I want to be the best and have fun in whatever I do. I am very satisfied with my life so far, and I want to help you be so as well. I am quite unconventional, unlike the usual run-of-the-mill types of professors you would likely encounter. If you are interested, read on.I would also encourage you to talk to my collaborators for further reference; you can find them on my publication list.
I also blog about various issues related to research; look for tags "advice" and "research".
There are two very important things that you need to know: you must love what you do, and you must strive for nothing but the best.Love what you do
You must be really interested in research. And you must be doing this for your own sake rather than other external reasons, like family expectations or peer pressures. Just like every other profession, your long term success and happiness depend more on your passion than your talent. I can excel in research primarily because I absolutely love what I do, and I never think I am particularly talented.You should not pursue a research program if you are not all that interested in research. Ignore this advice at your own peril.
Be the best
I am only interested in doing the best kinds of work. If I want to publish a research paper, I go for the top venue (e.g. SIGGRAPH for graphics or CHI/UIST for human computer interaction). There is nothing inherently wrong about not aiming for the best; it is just not what I want to spend my time on. Life is too short. (My first name literally means "the best" in Chinese.)I do publish in other venues for special reasons. For example, I have several papers in I3D, which is more suitable than SIGGRAPH for practical little cute ideas. I also use smaller venues for the initial phase of student training, but I would expect you to ramp up and do top publications eventually.
I do not care where you came from or what kinds of performance you had before; if I take you I am going to help you be the best. If you are not fully committed for that, you should find someone else to work with.
As a follow up to my main points above, I look for people who are truly interested in (and suitable for) research, and who can produce first class works. I evaluate your personality (especially integrity and propensity to get things done), intellect (especially creativity and critical thinking), and communication + inter-person skills (which involve both personality and intellect).I care less about anything else, like the school you graduate from or your statistics (GPA, GRE, IQ, class ranking, whatever). When I look at your resume, I usually focus on the *actives* like projects or publications instead of the *passives* like grades or rankings.
It doesn't even matter much which fields you came from. I like to try different things, and I have the capability to create ideas that fit your backgrounds and interests.
Work for yourself
My favorite collaboration model is "one mentor, one apprentice". My design is to have each student working for him/her-self, so that you reap what you sow and how much you achieve is proportional to how hard you work. This will give you the right incentives, and is good for your training (you could not lean on other students) and your visibility (the fewer authors on a paper, the more likely you will get noticed; most of my papers have relatively few authors compared to the norm in my fields). However, after you are sufficiently mature, we can consider putting you in larger teams so that you can learn how to work with others.
Consequently, you can work anytime you like. What matters is your output productivity, not your input efforts. Plus, you are really working for yourself, not me, or anybody else, so it is up to you to decide how to allocate your time.
Asynchronous communication
I operate in a continuous and asynchronous style through mechanisms such as svn paper drafts and Google sites, due to how my brain works and the fact that I travel all the time and my collaborators are all over the world. I do not like discrete and synchronous meetings as practiced by almost everyone else. My brain works on all projects all the time, so discrete sync points are less productive for me. I also do not find face-to-face meetings very effective, as people tend to think with their mouths rather than their brains.
Consequently, you can work anywhere you like. I am an expert on remote collaboration. Most of my co-authors are located in different time zones from me. In one case, I did not even meet the first author until after the paper was accepted. This also eliminates the facility and environment issues; if you do not like the office, you can work elsewhere. (Never in the human history can an individual do so much with so little; a nerd can sit in the bedroom and conquer the entire world with one mobile computer + internet connection.)
Have fun
I advocate a healthy life style. I get enough sleep every night and enough exercise every morning. I never stay late, as I find it detrimental to my creativity. (None of my ideas are produced while sitting in an office.) You can have a similar healthy life style while being productive, if you are willing to follow my advice. (It is actually very simple: start early, and never procrastinate. But you will be surprised on much difficulty people have in following such simple rules.)
I am not a fan of micro management. I want to give you as much room as possible to think, and I do not mind that you make mistakes as they are the natural part of the learning curve. If you prefer to be told exactly what to do, you will not be a good fit with me. In fact, you will not be a good fit for research at all, and you should find a programmer job. (You will be much happier, and get paid well.)
I want you to work on a subject that you are really interested in. You will be happier and more productive, and I will be motivated to learn new things. I would prefer you to tell me what you are interested in and what you would hope to do, and I will try to come up with projects that can fit our interests and backgrounds. I can pull this off, and I know it is very uncommon as most other professors would prefer you to work on their stuff.