A meeting at Exeter Uni to discuss the future of the vegetation component/s of JULES. Interesting to see all the developments being made, and to discuss possible advancements.
This was three days of workshop with some of the most interesting and informed people about Nitrogen – I learned a huge amount! Hosted by Beni Stocker and Colin Prentice, there was just a few global modellers – myself, Andy Wiltshire, Beni and Sonke Zaehle. The benefit for me, of course, was that so many people who did measurement work, could really talk about the nitty gritty of the real world, which is so difficult to get into our models in a meaningful way.
Beni has written up a summary of the meeting for New Phytologist, which gives a flavour of the breadth and depth of our discussions, but isn’t as fun as the science discussions we had over dinner. 🙂
Billed as a comprehensive introduction to planning and managing a research project as well as commercial projects, this seemed like a useful course in light of the project plans that I have been/am putting together. It covered a few days worth of course in a short day, which is exactly how I like courses – very little messing around or padding, a lot of information.
My key take homes: have a goal and a plan; time, quality and cost, which one can you sacrifice?; use three point estimates to calculate time; watch the critical path; you need a WBS (work breakdown structure) and a network diagram; identify risks and disagreements early; all projects fail in planning, not execution.
Peer reviewing effectively and efficiently. The biggest take-home for me was that beyond 3 hours spent on a review, a review doesn’t improve. 2 hours is about the right amount of time.
The other nice bit of advice was to look for the message and claim in the abstract, then look for it again in the introduction, evidence for it in the results, and see if it is in the conclusion. If it is missing from any of those, probably there are problems.
The key things that I took from this are:
- cntl R in shell provides a search of recent-ish commands. Worth it just for this.
- Use documentation (triple double brackets in python) to make help files for future you and others. Provide examples. Say what the input and output is. Write documentation before you start – can help with sorting ideas and when you get interrupted. Everyone else doesn’t comment. Be the change. 🙂
- Revision control. It really is important.
- Testing. Without testing, how do you know that you’re getting what you think you are?
- Automate things. Save time and mistakes.
A tutorial on using the met office unified model for climate and weather prediction.
A whole range of tips on giving effective interviews infront of the camera, including what to wear, where to look and how to get across the message that you want told.
A vox training course, looking at how to be a good presenter.
My key take home: think about how you want the audience to feel, not just what you want to say. e.g. When you’re re-capping, you want to re-assure them. So use a re-assuring tone of voice. That’s totally different to the voice and gestures you’ll use when you want to inspire the audience.
Also: fall in love with the full stop. Silence increases the importance of what you’re saying. It gives gravitas.
The take home messages:
- Make sure that everyone gets a career boost from the project. Especially you. You need to make sure that you are developing your career, taking on new responsibilities and tasks. It’s necessary to have great new papers with each project, but you do need something more.
- Time planning is worth about x10 the time spent faffing and messing up later without sufficient planning.
- Learn from others. Someone has already done something similar to your research project. They can tell you about potential pitfalls.