Could Grad School Be Fixed?(false-epiphany.com)
false-epiphany.com
Could Grad School Be Fixed?
http://false-epiphany.com/2009/08/could-grad-school-be-fixed/
6 comments
This is quite a good observation: something reasonably close to apprenticeship is already in place.
Grad students are given enormous flexibility in their assignments—a little trick to buy them time to try out their own research ideas long before they get to the thesis. Grading is extremely lenient; it's just a formality to keep things within reason. Professors are always eager to help you work on something publishable.
Grad students are given enormous flexibility in their assignments—a little trick to buy them time to try out their own research ideas long before they get to the thesis. Grading is extremely lenient; it's just a formality to keep things within reason. Professors are always eager to help you work on something publishable.
Having both worked in industry and academia, I think the author considers so called "real world knowledge" as the only kind of useful knowledge. For example, a working programmer might learn Ruby on Rails very well by working with the framework but be completely oblivious to other frameworks or programming languages, not to mention theoretical computer science.
Academia, on the other hand, focuses more on a high level knowledge by focusing on publishing results in the form of short papers with the most central results and references to the most central knowledge about the problem in research literature.
I think that for someone who wants to work as a programmer grad school is indeed "broken", since it underemphasizes the first kind of knowledge. On the other hand, many advances to technology come from academia, so apparently that system also has its advantages.
Academia, on the other hand, focuses more on a high level knowledge by focusing on publishing results in the form of short papers with the most central results and references to the most central knowledge about the problem in research literature.
I think that for someone who wants to work as a programmer grad school is indeed "broken", since it underemphasizes the first kind of knowledge. On the other hand, many advances to technology come from academia, so apparently that system also has its advantages.
I disagree that the author considers "real world knowledge" the only kind of useful knowledge. If you read the preceding entry on "The Agony of Grad School" it clarifies that the main problem here is this person has a really high cost for context switching, plus a really low tolerance for distraction. That's a bad combination, because you end up thrashing all the time. In the preceding entry, he talks about how he wants to know what's going on in a theory of programming languages course (not really "real world knowledge" ;) but he can't due to the context switch overhead from real analysis.
It sounds like the author unfortunately hit a place that requires more classes and teaching before getting to the point where research becomes the full time 100% job. It also sounds like the author was trying to take several courses and be a teaching assistant at the same time. That's a lot of work for anyone, even people who are good at multi-tasking.
It sounds like the author unfortunately hit a place that requires more classes and teaching before getting to the point where research becomes the full time 100% job. It also sounds like the author was trying to take several courses and be a teaching assistant at the same time. That's a lot of work for anyone, even people who are good at multi-tasking.
I suppose I was wrong about the author's reasoning, but I still think my description applies to many people.
Personally, I think both camps have their problems. Academic people can be quite annoying when they talk about something like they know everything about it just because they've read some paper about it. On the other hand programmers knowing only one language and framework are equally limited in their view.
Personally, I think both camps have their problems. Academic people can be quite annoying when they talk about something like they know everything about it just because they've read some paper about it. On the other hand programmers knowing only one language and framework are equally limited in their view.
I feel the author is saying that knowledge should be tested against reality. The test should not be a quiz administered in a closed classroom. Knowledge should lead to outcomes in the real world. It should be hardened against the real world.
This post is interesting but hard to take too seriously. First, I don't understand the author's hostility to the idea of graduates taking courses. Grad courses supply general knowledge not acquired in undergrad and in-depth knowledge to prepare the student for research. There are reasons why these courses follow the format they do - certain disciplines are best learned by an approach closer to 'lecture than to 'project (Real analysis? Lambda calculus?). The author claims: "You pick up all that needed stuff by doing real work. When you find a gap in your knowledge, you fix it. The instructor might give you an exercise to help you fix it." But this is not always feasible in abstract disciplines - try to pick up real analysis in this way, and you'll probably end up doing a lot of proofs that look very much like the standard real analysis sequence but seem a lot more frustrating. This isn't to say we need to sit in classrooms for six hours a week, but teaching certain subjects (but not all) via projects is a waste of time.
The call for 2-week sessions instead of semesters seems to be based -- like several other suggestions -- on the peculiar preferences of the author rather than on sound argument.
It's easy to come up with one's own wish list of how graduate school should operate, and the author is rightfully cautious about the implementation. But it seems to me the problems raised aren't fundamental to graduate school. Maybe the classroom lecture is becoming obsolete; it arguably helps demoralize students by removing from them responsibility for their own edification. But this doesn't demand top-down reform of everything but the thesis.
We should save that for the undergrad programs.
The call for 2-week sessions instead of semesters seems to be based -- like several other suggestions -- on the peculiar preferences of the author rather than on sound argument.
It's easy to come up with one's own wish list of how graduate school should operate, and the author is rightfully cautious about the implementation. But it seems to me the problems raised aren't fundamental to graduate school. Maybe the classroom lecture is becoming obsolete; it arguably helps demoralize students by removing from them responsibility for their own edification. But this doesn't demand top-down reform of everything but the thesis.
We should save that for the undergrad programs.
"Lecture-and-exam classes don’t work. We’ve been doing them for hundreds of years now, so there is no more doubt about that."
That's funny, they seem to have been working fine for hundreds of years. I know they worked fine for me, even if they were a bit boring at times.
Seriously, though, at the end of the day, no single system will work for everyone. Many colleges have tried different educational models. Many of them failed, and those that survived mostly went back to the traditional model for a reason: it works.
That's funny, they seem to have been working fine for hundreds of years. I know they worked fine for me, even if they were a bit boring at times.
Seriously, though, at the end of the day, no single system will work for everyone. Many colleges have tried different educational models. Many of them failed, and those that survived mostly went back to the traditional model for a reason: it works.
And students asking "why" is a taboo in current educational model.
That's simply not true for the most part. Sure, some professors don't like to be questioned, but most do try to encourage this. Almost every educator I've had or worked with wants students to ask the hard questions because it generally shows that they're listening, learning, and thinking.
From the article:
"Ph.D. students should spend most of their time doing real, publishable research alongside their professors, working as apprentices."
Most computer science PhD programs I know of have as a goal to get you to this point. Well, sometimes you end up working alongside your peers instead of seeing your professor. Different places do have different ideas about how early in grad school you finish classes and how many you need, but after that research is your full time job.
Most computer science PhD programs I know of have as a goal to get you to this point. Well, sometimes you end up working alongside your peers instead of seeing your professor. Different places do have different ideas about how early in grad school you finish classes and how many you need, but after that research is your full time job.
yeah, everything I know about CS Grad School says that classes are a necessary formality. For the most part, TAing is too. Grad students live to do research
Not included in the extensive list of objections is the fact that any redefinition of the semester schedule would meet gigantic political resistance. Colleges have been doing that lecture-homework-exam thing for centuries now, and they're not going to change just because it doesn't work and some guy on a blog has pointed out an obviously better way.
Really, the current system does work. It just works somewhat poorly. But well enough to get stuff done. As bad as U.S. universities are, they are the envy of the world, and their list of accomplishments is truly vast and magnificent. That's why there would be so much resistance.
The political battle might take ten years or more of relentless effort, and still might not dislodge "the first mover". As usual, marketing trumps engineering.
Really, the current system does work. It just works somewhat poorly. But well enough to get stuff done. As bad as U.S. universities are, they are the envy of the world, and their list of accomplishments is truly vast and magnificent. That's why there would be so much resistance.
The political battle might take ten years or more of relentless effort, and still might not dislodge "the first mover". As usual, marketing trumps engineering.
The better part of learning is making mistakes, and most grading penalizes mistakes.
This is true; it makes students unadventurous. Fortunately, most science PhD programs already work nearly the way he's proposing:
- Students are attached to a lab early on, and mentored by a professor for the rest of the program
- Classes related to the degree usually allow enough flexibility in the assignments that a student can find a way to apply their lab's line of research to the current topic
- Students take 2-3 lecture-style courses and a scattering of seminars each semester for the first two years or so, then focus entirely on research (mentored, of course) for the rest of the program
- Grades are implicitly just a formality. Maybe not everyone figures this out, but you're doing things properly in science course, you will get an A. Don't even worry about it; just focus on your research and let the prof know you're learning something. In a PhD program, your report card is really your list of publications -- and that's another can of worms.
The biggest departures from this I see are in law, business and med schools, where there's a secondary goal of filtering students up front. (And they charge so much for the honor. Science grad school isn't so much a hefty up-front investment as a temporary oath of poverty.)
My program is a mix of CS, math/stats and biology, and it's interesting to see how the disciplines collide in the curriculum committee. CS and math/stats teachers do have a lot of material that simply must be lectured to students, with lots of targeted exercises along the way -- these professors push for more core courses. Science teachers constantly try to trim the core requirements so students can spend more time in the lab, doing real research, without distractions. I don't think this dynamic would have evolved if the department was simply following the rote lecture-and-exam tradition.