Today in the Computer Science building, Immanuel asked us if we “would hire someone with a M.S. in computer science?” This sparked a conversation about the general state of computer science graduate school with a special attention to the M.S. Tyler said that it seemed to be getting harder to find good research topics for graduate work and he wondered how that would affect computer science in general. This basically boils down to the observation that it’s easy to find research in C.S. and this can’t last forever.
Compared to physics and math, Computer Science is a new field. This means two things. First, there is a lot of “low hanging fruit,” as Dr. Atallah so eloquently puts it, meaning there are many topics completely unexplored in C.S. (right now the big branch is security). As C.S. matures, people will have to work harder to find fresh ideas for research. I think this is what Tyler was talking about when he noted that research seems to be running dry. Maybe research is running dry by today’s C.S. standards but, compared to physics and math, C.S. has more research to go around than can be published! In fact, it’s not uncommon for undergraduates to have one or more publications. This precedence is definitely something unique to the youth of a new field. So as research topics become more scares, the top C.S. schools will adjust their expectations accordingly.
Another point which draws from Tylers observation is that, in early C.S., most research being done could almost directly be applied to industry. MIT, Berkeley, CMU, …, were all doing research on topics like operating systems. This is the obvious place for research to start because this is where the case studies lie. As the industry related research topics are published, academia will become more distinguished from software engineering. Even now it is becoming easy to see the difference between computer science research and software engineering. Of course there will always be research done in industry and vice versa but still, the areas will have matured into distinct entities. The same that happened with math and physics many years ago.
So why won’t C.S. research dry up completely? Because research isn’t really drying up, it’s just becoming more academic. It seems that the undergraduate computer science degree is leaning more towards the software engineering side to balance this out and the deeper these distinctions grow, the more space will be left for another degree. The Masters of Science. I don’t know much about the history of the U.S. education system but I doubt that the M.S. emerged along side the Ph.D. like we think of them today.
Ironically, today on reddit.com an article by one of Purdue’s most distinguished professors made it to the top of the list. Comer’s article was about how a Ph.D. should be about research.
“research means searching to uncover the principles that underlie digital computation and communication. A researcher must discover new techniques that aid in building or using computational mechanisms. Researchers look for new abstractions, new approaches, new algorithms, new principles, or new mechanisms.”
If this isn’t your goal, you shouldn’t be getting a Ph.D. From Mor Harchol-Balter’s famous Applying to Ph.D. Programs in Computer Science paper:
“A Ph.D. is a long, in depth research exploration of one topic. By long we’re typically talking about 6 years. By in depth we mean that at the end of the Ph.D. you will be the world expert or close to it in your particular area. You will know more than your advisor about your particular research area. You will know about your research than anyone at your school. By one we mean that by the last couple years of your Ph.D., you will typically be working on only one narrow problem. The Ph.D. is not about breadth, it is about depth.”
Conversely, this is where the M.S. fits in higher education. Computer science has it’s places for all types of degrees and separations of areas and as the field becomes more mature, hopefully this will become more distinct and lead people to further specialize in what they really love.
- None Found