Ask HN: How to think like a researcher?

https://news.ycombinator.com/item?id=1749125

by mih • 16 years ago

8 7 16 years ago

news.ycombinator.com

Greetings!

So, I have started with a PhD program recently and am working on my research proposal. I put my idea on paper and it was shot down by my advisor for not having enough theoretical 'research questions'.

Also, I'm based in Europe. The trend I have observed is the US universities will tolerate research with maybe not enough theory and having more practical approach. In EU, a broader and more theoretical approach is required and some Profs. are quite obstinate in this aspect. This is just speaking of my personal experience and I could of course be mistaken.

Coming from a programming background, I am tuned to think in terms of what's technology is available rather than what can be conceptualized or assumed. This I feel is hurting my thinking ability as a researcher and I need to unlearn a few things.

Can any of the experts on HN offer advice/pointers/tips on how to fine tune thinking ; to start thinking more like a researcher and less like a programmer ?

mhewett• 16 years ago

I had exactly the same problem and never quite solved it despite ending up as a research scientist at a world-class university.

My best advice is to find an engineering-oriented advisor, even if they are from another department.

If that fails, find an interesting engineering problem, say "Execution of a distributed rule set across a distributed knowledge base". What are the fundamental problems? In this case, they are speed and consistency. Researchers don't care about speed, but they do care about consistency. So your topic becomes "Maintaining consistency in a dynamic distributed rule network". You solve this one theoretical problem, engineer a cool system to illustrate it, and get a PhD.

Best wishes and good luck.

ggchappell2• 16 years ago

I have a rather different viewpoint.

In my experience, people who belong in PhD programs generally do not need to be taught to "think like a researcher". Rather, they enter a PhD program because they already love to do research. They spend their spare time thinking about how to solve problems, and they attract friends with interests along the same lines. Thus, a PhD, which is essentially training and a credential for a professional researcher: it is a step toward the goal of spending their professional lives doing what they already love to do.

Based on your post, this does not appear to describe you. Why, then, are you in a PhD program?

I'm not trying to discourage you from pursuing your dreams, but I think the question is an important one.

mih2• 16 years ago

Unfortunately where I come from lack of true research opportunities are far and few. Just because I have not been exposed to research methods so far does not mean I cannot develop the necessary skills, when given the chance. Agreed, to some people research is natural and effortless. This however should not mean the less gifted ones among us give up without attempting.

My problem is this - when given a question I can give a solution. It is finding the question in the first place that is proving to be a challenge.

ggchappell2• 16 years ago

> Just because I have not been exposed to research methods so far does not mean I cannot develop the necessary skills, .... This however should not mean the less gifted ones among us give up without attempting.

Certainly. I wasn't questioning your ability, but your interest.

> Unfortunately where I come from lack of true research opportunities are far and few.

I must disagree. Any 8-year-old with a pencil and paper can do research in mathematics and the theory of algorithms. Give him a piece of glass, and he can get into optics. Send him outdoors and he can work on biology, chemistry, etc.

Of course, this won't be groundbreaking research. But the point is that people who like doing research are probably already doing it. They just need help getting to the front lines of a field, so that they can make significant, truly original contributions. That is part of what a PhD program is for.

I can't tell much about you from a couple of paragraphs--and you don't need to justify yourself to me, anyway. But some of what you said earlier raised a "red flag" in my mind. As a PhD student, you have, no doubt, been exposed to some of the important and/or difficult problems in your field. Have you tried solving any of them; if not, have you spent much time thinking about how you might go about solving them? More generally, do you enjoy problem solving? Do you work on problems when you are not obligated to? Do you talk about them with your friends (or wish you had friends that liked talking about them with you)?

If so, then, great, figure out how to overcome your current obstacle, finish your degree, and have a fun life being paid to do what you love.

But if not, then you might be investing a huge amount of time & effort preparing for a career that you'll hate. And that is hardly an outcome to be desired.

anigbrowl• 16 years ago

find some people who are new to this field (perhaps sit in on a first-year lecture?) and take note of questions that get answered with 'because you can't...'. Also, just because a promising idea fell at the first hurdle in the past doesn't mean that hurdle is still a barrier. If nothing else, reviewing the history of your field and seeing that some ideas went nowhere will get your imagination going. Perhaps even talk to retired practitioners and keep an ear out for stories about 'the one that got away' or that made them wonder 'what if I had done x instead' for years afterwards.

Alternatively, look for some imagination-based recreational activity, preferably one you're not familiar with so you find yourself generating ideas without necessarily knowing whether they are practical.

PS what are you studying?

mih2• 16 years ago

Thanks. I am studying geoinformatics, though I do have the freedom to solve problems in the field from the perspective of computer science.

CyberFonic• 16 years ago

The research proposal is one of the hardest things to write. You are expected to know what problems exist and how they could be solved. Assuming that you have a specific area of interest, use Google Scholar to look at the papers written on that topic in the last 2 years. You will probably need to invest 40-60 yours to do this well. Reading some recent PhD theses is also an eye-opener.

Having done this step you should have a better idea of the sorts of theoretical research questions that others have asked and thus provide you with ideas for yours.

The core point of a PhD is to make a significant contribution to the body of knowledge. It is very different to that of a Masters where you learn and apply advanced knowledge (developed by others). Looking at available technology is only likely to have merit if you find a deficiency and find a way of solving it. And a sound theoretical justification for your solution.

Your comments on US universities might apply to some you may have looked at. From my experiences Stanford, MIT, Berkley, Princeton have very high standards and demand for theoretical foundations. Edsger Dijkstra had noted the trend for stronger mathematical foundations in the European universities than the USA ones.

A PhD is an apprenticeship for becoming a researcher. So you will learn to think like a researcher as you progress. Most universities have introductory courses on how to do research, literature searches, etc. There are also excellent books are resources on the internet. Seek and you shall find.

Ask HN: How to think like a researcher?

Related Stories