Sunday, December 14, 2008

The Tao of Grad School part 5

Part 5: Approaching your project
You are embarking on the journey to compiling your dissertation, defending your work and getting your Ph.D. Obviously, you should have a strong sense of ownership of the project, and the sooner you realize this, the more likely you are to maintain motivation throughout the journey.

Before I go on, I guess I should include a disclaimer where I outline how different advisors and project types afford more or less stringent guidelines on how much creative wiggle room you've got within a given project description, so find the outer perimeters of your personal scientific freedom by inquiry and caution rather than by crossing boundaries. That being said, you still need to feel an ownership of the project in order to function as a researcher rather than a lab technician. Not a bad word about lab technicians, but if you're a grad student, you are supposed to learn how to become an independent researcher. That's kind of hard to do if you're not accustomed to bringing your own ideas to the table.

In my humble opinion, your Ph.D. project can be divided into three separate levels: Big Picture Science and Detail Science. Obviously, the former category refers to research which moves the project frontiers towards the overall goal of the project and how the results of this project can be used to (hopefully) move the frontiers of your chosen scientific discipline. A caveat here would be that moving the frontiers of science is not necessarily a good thing; all things considered, you'd rather not be known as the guy (or gal) who set your scientific discipline back 20 years. Big Picture Science is (probably) how the project was pitched to the funding agency, and what will lead to some applied aspect of your work down the road. However; in order to understand your system well enough to advance the overall knowledge of the field, you need to perform a lot of Detail Science. This type of science is what you'll do in order to complete various subgoals in the project description, and this is where you'll be spending most of your time by far. For every time you get to publish some Big Picture Science, multiple minor issues/challenges/learning opportunities (formerly known as problems) arise which need to be solved in order to move things to the next level. If you do your job well and the scope is wide enough, you'll get more publications from this category of research.

What's really important is to never lose track of how these three levels of research are interrelated. When you do some specialized work, always keep your mind on how this fits into the big picture. Whenever you get to move towards the larger goals of the project, know the limitations, and what fundamental issues had to be resolved to get where you are. No scientific results are unconditional.

The keen observer might have noticed that I've referred to three levels of research yet defined only two. That's because the third level is what I call Stealth Science. Not everything you do is going to pan out the way you or your advisor planned it. As you get to know your system(s) better, you might start to think of directions which can move the project forward, but which do not necessarily exist in the subgoals. Longshots. Small-level research you can try on the down low to see if it's feasible and if the results derived from it align with your initial hypotheses and with the project goals. But also additional research within the defined subgoals, where data is collected for the purpose of having results on hand for a rainy day, when you're expected to show results but the initial guesses as defined in the project plan didn't work out. Either way, before embarking on your stealth science mission, make sure that you know if your advisor and/or project allows for this type of off-the-books independent research. I personally always did it myself and encourage it now, but this might not be the case for your advisor. To quote Ice Cube; check yo'self before you wreck yo'self.


Anders said...

In my humble opinion, your Ph.D. project can be divided into three separate levels: Big Picture Science and Detail Science.

There are three kinds of people: Those who know math, and those who don't.

(yeah, I saw the stealth science part, but this was too good to pass by).

Anders said...

Or the nerdy one: There are 10 kinds of people: Those who know the binary numeral system and those who don't.

Wilhelm said...

...better you than me, bro...nerd powah'