tell charlatans and snake-oil sellers from legit, cutting-edge players, even for researchers within the field. Standards will be an important tool for trying to solve this problem.
4. Quantum is rapidly developing: Progress is moving so fast that it can be difficult to keep up, even for researchers working in the thick of it. Standards could help smooth that path and avoid some of the worst roadblocks. However, one of the big challenges for developing standards in such an environment is to do enough to be of benefit, but not so much as to impede progress.
5. Structures and principles are so new and different that it is hard to both imagine and plan for what it might be possible to do with quantum technology, especially quantum computing. We are already starting to face problems with how AI has been developing way ahead of discussions about ethics and guiding principles and regulation, which are really struggling to catch up. And this is causing big problems, both in terms of regulation and societal backlash. Quantum tech is potentially in a similar position and having clear standards could help avoid some of these pitfalls.
What is the future of standards development in your area of work?
Hopefully, the future of standardisation in quantum computing will see increasing engagement and broad consensus from across the community. It is hard to start developing standards from scratch in such a radically different technology and there is a bit of a chicken-and-egg problem in that it will probably be hard to get broad engagement if the standards being produced are not high quality, but that is hard to achieve without good engagement. I guess the most obvious places to see standardisation developed initially are in the areas of terminology definitions and performance benchmarking. Other areas that might emerge sooner than later include how to program and control quantum computers and creating updated specifications for noise and performance requirements for enabling technologies (like lasers, optical fibres and microwave electronics).