Valerio Scarani CQT

Personal answers to FAQs

TOPICS: A career in research: tips for undergrads, Honors, Ph.D., post-docs. To disappoint or not to disappoint your supervisor? Presentations, talks and the like

A career in research	Summary: seen from a professor's point of view, undergrads are a hope for the future, Ph.D. students are work-power, post-docs are brain-power.
Undergrads	Don't rush too fast on "research" projects. It's good that you keep an eye on hot topics; but if a topic fascinates you, learn the bases first. Personal experience: time to publish will come. This is also true at later stages: many times, upon learning the great ideas of some colleagues, I said to myself "I would really like to work on that topic". Then I worked hard, and a few years later I produced my personal contribution. You can rarely be THE pioneer, but you can become an important actor in the subsequent developments.
Honors, Master etc	Your Honors or Master project should be your first contact with real research (if you don't like it, it may be the last). This is not always obvious, for at least two reasons: At this stage, your initial enthusiasm has probably disappeared. The only thing you know, like good Socrates of old, is that you don't know anything. Whence the tendency towards "filling gaps" in your knowledge, rather than setting for research and discoveries. There is a fear that the choice of research project may orient your career forever, whence the hesitation in going for something too specific. These are valid concerns. My answers are: Some gaps will indeed be filled at some point: practice in research is actually one of the best ways to become acquainted with the basics as well. Some other gaps will sadly never be filled. In any case, the happy time where you were only asked to solve well-defined exercises is gently fading: your Honors or Master project may well be the last of such exercises. The choice of these projects does NOT orient your career forever. Personal experience: my Master was in mathematical physics, my Ph.D. in experimental solid state physics, after that I became an "applied theorist" in quantum information science. Not that my career is exemplary in any way. But it shows that, with some effort and the good amount of luck, one can re-orient it.
Ph.D.	A Ph.D. is definitely about research. Sure enough, there will be graduate school to improve your general knowledge. Sure enough, your supervisor will provide you with basic training in your discipline and maybe send you to some summer school. Sure enough, you will not be requested to publish a research paper in the first six months, nor in the first year, maybe not even in the first two years. But ultimately you are expected to become a researcher, not just to "learn more", and even less to "learn from a renowned master". This has obvious implications on the choice of the topic and the supervisor. As a student, you liked some fields, you appreciated well-defined problems and very pedagogical teachers that never forgot hbar in their equations. Now: Field. In some fields, almost all that could be done has been done. Most probably, you (and I) know very little about those fields; ideally, it would be good to know more. But you should not ask yourself "what am I going to learn?"; rather, "What are the typical contributions to research in this field? What is the content of the papers that are published these last years?". Project. In research, a well-defined problem has already been solved. You have to look for a problem waiting to be defined. In other words, your Ph.D. project should sound like "we are going to explore this and that, not exactly knowing what we are going to find, with the hope that this is the right approach to [particular open question]". Supervisor. Of course, the "Best Teacher Award" does not tell anything about research. But the equations "good researcher = bad teacher" and "good teacher = bad researcher" are equally wrong. If your candidate supervisor is not able to explain you what his/her research is about, this is not deep thought, but foggy thought: go and look for someone else!
Post-doc	Under the guidance of your Ph.D.supervisor, you have brought a research project to its completion. Now you are a researcher. Next step? Learn to define your research projects yourself! The person who hired you as a post-doc is no longer your "supervisor": he/she is the head of the research group in which you are integrated. You shall surely learn from him/her, as well as from the other members (after all, we all learn all the time from each other). But what used to be enough during the Ph.D. (namely, to take vague indications, make them concrete and get the results) is no longer enough: you are now supposed to contribute with your own ideas. ... or didn't you know that the brain energy of your boss, when it is not spent in bureaucracy, is indeed mostly spent in trying to find new ideas? You are going to be appreciated if you help him/her in this task.

To disappoint, or not to disappoint?	Summary: There are three situations in which students may feel afraid of their supervisors: "I have made a mistake", "I do not understand" and "I am not progressing fast enough". One should learn to handle these situations. Unfortunately, many students end up doing exactly the opposite of what should be done. Learn why!
Mistakes	Mistakes do happen to anyone, including your supervisor... and including you! Let's put it dramatically: you discover that your latest result, one your supervisor was very excited about, was wrong. How to react? Correct: sleep one night over it, no less no more: maybe it can be easily saved. If it cannot be saved after one night of sleep, go and tell your supervisor. Wrong: do not mention the issue and hope that either you'll fix it by working hard the next month, or your supervisor will forget about the whole topic. Reason: your supervisor is always thinking about the progress of the group, planning the next step etc, based on what you reported. Also, he/she may be going to conferences and mentioning your result. You should stop this as soon as possible. Of course, if the mistake is big and/or costly, there will be a moment of disappointment: you can't avoid that. But if you try to hide the fact, the consequences are far worse: and, as a bonus, your supervisor will not trust you any more.
"No clue"	You feel you have not understood what the goal of your project is. Correct: go and tell it to your supervisor: "I have not understood this". The supervisor will gauge whether it's only a matter of reading one or two nice articles, or if something deeper is at stake Wrong: execute the orders (compute what you were asked to compute, measure what you were asked to measure). Go back to your supervisor with the brute results; hide the fact that you do not understand what you have done. Reason: if you don't see the reason yourself... I strongly discourage you to work in research.
Slow progress	If your project is in the "slow progress" mood, and you fear the disappointment of your supervisor for that, check this list: (i) you are working hard to fix a mistake you want to keep secret: see above; (ii) you have no clue, so you are just waiting for the next input of your supervisor: see above; (iii) you are working far less than the expected hours. If it's sheer laziness, wake up; if you have some real worries, mention something to your supervisor. If you feel that none of these point applies, then maybe you are working on a really serious project, in which progress is slow by definition: do not worry!
Final disclaimer	On all these topics, a little bit of "finesse" is required to gauge a concrete case (and a given supervisor).

Presentations	Summary: whom are you going to address? Which message do you want people to take away?
The message	The success of a presentation depends almost entirely on the following two issues: A correct gauging of the audience. The choice of the message to be conveyed. At the moment of preparing your presentation, ask yourself: what do I want people to remember after having listened to my talk or seen my poster? Most often, ONE message is enough. Oh, and don't forget that you are always going to convey a message: if you don't manage to convey yours, most probably it will be "we did not understand what the presentation was about".
The structure	At any level, I strongly suggest a tri-layered structure: Part 1: everyone in the room should be able to follow; Part 2: people who work in the field should be able to follow; Part 3: only the very experts should be able to follow. A division 1/3-1/3-1/3 is a good balance. There are of course exceptions: Part 1 should be shortened if you are participating in a specialized workshop; Part 3 should be shortened in front of a general audience, or if your result is actually very minor (e.g. in a Honors presentation).

more soon...