3 ms·

This article is long and makes a lot of good points about things that you can gain from receiving doctorate. It sort of obscures one point that I think is the m

This article is long and makes a lot of good points about things that you can gain from receiving doctorate. It sort of obscures one point that I think is the most important thing. You learn an incredible amount about what you're studying and what you're interested in, and an academic environment is where some people learn best.

tensor 14y ago
Regarding your last comment, "an academic environment is where some people learn best", I would put forth that an academic environment is useful for all people. There are very few environments in the world that give you the kind of access to information that an academic environment gives you. Being a graduate student at a major university gives you access to nearly every journal article in the world, numerous domain experts, and advanced labs for the physical sciences. There are few to no industry jobs that provide this level of information access to someone right out of undergrad. That said, graduate school is not for everyone. If your goal is to learn, and learn how to learn, then graduate school might be for you. If your goal is to make money, or learn enough to make money or produce popular apps, then graduate school may not be appropriate. It's also the case that for some areas, like computer science, there is an increasing amount of free resources that one can make use of outside of graduate school. Many journal articles are free, and one can always buy and read through textbooks outside of school.
- _delirium 14y ago
  On the latter point, many public university libraries also allow anyone to come in and read books, download journal articles, sometimes even ILL. The main question in CS, imo, outside of certain capital-intensive areas (robotics, HPC), is learning styles. Do you learn better taking formal courses, having a supervisor oversee your project, having formal evaluation milestones (quals, proposal, thesis, defense)? Some people do, some don't. I do think the process of doing research and thinking of your goal in terms of "research contribution" rather than "product" is valuable for many intelligent/curious people. I'm less sure (despite being an academic myself) that the formal academia route is the best way for everyone, though. Imo the most valuable part of academia is the interaction with a research community in an area that interests you: work on a project, collaborate with some people, submit papers to conferences, get reviewer feedback, revise things, present papers, chat at the conference banquet, get inspired with new problems or ideas, etc., etc. You don't strictly have to be an academic to do any of that, especially in CS, although a large proportion of people who do are. Some of it is just culture I think: I think a lot of people outside academia who could submit interesting conference papers just don't even have that possibility on their horizon as something they could do, or would want to do.
  - tensor 14y ago
    These are all good points that I generally agree with. It is also the case, however, that experiences do vary. For example, my supervisor was hands off to the point where I had to largely fend for myself. The quals and proposals were more things that got in the way of research rather than helped. This was the case so much that I ended up doing my proposals long after research was underway. The thesis and defence is useful in the same way that publications are. However, even there, if you get a bad committee with political or personal axes to grind, it can be very much not that helpful. My experiences with publications are the same. I've had good rejections that help me improve the work as well as several bad rejections (from conferences where you cannot respond) where the reviewer simply did not understand the research. In short, for some people a PhD is very hands on rather than overly formal.

tensor 14y ago

Regarding your last comment, "an academic environment is where some people learn best", I would put forth that an academic environment is useful for all people. There are very few environments in the world that give you the kind of access to information that an academic environment gives you. Being a graduate student at a major university gives you access to nearly every journal article in the world, numerous domain experts, and advanced labs for the physical sciences. There are few to no industry jobs that provide this level of information access to someone right out of undergrad. That said, graduate school is not for everyone. If your goal is to learn, and learn how to learn, then graduate school might be for you. If your goal is to make money, or learn enough to make money or produce popular apps, then graduate school may not be appropriate. It's also the case that for some areas, like computer science, there is an increasing amount of free resources that one can make use of outside of graduate school. Many journal articles are free, and one can always buy and read through textbooks outside of school.

_delirium 14y ago
On the latter point, many public university libraries also allow anyone to come in and read books, download journal articles, sometimes even ILL. The main question in CS, imo, outside of certain capital-intensive areas (robotics, HPC), is learning styles. Do you learn better taking formal courses, having a supervisor oversee your project, having formal evaluation milestones (quals, proposal, thesis, defense)? Some people do, some don't. I do think the process of doing research and thinking of your goal in terms of "research contribution" rather than "product" is valuable for many intelligent/curious people. I'm less sure (despite being an academic myself) that the formal academia route is the best way for everyone, though. Imo the most valuable part of academia is the interaction with a research community in an area that interests you: work on a project, collaborate with some people, submit papers to conferences, get reviewer feedback, revise things, present papers, chat at the conference banquet, get inspired with new problems or ideas, etc., etc. You don't strictly have to be an academic to do any of that, especially in CS, although a large proportion of people who do are. Some of it is just culture I think: I think a lot of people outside academia who could submit interesting conference papers just don't even have that possibility on their horizon as something they could do, or would want to do.
- tensor 14y ago
  These are all good points that I generally agree with. It is also the case, however, that experiences do vary. For example, my supervisor was hands off to the point where I had to largely fend for myself. The quals and proposals were more things that got in the way of research rather than helped. This was the case so much that I ended up doing my proposals long after research was underway. The thesis and defence is useful in the same way that publications are. However, even there, if you get a bad committee with political or personal axes to grind, it can be very much not that helpful. My experiences with publications are the same. I've had good rejections that help me improve the work as well as several bad rejections (from conferences where you cannot respond) where the reviewer simply did not understand the research. In short, for some people a PhD is very hands on rather than overly formal.