How I work

This essay serves three purposes. First, it helps me to know myself, which in turn helps me to improve the way I work. Second, it’s a way for other people to know me, and improve the ways in which we interact. And third, it’s aspirational. It sets the standard I want to maintain.

Daily Routines

My work breaks down into three kinds: research, writing, and running the London Institute. I try to work on all three every day.

My daily routine is to get to the office at 6:30 a.m. and work till 6 p.m. I spend the first part of the day—6:30 till 1:00—on research and writing, and from 1:00 onwards running the Institute. The first part I’m mainly on my own, and the second part mainly with others. I don’t take meetings during the first half of the day, so I can rest assured that this time is reserved for creative work. I don’t have lunch, but on most days I interrupt my work with a trip to the gym or Mass. 

When I arrive in the morning I have a black coffee and a Diet Coke and jump straight into research. This involves a mixture of paper and pencil calculations, testing my ideas in Mathematica, and writing up my results in Latex. I find that when I keep a problem at the top of my mind, my subconscious mind works on it behind the scenes. So when I come back to a calculation, more often than not I’m able to make progress where before I was stuck. But this only works if I come back to the same problem every one or two days. So I try to work on just one or two research projects at a time.

My second kind of work is writing. This includes writing for our website and the press and other internal documents. I take the view that writing is thinking: by trying to express my beliefs on paper, I figure out what it is I believe in the first place. (Writing this essay is a case in point.) A distinctive aspect of my writing process is that it’s very iterative: I do a first draft, then a second draft, and so on, sometimes into the dozens. In general I find I can come back to a piece of prose after one or two days and be sure of making it 10% better without too much effort—though it does take time.

At 1:00 I start my third kind of work: running the London Institute. This is a mixture of planned and ad hoc meetings with colleagues; reviewing the work of other members of the Institute; meeting visitors and members of our Council; and corresponding with people outside of the organisation. A big part of what I do is establishing our routines and rituals and protecting our standards—though of course it’s for others to enforce them.

In general I don’t work from home. As with exercise, I find it takes a lot more willpower to work from home than in a dedicated space surrounded by like-minded people. So when I want to work over the weekend, I come into the office. I do this at least one day most weekends.

Simplicity

I have adopted a few key habits that simplify my life and allow me to be more productive. The first is that I don’t eat anything until 6 p.m. Before then, I just drink black coffee, Diet Coke and water. I started this years ago when one of my American relatives urged me to try it for a week. To my great surprise, after five taxing days, I stopped feeling any hunger at all. I’ve stuck with the practice ever since. When I do eat, however, I eat a lot, usually the equivalent of two dinners.

This has brought a number of benefits. One is that I save time—it takes a lot of time to find, cook and eat food, and then clean up. Two, I’m more creative, which I suspect is linked to maintaining longer periods of intense focus. Three, I’m more trim. Importantly, this diet hasn’t affected my energy levels or performance in the gym.

The second simplifying habit is that I don’t watch the news. This includes TV, newspapers and any other news sources—and I don’t use any social media. So I’m largely unaware of current events, such as who won a major election or when the Olympics were on. Most of what I need to know I learn through word of mouth. Looking back after ten years of news abstinence, I find our obsession with the news similar to our obsession with our favourite sports team—that meaningful and that meaningless.

This has brought a lot of upside. One, I’m able to focus more fully on the activities that I love: my research, running the London Institute and spending time with others. Two, I save time, probably 45 minutes a day. Three, I’m more likely to engage in what I call “local good moves”. Instead of talking about, say, US politics, I focus on things where I can make a concrete difference: helping out someone in need or visiting a lonely friend. Fourth, I find that notions such as left-wing and right-wing have faded away, and I rely more on common sense in how I see the world. In other words, I find news to be an implicit form of peer pressure, and avoiding it makes it easier for me to follow my conscience.

The third simplifying habit is that I make a point of living close to work. At all of the places where I’ve worked—Caltech, Cambridge, ENS, the Curie Institute and the London Institute—I’ve lived at most a 20-minute walk away.

A long commute could easily eat up 20% of my work day, not to mention deflating my mood. So I’m willing to compromise to be close: living in a smaller apartment and paying more for it. By the same token, I would be wary of working for an organisation located in an undesirable location. I want to both live near work and live in a part of town that I like.

Tactics

I’ve developed a number of specific tactics to be better at what Cal Newport calls deep work: professional activities performed in a state of distraction-free concentration that push my cognitive capabilities to their limit.

I sleep for 5 ½ to six hours per night. I’ve experimented with different sleeping routines, and found that a lot depends on habit—I can get used to sleeping anywhere between 5 ½ and 8 ½ hours a night. I normally think of myself as a night owl—in the sense that I like to stay up and am the last to leave a party—but I have found with time that I prefer mornings over evenings for creative work.

To keep my focus when there is noise around, I often resort to two types of noise cancellation. First, I have Apple noise-cancelling AirPods. On top of these I wear ear defenders—the ordinary kind seen on the shooting range. If these interventions don’t do the trick, I play white noise in the form of a thunderstorm through my AirPods. I sometimes listen to music while working, but only when solving equations rather than writing. I’m pretty broad in my musical taste, but my preferred music for working tends to be deep house.

I like to read through a piece of prose or a mathematical derivation, printed out in hard copy. That way, I can jot down quick notes while I read it, so as to preserve the flow.

In my early years I would frequently show up late to things, but a friend of mine convinced me that it’s cooler to be on time than fashionably late. At work I like people to show up on time and meetings to start on time. I also think talks should end on time. It’s only in the academic subculture, where there are no consequences for inefficiency, that it’s considered a feature rather than a bug for talks to run over schedule.

I try to avoid being on external committees. For one thing, they cut into my time available for research and running the London Institute. For another, they are almost always boring. Third, the sort of person who thrives on a committee tends to be diplomatic and hold soft views, whereas I am often blunt and hold strong views. The best advice I know for how to avoid committees comes from Richard Feynman: adopt an air of unreliability. That way, instead of having to say no to committees, they say no to you.

A Cambridge don and journalist once taught me a trick he used to hit tight deadlines. He would light a cigar, start working, and know that he had to finish his article by the time he finished the cigar. This would give him one to two hours, depending on how often he puffed. I’ve used this to great effect, especially for tasks that I tend to put off, such as proof-reading a paper or writing a letter of reference.

Space

I think a lot about the space in which I work. I spend a lot of time in my office, and its design affects my motivation and efficiency.

My office is on the second floor of the Royal Institution building and looks onto Albemarle Street. I tend not to use a desk, preferring instead a small portable writing table. For meetings I have a pedestal dining table for up to six people. I have an Edwardian three-by-seven bank of drawers for storing practical things. Above my fireplace there is a blackboard, painted directly onto the plaster wall in blackboard paint.

I keep a lot of books in my office on two large bookshelves. In the past I would keep most of my books at home, with only my science books at work. But now, if I like a book, I get two copies, one for work and one for home. Partly this is because I take inspiration for how to build and run our Institute from many disciplines. As well as physics and maths books, I also have a lot of books on biography, philosophy, design, business, writing, strategy, typography, psychology and advertising.

Inside my office there is a cupboard under an external flight of stairs, not unlike the cupboard that Harry Potter grows up in. I do most of my creative work in this cupboard. I first started working in it to be more focused—working in my main office, I was too accessible and kept getting distracted by others (at the London Institute we make a point of keeping our doors open). Now I’m still accessible but there is an extra layer of navigation required to find me.

I like working in my cupboard, and spend half of my time at work there. It is 6 ½ feet by 4 ½ feet, with the height of the ceiling ranging from three feet to eight. Inside it are a small Georgian writing table and a compact armchair. To my right are three fifteen-drawer cabinets for papers, which I can easily access from my chair. For storing larger things I have a chest of drawers, and on top of the chest is a printer. The cupboard is lit by two brass table lamps.

Sometimes for a change of scene I like to get out of the office and work from a coffee shop. I do this a few times a week at a restaurant across the road. The staff know me and are happy to let me work over a black coffee for 90 minutes or so.

I love to work by a fire. I get to do this when I visit the English countryside or Texas where I grew up. In the old London Institute building, on South Street in Mayfair, I had a wood and coal fire going in my office most of the year.

Communication

A big part of work involves communication, so I’ve tried to capture how I best communicate.

As I mentioned above, I spend the first part of my day on creative work. While I can be reached by people during this time, in general I prefer to meet after 1 p.m. However, I don’t interpret this rule too strictly—sometimes it’s best for others to find me for a quick question or if something important comes up.

I prefer speaking over writing with people I know. It’s a much faster way to understand and resolve things. If you know me well or are in the building and send me a long message, it’s unlikely I will read it—better to find me or ring me and talk it through.

In general I prefer small groups to being on stage. I’ve managed to gain a certain proficiency at giving scientific talks to peers. But in general public speaking is not one of my strengths, and I’m happy for others to have the limelight whilst I forge one-on-one relationships or put my ideas in written form for the public.

I don’t like email. This is partly because I don’t like switching between tasks, and partly because most email is shallow work, whereas I prefer to spend my time doing deep work. So I try to minimise my email activity. One way that I’m able to receive fewer emails is by sending fewer emails. Another is that I try to avoid cc-ing others, on the assumption that fewer people will cc me back. At the London Institute, because everyone shows up full time, we don’t send a lot of messages—and in any case we mainly use Whatsapp instead of email.

A long time ago I realised that email was the bane of my life. So I took the extraordinary step of hiring someone to manage my email for me. Now that I run a research institute, I usually have an assistant who helps with this. My assistant reads my email and responds as me when possible. For emails that need my input, she gives me the gist of the message and then responds as me according to my overall intention. Occasionally I write emails myself.

I try to avoid video calls, preferring instead to speak on the phone or meet in person. My reasoning is that, if you know someone well, video doesn’t add much, and is sometimes even distracting. It also makes it impossible to simultaneously do mindless chores, such as walking to do an errand or tidying my office (most of my tidying is done on the phone). On the other hand, if you don’t know someone well, a video call is a poor approximation to meeting in person. I find it offers little of the intimacy that comes from physical interaction, but misleads us into thinking we’ve properly met.

Tools

I use a number of tools to do my work, from a computer to chalk.

For my computer I use a 13.6 in MacBook Air with the M2 chip. I elevate it with an aluminium laptop stand when I’m at the office or a Nexstand when I’m on the move. I tend not to plug into a big external screen, but I am currently thinking about whether this could make me more productive.

My phone is a black iPhone 16 Pro, and I use AirPods 4 for headphones and noise cancellation. I often read books on my phone, especially when travelling. But if I like a book enough to read it, I buy a physical copy as well.

I do my mathematical work in a black ruled 19 ⨉ 25 cm Moleskin notebook. I use a 0.5 mm mechanical pencil, with 2B lead, which is softer than the standard HB. My current favourite is the Tombow Mono Graph multifunction pen (black and red) and pencil. It has a spiral eraser which I use a lot. I like to print out my text when editing, and use the red ink to make corrections and the black ink to cross out parts I’ve finished. My favourite stapler is the Swingline 43 plier stapler, which is only available secondhand.

At the blackboard I use Hagoromo Fulltouch chalk, widely considered the world’s best. I store it in Wedgwood basalt dishes, with a larger supply kept in an ebony wood tray. My current blackboard is the plaster wall above my fireplace covered by several sanded coats of blackboard paint. However, I am looking into upgrading to the real deal: black honed slate, which is not particularly expensive, though it is heavy.

There are a handful of software packages that I use extensively. I use Mathematica for symbolic computation and data and function visualisation. For general programming I also use Mathematica. I used to use C when I needed more speed, but computers are fast enough for my needs these days, and if I do need more speed, I work out a faster algorithm. I use Adobe Illustrator for making vector graphics for my research papers and Adobe InDesign for typography and document layout. For Latex, I use TexShop and Overleaf. I use Google Docs for writing outside of research, such as for this document and collaborative writing for the press or our website.

I buy most of my stuff on eBay, and have become pretty proficient at it. I have a few dozen saved searches and each day I get an email alert if something shows up that meets my criteria. I use Auction Sniper to automate my bidding, which puts in my bid two seconds before the end of the auction.