On September 10 and 11, five researchers are giving half an hour each to our students. Thirty minutes, a short interview, then the floor opens.
- Thursday, 1:00 PM — Igor Gilitschenski, University of Toronto. Machine learning and computer vision.
- Thursday, 2:00 PM — Art McDonald, SNOLAB. He shared the 2015 Nobel Prize in Physics for the discovery that neutrinos have mass.
- Thursday, 3:00 PM — Jason Edward Lewis, Concordia University. Computational media and Indigenous futures.
- Friday, 11:00 AM — Nikita Dhawan, University of Toronto.
- Friday, 2:00 PM — Marsha Chechik, University of Toronto. Bell University Labs Chair in Software Engineering, and Chair of ACM SIGSOFT.
All times are Toronto time. But the schedule is not really the point of this post. The point is that most students waste this kind of access, and it is worth knowing how before you are sitting in the room.
The question that wastes your half hour
It is some version of how do I get into research.
It is a fair thing to want to know. It is a terrible thing to ask a working researcher, because the honest answer is a shrug. They got in through a specific accident twenty or thirty years ago, in a system that has since changed, and they know it does not generalise. So you get a polite non-answer about being curious and working hard, and thirty minutes evaporates.
The same goes for anything you could have found on their faculty page. Where they did their PhD, what their lab is called, how many papers they have. You are spending live access on a search query.
Ask about the decisions, not the career
A researcher's career is not transferable. Their decisions are, and nobody writes those down.
Nothing in a published paper tells you why that problem, out of the hundred they could have picked. Nothing tells you what they tried for eight months that did not work, or how they knew when to stop. That is the material you cannot get anywhere else, and it is the material that is actually useful to a sixteen-year-old choosing what to spend a year on.
- How did you choose this problem instead of the one next to it?
- What did you believe when you started that turned out to be wrong?
- What does a normal Tuesday look like?
- When something stops working for months, how do you decide whether to keep going or drop it?
- What is the last thing you read that changed how you work?
Every one of those is answerable, specific, and something the person has genuinely thought about. None of them can be looked up.
Do fifteen minutes of homework
Find one thing the person has actually worked on and read enough to say a sentence about it. Not the whole paper. The abstract and the figures will do.
This changes the conversation more than anything else you can do. "I read that you work on X — how did you end up there?" gets a real answer, because you have shown the person you are worth spending the effort on. It takes fifteen minutes and almost nobody does it.
Ask it even if you have to type it
The most common regret we hear after a session is not that someone asked a bad question. It is that they had a good one and sat on it.
Nobody in the room is judging your question. The speaker gave up half a working day to talk to students, which is not something you do if you are looking for a chance to be dismissive. And if you would rather not speak, type it into the chat. A typed question is a real question.
Show up for one, not all five
Five sessions across two days is a lot, and treating it as an attendance obligation is the fastest way to get nothing out of it. Pick the one or two where you have a real question. Prepare those properly. That is worth more than sitting through all five with your camera off.
Speaker Day is for students enrolled in Research 101. The schedule and the meeting room are in the classroom, on the Live sessions page — the link is not published here.