The full blog post can be found here, on the ICCS blog.

• In September, “From C to Comonads to Climate - A Functional Programmer’s Journey in Array Programming” as the keynote at the 1st ACM SIGPLAN Workshop on Functional Programming for Productivity and Performance (FProPer’24), co-located with ICFP 2024 in Milan, Italy.
• In November, “Computing for Climate Science and Programming for the Planet” at the departmental seminar of the Department of Computer Science and Technology at Cambridge, where I spend half my time, and in short form at Queens’ college at the Monday night SCR talks.
• Also in November, “Writing Computational Models that are FAIR and VRAI” (slides on request) at the Centre for Atmospheric Science, Yusuf Hamied Department of Chemistry, University of Cambridge.
• And again in November (busy month!), “Taming the Black Box - Effectively Leveraging ML Sub-Models (some thoughts and case studies)” at the Cambridge Centre for Climate Science (CCfCS) Winter Symposium 2024, University of Cambridge.

Some of the content of these talks is currently in draft stages of a paper I hope to submit in the early new year.

1. Typing Python with mypy

Typing Python with mypy Many compiled languages include a ‘type checker’ as part of their compilation process which applies automated checks to source code to rule out potential runtime errors due to mismatches in the format of data (‘type errors’). The Python language does not include such a check: its types are ‘dynamic’, with type errors occurring only if encountered at runtime. Python however supports type annotations (since Python 3.0) which allows a programmer to insert optional type information into code which external tools can then use to type check a program. This session will teach how to use Python types alongside the mypy tool for ruling out program bugs and better documenting source code. We will also talk about some fundamental concepts in typing and program verification.

1. What can abstract mathematics tell us about programming climate models

What can abstract mathematics tell us about programming climate models?

Category theory is a subfield of mathematics that seeks to expose common underlying structure in other areas of mathematics. It has since also became a foundational technique for understanding logic and programming, with its use both in semantics of formal languages and as a tool for structuring programs. Many concepts in computer programming can be explained from a category theoretic perspective, yielding new insights about how to reason about programs and generalise their definitions. In this session, I will give an overview of a few key ideas that have applications to numerical programming tasks familiar in earth systems modelling. This will provide some fresh perspectives about how to structure and reason about programs both for correctness and efficiency.

1. And a contributed a bit to a training/tutorial session on Introduction to Computational Science in Julia .

I was particularly happy to deliver the second talk here which was a lot of fun, using ideas from category theory to think about how we specify the behaviour of key patterns in numerical modelling and also uses the ideas to motivate optimisations. This was all done in Python, including generating tests from categorical laws. The material is online.

Details can be found here. Closing date 9th September 2024

Here’s a video about the position and potential ideas that a succesful candidate could explore:

Potential topics include:

• Static analysis tooling for numerical models
• Verification of critical libraries used by scientists
• Novel language ideas for future computational models
• Extending existing languages with novel features to support science
• Developing new tools to support climate scientists

Other ideas possible that I’d love to hear. I’m looking for someone who is excited about applying PL ideas into the climate modelling domain. There will be a chance to work alongside scientists to try out ideas ‘in the field’ through my network.

The event was co-located with the 51st ACM SIGPLAN Symposium on Principles of Programming Languages (POPL 2024) which gave an excellent opportunity for this community to consider how its research could be targetted and leveraged to address the multiple overlapping planetary crises facing us. The workshop had two keynotes: from Drew Purves of Google DeepMind on ‘Setting the stage for AI biodiversity’ and Lisa Rennels of UC Berkley on ‘Building Open Source Software for Climate Change Research - Lessons Learned from Mimi.jl’. The contributed talks were grouped into four tracks on ‘Modelling and analysis’, ‘Energy and efficiency’, ‘Software engineering and ecosystems’ and ‘Policy and decision making’. This demonstrated the many angles from which programming languages research could help address our current planetary challenges. See the list of talks here.

The event highlighted many exciting works already coming from the Programming Language community but also many opportunities for further cross-disciplinary work. Two discussion session highlighted the need for thinking more broadly and stepping outside the core discipline to look for opportunities where expertise could be applied or where general assistance was given. Co-organisers Dr Dominic Orchard and Prof Anil Madhavapeddy are now working to synthesise lessons and a way forward for the community to have transformative change in this area.

Dominic noted that “it was particularly exciting to see how engaged the students and early-career scientists were with the issues, and the energy and enthusiasm for leveraging their skills to address the significant challenges facing humanity”.

Alongside this, he helped deliver a one-day training workshop at the African Institute for Mathematics on software engineering good practice and programming in the context of climate science. Particularly, Dominic gave a session on using Using types to rule out bugs - Python vs Fortran perspective, giving an overview of concepts in verification and correctness, and giving practical examples of how to leverage type systems for correctness.

At the conference, Dominic gave a talk on Research Software Engineering for the New Generation of Climate Models giving an overview of the work of the Institute of Computing for Climate Science and a spotlight on some of the interesting challenges going forwards.

Read more on the ICCS blog

• Gilberto Pastorello from the Lawrence Berkeley National Laboratory presented our collaboration on the ONEFlux data processing tool for land-atmopshere cabord flux, with a poster presentation (corresponding abstract).

• At the DataWave side meeting, Jack Atkinson from ICCS presented our work on improving interoperation between Fortran and pythonic machine learning libaries, applied to the Gravity Wave Drag parameterizations in atmospheric models.

• At the DataWave side meeting, I presented a short talk about Analysis and lightweight verification of Fortran code using the CamFort tool and demonstrated some of its lightweight static analysis tooling by running it live on the MiMA moist atmospheric model.