Jeremy Côté

Exciting Science

When I was a budding graduate student looking for a supervisor and a research direction to pursue, I had little idea of the scientific landscape. I had a bit of research experience as an undergraduate, but not much. I wasn’t sure what to do next.

I began by looking at the websites of several researchers, trying to see if anything excited me. I spoke to some of them about how I was thinking about joining their group in the future.

When it came to talking about projects, they almost always said the same thing: “Take a look at my recent publications.”

This sounds reasonable. Looking at a scientist’s papers should give you a good idea of what research topics the person cares about. But this is a bad idea, particularly for someone with little experience in the field and wanting more information.

Why am I so against this? Let’s start with an analogy. I love running. It’s the sport I do every day, and running fills me with joy. Now imagine you asked me to send you some resources on running. First, I could send you a playlist of running videos showcasing the beauty of trail running, the huge workouts by elite marathoners, and even the all-out mud fest which is cross-country running. Or second, I could send you my training log, complete with details about repetitions and how I was feeling that day.

Which one would you prefer?

The second option absolutely gives you information. But it’s specialized to the point of being useless. Unless you’re already familiar with running, the training log won’t tell you much. Worse, it’s so cryptic and besides the point of running that it makes the whole activity seem terrible.

I think there’s a similar situation with science. If I want to convince you that my area of research is exciting, showing you the equations and theory isn’t going to cut it. I should show you these at some point, but not at the beginning. I need to build up your knowledge first, and doing so means getting you to pay attention for long enough that you want to learn.

Trying to learn and get excited about a new topic by reading research papers is the equivalent of staring at a training log to get motivated about running. The log is technical, specific, and probably not going to tell you much.

If you want to get a grasp of a researcher’s work, look at videos they’ve made, blog posts they’ve written, and exposition on the topic. Those won’t make you an expert, but they will give you the context to dive deeper.

When I was trying to find a new research area, I skimmed through the papers of researchers. I learned almost nothing. Or rather, I did learn about a bias I held: I expected scientific papers to entertain and excite me about the science1. In an ideal world, that would be the case for every paper. In practice, papers are often records of work, and less pleasant reading experiences.

Don’t get me wrong: the scientific paper is an important part of science (though it shouldn’t be the only thing). But when we’re asked the question, “What do you do?”, answering with, “Look at my publications.” isn’t going to cut it. Instead, we should have different artifacts we can point to2.

If you talk to a scientist about their work, chances are you will be excited. Their eyes light up, they can tell you about the big-picture aspects of the work, and they provide the context to make the science exciting. Often, that gets lost for a person reading about an unfamiliar area of research. Particularly when you’re a student.

I think the answer here is straightforward: Don’t read people’s papers to get excited about science. Instead, search for their artifacts such as videos and blog posts that explain the ideas, providing a gentle way into the topic. I predict this will help you make much more informed decisions about what you want to do.

Yes, the day-to-day work does involve the technical details, and we as students will have to deal with them on a regular basis. But it’s worth seeing the reason for all of this work, and that’s where the scientific paper falls short.


  1. Of course, there are some scientific papers which are great to read. Just as there are some which are unreadable. My point is more that, for the busy scientist wanting to move on from a project, there can be an incentive to get the paper readable enough so that people can understand, but not make it a joy to read. 

  2. I have several places I look to for information about research topics. These include: Quanta Magazine, the Google AI Blog, the Amazon Science blog, Symmetry Magazine, as well as blogs from individual researchers. The point is to look for researchers talking about their work.