Advice for Students
This text started as a somewhat random list of things, which Florian has ever noticed are interesting or important for a student to know but which sometimes nobody tells them. Later Deyan Ginev put it on Authorea to make it accessible to a wider audience.
This text is not particularly well-organized but still an extremely valuable resource for students. Many points apply in general; some are specific to Florian’s field of theoretical computer science; some are specific to him.
Note that this was advice was collected over a decade, originally without any intention to publish it. Therefore, it lacks attributions even though some of it is advice Florian got from other people.
There is a natural synergy between us. I want to (in the beginning) educate you so that you can contribute to research (usually later). You want guidance and opportunities to do independent research. That usually works out very well, and you should always remember that we are working together.
Do not try to impress me by pretending you understand something when you don’t.
Do not hide problems from me.
Tell me when you don’t have time to work on something.
Come to me when you need help – speficic to our joint work or general academic help.
Tell me when you do not like to do something – there are always lots of other things you can do instead.
Do not pretend you are interested when really you are not committed. Then we are both better off working with someone else.
I like meeting with students, and I will usually offer you some kind of regular meeting. But you should be aware of some things:
I have very little time, and the time I offer you is valuable. If I feel, I am not getting anything out of it, I will stop advising you.
I will arrange all my meeting slots to optimize the use of my time. Thus, even when you arrive on time for a meeting, I might still be in another meeting, in which case you may have to wait a bit. Always come in in that case
otherwise, I might not notice or forget that you’re waiting.
If I reserve a meeting slot for you, I expect you to show up. If you have a good reason to skip a meeting, you usually know about it well in advance.
It is a common mistake of students to skip meetings with their advisor because they feel they have not done or understood enough. In fact, those are the most important points at which to meet.
It is very useful to bring a voice recorder and a camera to meetings so that you can record what we say and draw on the board. That gives you more time to think during the meeting and maximizes how much you get out of it.
My preferred means for quick communication is skype. My user name is florian.rabe. When you contact me, start by stating your question or problem. I will usually attend to it on the same day. Sometimes I don’t have time, and I will forget to answer; you should wait around one day before reminding me. However, if you come to me with questions that could be asked using Google or reading lecture notes, I will tend to ignore your message.
If I write rather curt emails to you, that (usually) does not mean I’m mad at you – it’s just an efficient way to communicate.
As you make progress in your research, your work will become more and more challenging and independent. You will notice that more and more things do not work – that’s because you’re reaching the end of current research where some things just aren’t done yet. Talking to me (or someone else) will often be the only way to get the information, understanding, or help you need.
On the other hand, I have very little time to talk to you. You will be most successful if you learn to optimize the benefit you get out of conversations/emails with me:
Try to ask questions in a short and precise way so that I can answer fast and conveniently while giving you the important information.
The time it takes me to reply to an email or IM is exponential in the length of the message. (quote from Alan Bundy)
If we have a meeting, try to write your problem down before, and take a print out to the meeting.
If you think you know what I should do to help you, don’t be too shy to give me an exact list of instructions.
I (or other advisors) will occassionally tell you briefly that certain things may be interesting to you. Cherish these ideas and pointers even if their purpose is not immediately obvious to you. They are almost always valuable to you, but often you need time to understand why and how.
Talk as much as possible to your advisors and your peers about your and their research.
If you don’t talk to your advisor about either your results or the problems you encountered, you probably did not do anything valuable (and your advisor will assume that!).
From your advisor’s perspective, there are two problems when communication is not frequent:
I get worried. Many students have difficulty coping with stress (course work, jobs, research projects; family, relationship, performance pressure, etc.). This leads to phases of psychological exhaustion at least occasionally. Then I need to support them. Moreover, students are often in denial about that; so it’s difficult to judge what it means when they tell me they’re doing fine. Keep in mind that a lack of communication is my only indicator of such situations.
Doing any kind of useful research work requires a tight feedback loop between student and advisor: The quality of a student’s work is exponential in the amount of interaction, and crossing the usefulness threshold is hard. I will usually not even want to look at what a non-tightly-supervised student did.
I can’t force when I have ideas for your work. Often they come a day after we talked about a problem that occurred in your work - so if you have an office space close to me, it’s good to be around after a meeting.
You should seek to form cross-connections with other students in the research group. Often you will be able to solve problems faster if you don’t have to wait for an answer from me. In some cases, this can lead to collaborations that produce joint papers. (Writing two papers together can be easier than one paper alone.) Or it can lead to older students coadvising younger students (which is very helpful for both). In any case, you will benefit a lot from broadening your knowledge. Being able to give a 1-minute summary of someone else’s research is often the difference between a good and a very good student.