Home
Projects
Research
Publications
Teaching
Activities
Software
Other Stuff
Bio
|
So, you want to do a PhD...
In the beginning of your PhD (oh, early
days...)
So, you've chosen a supervisor and a topic for your PhD. Now that
you're about to embark on this journey it's important to clarify a
number of things that will help you make sure you know where you're
going.
- Discuss with your supervisor what is expected from your
PhD:
- What are the main expected outcomes of your PhD? Together
with your supervisor put together a plan for the rest
of your PhD that states exactly what is
expected when. Learn to stick to these
deadlines. Obviously there will be situations where you miss
deadlines, but having a clear plan for your PhD gives you a
better picture of how far away you are from your 'ideal' PhD.
If you have to revise parts of your plan, do so carefully and
always after discussing it with your supervisor.
- What are the key conferences and journals you are
supposed to publish in? This differs from area to area and
your supervisor can provide you with this information.
- How many key publications are enough? Again this
differs from area to area. As a rule of thumb: 2-3 conference
publications, one or two of which are extended into journal
publications. These publications are essential. They feed
directly into your PhD thesis later on, you get peer feedback
on them giving you a good idea how your work compares to the
state of the art in your area, and last but not least, they
keep your supervisor happy.
- Agree on a slot for regular meetings with your
supervisor. These meetings can be held weekly or bi-weekly.
Personally, I prefer to meet with my students weekly, but in
any case, make sure you meet with your supervisor frequently
and on a regular basis. The purpose of these meetings is to
discuss your progress, new ideas, technical problems,
etc.
- Estimate your financial situation. PhD stipends or
salaries are certainly not known to be generous anywhere in the
world. Normally, it should be enough though to allow you an okay
standard of living. If you don't get a stipend or have an
expensive fine wine habit to feed, and you don't come from a
well-heeled family, you probably have to earn some money on the
side. Most PhD students do so by working as teaching assistants
(TAs). Although TA-ing in moderate amounts does not interfere too
much with your PhD, don't overdo it as this can hinder progress
with your PhD. Start with as little TA-ing as you think will give
you enough money to survive. Remember, a PhD is a full-time
job, and your supervisor will expect you to treat it like
one. Despite its flexible working hours, you should expect to put
in at least 8 hours a day.
- Read, read, and read. It's important that you get a good
understanding of the research problems in your area early on. You
don't want to find out half-way through your PhD that somebody else
has already tried out your idea. Your supervisor should of course
know the area very well, but unlike what some of them like you to
believe, they are not infallible. I initially point my students to
the most important and relevant publications in the area which are a
good starting point for their own literature research. We then
discuss regularly the other approaches, and how the PhD student's
approach differs from them, either in one-on-one meetings or in our
reading group. In addition to just reading it's also a good idea to
write short summaries of the most relevant papers. All PhD theses
I'm aware of include a literature review or background chapter,
which should be written early in your PhD. Of course, you have to
revise this chapter later on, as the world around you does not stand
still, but having a first version of a literature review early on is
yet another thing you can tick off on your to-do list. Do not only
read papers but also one or two PhD theses by others. You can ask
your supervisor which ones he or she recommends. You might not
understand all the details at this point, but you get an idea what a
good PhD thesis looks like.
- Try to write your first paper within a year after you've
picked your supervisor and PhD topic. Since conference submission
deadlines tend to be not during the summer, this could mean that you
have to submit your paper before (preferably) or after the first
year has passed. Getting your first publication out in time is
important for two reasons: Firstly, your first publication will fill
you with immense pride (and rightly so) and it's a huge motivational
boost. Secondly, time is limited and getting things done in time
will prevent you from being stressed later on.
- Talk to the other PhD students in your group, especially
the ones that are a few years ahead. They have learned many of the
lessons the hard way. Talking to them can help you avoid falling
into the same traps. They also know the, lets say, peculiarities of
your supervisor which will help you convince that it's actually your
supervisor who's the weird one and not you. Talking regularly to
your fellow PhD students about your research also allows you to bat
around crazy ideas. Although these discussions should not replace
the discussions with your supervisor during your regular meetings
they are an important addition.
- Use any opportunity to present your work at group
seminars. This is a friendly environment where you can present
your ideas, your progress, and your results and get valuable
feedback from a wider audience, including colleagues of your
supervisor, you normally don't talk to on a regular basis. Plus,
practicing talks is important. These dry-runs will improve your
presentation skills considerably.
Next: Knee-deep into your PhD (in other words: 'being
stuck!')
| |
|
