FAQs 常见问题

The most important thing is to enjoy learning. I don't mean this in the poetic 'live laugh love' sort of way (though that's also true) — I mean that to thrive in physics (or, really, anything for that matter) you have to love the process of doing it and not simply the idea of being it. For physics, that means solving problems and being curious about things that do not currently make sense! Don't always simply chase the answer; enjoy sitting with a problem and taking time to solve it. With the internet and AI it's so easy to get an immediate answer. Sometimes that can be great, but when it comes to getting better at problem solving, avoid it! At every level, physics involves trying to overcome problems. That's kind of the whole job. If you learn to love working through a problem that doesn't fully make sense, or that isn't immediately obvious, then you will set yourself up well for research.

In practical terms, take mathematics and physics at the highest level available to you, and if further mathematics is offered, take it. Mathematical fluency is something you will draw on for the rest of your career, and it is much harder to develop later than it is now.

I would also strongly encourage you to learn to code as early as possible. Python is the best language for most of astrophysics, and even a basic familiarity with it will give you an advantage. There are excellent free resources available online. I taught myself over a summer using a book I picked up from Waterstones (Python Programming by Michael Dawson). One important caution, though: resist the temptation to reach for AI coding assistants while you are still learning. They are useful tools once you understand what you are doing, but using them before that foundation is in place will leave you unable to debug or reason independently about your own code. It's plausible that AI will be writing a lot of code in the future — but while humans are still in the loop (i.e. involved in the research process), you will still need to be able to read, modify, validate, and interpret it. And this is only possible if you work through problems yourself! You have been warned!!

Beyond that, read widely. Popular science is a great place to start, but don't limit yourself. Follow your curiosity: attend talks, read papers, watch videos, ask questions. Don't feel you have to 'wait' until you are at a certain level to engage with something that interests you.

First: be curious, and resist specialising too early. Keep taking the courses that really interest you. My philosophy has been, and will always be, to do what interests you, not what is necessarily easy, or strategic for your grades. Wherever you can, do what you love.

Learn to code seriously. Become familiar with Python, version control (Git), and numerical methods. Learn to use LaTeX.

Make the most of the research opportunities available to you during your degree. Most physics departments offer summer bursaries, and some universities run formal placement programmes. This will give you an honest picture of what a research career involves, and will let you build relationships with academics who can support your PhD application.

Finally, the same caveat about AI applies here as it does at school: these tools can have great value once you know what you're doing, but they provide the easy way out. There is no substitute for struggling through a problem. Think of every coding bug or physics problem you have to solve as an extra rep that will strengthen the physics part of your brain.

Almost certainly not, but it may require some deliberate effort.

Ultimately, coming from a physics background will give you a significant advantage. Some PhD candidates come in through adjacent degrees like mathematics, engineering, or computer science and find themselves well-placed to do astrophysics.

Whatever your background, the skills that matter most are mathematics, physics, and programming. There are lots of free resources available online to help you learn these, though it will take time and may be harder than doing a university course.

Get research experience wherever you can. A summer project, even in a related field, demonstrates commitment and gives you the academic contacts you'll need for references.

Video

• 3Blue1Brown — I cannot recommend this channel enough. The videos are brilliant!
• PBS SpaceTime
• Veritasium
• VSauce — really formative in getting me curious about the world and thinking about connections across disciplines

Lecture notes

• David Tong — all of his notes! (Cambridge DAMTP)
• Harvey Reall — general relativity and black holes (Cambridge DAMTP)
• Sean Carroll — general relativity

Books

• The Feynman Lectures on Physics — Richard P. Feynman

Here is a highly non-exhaustive list of books and films I like.

Non-fiction

In terms of physics books, before I started studying physics I'd tend to just pick up whatever sounded cool to me at the time. I ended up reading a lot of non-fiction (not limited to just physics — I really loved biology, chemistry, philosophy, maths, psychology, etc.) but I can't say that the books I read were the best resources available, and I'd be reluctant to recommend them as such. So my advice instead would be to just go for whatever looks interesting to you. Searching around in a bookshop for something that looks exciting is a really fun thing to do.

I think popular science is a great way to get a flavour of the scientific landscape and to get excited about it. However, learning physics is always a more hands-on process than popular science books will allow, so if you are committed to learning physics, you will ultimately need to pick up a textbook and start solving problems!

That being said, here are a few books I have really enjoyed:

• The Fabric of Reality — David Deutsch (1997)
• How to Teach Quantum Physics to Your Dog — Chad Orzel (2009)
• The Moral Landscape — Sam Harris (2010)
• How to Teach Relativity to Your Dog — Chad Orzel (2012)
• Waking Up — Sam Harris (2014)
• Being You — Anil Seth (2021)
• What We Owe the Future — William MacAskill (2022)

Fiction

• The Stranger — Albert Camus (1942)
• The Wall — Marlen Haushofer (1963)
• The Old Man and the Sea — Ernest Hemingway (1952)
• Dune — Frank Herbert (1965) - I'd also recommend his standalone works (Destination: Void, The Godmakers, Direct Descent)
• His Dark Materials trilogy — Philip Pullman (1995)
• Stories of Your Life and Others — Ted Chiang (2002) — some of the best science fiction I have ever read!
• Ball Lightning — Liu Cixin (2005)
• The Three-Body Problem trilogy — Liu Cixin (2008) — my favourite science fiction trilogy!
• The Egg — Andy Weir (2009, short story)
• The Martian — Andy Weir (2011)
• The Expert System's Brother — Adrian Tchaikovsky (2018) - Really anything by Adrian Tchaikovsky!
• Exhalation — Ted Chiang (2019)
• Piranesi — Susanna Clarke (2020)
• Project Hail Mary — Andy Weir (2021)

Films

• It's a Wonderful Life — Frank Capra (1946)
• Back to the Future trilogy — Robert Zemeckis (1985)
• The Truman Show — Peter Weir (1998)
• The Matrix — The Wachowskis (1999)
• The Prestige — Christopher Nolan (2006)
• Inception — Christopher Nolan (2010)
• Interstellar — Christopher Nolan (2014)
• Arrival — Denis Villeneuve (2016)
• Get Out — Jordan Peele (2017)
• Uncut Gems — The Safdie Brothers (2019)
• Tenet — Christopher Nolan (2020)