A document originally written by the evolutionary biologist Stephen C. Stearns, that can be viewed at:
http://www.unibas.ch/dib/zoologie/stearns/gradadvice.html
or in a more formatted form at:
http://ethology.zool.su.se/modestadvice.html
In compliance with the Everything guideline Cut-and-Paste Writeups Will Die, I will only reproduce the headings of particular sections that appear in "Some Modest Advice..." and occasional quotations from the text in italics. I will add my own modest advice, based on my own experience. Why? Because there are some things worth saying that are not in the original text. (Besides, I haven't made wu for months.)
Disclaimer: Coming from evolutionary biology, I am similarly predisposed to writing about life in science. Because I have no experience being a graduate student in the arts, I will leave this to the discretion of a more qualified Everythingian.
1. Always prepare for the worst.
Be cynical.
The better half of graduate school is learning to be a good scientist. One of the easiest mistakes to make is to get yourself caught up in the excitement of a new research project that hasn't been subjected to critical and sober examination. As a rule, do not take a new idea seriously until you've sat on it for at least a couple of weeks, as well as done some literature research. Talk to your fellow students about the idea; being proprietary is bad science. Avoid presenting your ideas to faculty members until you have justified them as having merit, especially if they don't frequently have the opportunity to see you in action. You don't want to make a costly bad impression for the sake of a half-baked idea. I would have saved myself a good six months of time and funding if I hadn't dashed off on speculative pursuits. Caveat: weird ideas are also where scientific innovation come from; just because you're critical doesn't mean you should stop dreaming!
2. Nobody cares about you.
This is mostly bad though somewhat useful advice. It is bad because its wording is unnecessarily harsh. Graduate school is frightening enough without having to believe that nobody cares about you. Your family cares about you. Your friends do as well; make sure to let them know when you're having troubles -- and if they're fellow students, then reciprocate. For most people, graduate school not so much of an academic challenge (most everyone in graduate school had excellent grades) as it is a psychological one. Do not despair! Graduate school can really suck; I've had an emotionally tormentuous time, and chances are that you will too.
It is good advice because professors have a lot of work to do, and they cannot be approached four times a day to be updated with every new development in your graduate career. If you have a result that you want to show to your professor, then double-check its accuracy first and present it in a clear and legible form. Do not require your professor to wade through a mess of unnecessary information; this is especially bad if the result turns out to be wrong in the end because of a very obvious mistake. You do not want to convince your professor that you are sloppy. If you want to discuss a new idea, then similarly take some time and write it out until it is coherent and organised. Do a little literature research so that you can back it up with some related material.
3. You must know why your work is important.
"...if someone hands you a problem, you won't feel that it is yours, you won't have that possessiveness that makes you want to work on it, defend it, fight for it, and make it come out beautifully."- S.C. Stearns
This quote encapsulates one of the best of reasons to be doing science. Unfortunately, you will frequently also be required to convince someone else that what you are doing is important. Many disciplines seldom have the luxury of the few that impact the quality of human life in obvious ways, through the application of the end products of basic research. Funding for basic research is becoming scarce, and without funding, it becomes very difficult to do science, no matter how self-motivated you have become.
4. Psychological problems are the biggest barrier.
When you are applying to graduate schools, you should be evaluating them at least as much as they are evaluating you. The laboratory you join will be where you'll be spending most of your waking hours for at least four years, and as many as eight. The wrong professor can make your life miserable; do not make the mistake of being blinded by prestige, which is unfortunately often negatively correlated with their quality as an advisor. (I personally know one professor who frequently drove his students to tears.) Decide what you can tolerate and what you want out of graduate school; it is possible to thrive in a hostile environment, and is sometimes worth it if the professor is brilliant, well-funded and well known.
Even with the most sympathetic of professors, your results can deal serious blows to your emotional health. Do not let your self-esteem be tied to the success of your research, or that of others. You are not competing with anyone else. Take time off. Do all those things that you enjoy that have nothing to do with science.