In this article I'm going to say something about how to do research, especially how to find a good research topic.
Since I'm a PhD student, I need to do some research in order to get my degree. These research topics should not only come from my supervisor, also should come from me. But it is really hard to come up with original ideas like relativity (Albert Einstein), quantum mechanics (Schrodinger, Heisenberg, Dirac, Pauli, Born, Bohr etc.) and chemical bonds (Linus Pauling).
So the question that puzzles every researcher is: How do we tackle this major issue? I think the best way is to think of this question as a research topic. Now let's analyze it.
The question that we want to answer is:
How to come up with new and important ideas in research?
First we need to clarify the question:
What is new and important ideas in research?
It is really hard to define what is "important". It is very subjective. But it is relatively easy to define what "new" means, since there are only several things that could be new in scientific research (ordered by importance):
1. New scientific concepts (e.g. "second-order phase transition")
2. New experimental techniques (e.g. "STM")
3. New phenomenon (e.g. "super-conductive graphene")
4. New explanation (e.g. Feynman's explanation for liquid Helium)
5. New experimental / theoretical data
Usually, if we did something, and it hasn't been done before (or has been done a long time ago), then these data is valuable for level 5 "New experimental / theoretical data", it can be published, but only in a mediocre journal.
If a researcher want to publish in high-profile journals such as Nature, Science, JACS, Angew Chem, etc. (Since I'm a chemist, so I only list journals in chemistry)., they need more than just provide the data. They need to extract some new concepts, new theories, and new predictions from that, it would be much better if they found something that can shatter the old believes (e.g. T. D. Lee & Z. N. Yang).
But we cannot make graphene out of a tape everyday, so we must find a way to generate something "new", this comes to our second question:
What is the purpose of scientific research?
The purpose of scientific research is solely for the advancing of science itself.
If we agree on this, then make "new" discovery becomes less hard. Why? Because if we follow this purpose, then the "new" things means that "no one has ever done it before".
If we treat the whole scientific area as a pyramid, then different ideas and discoveries are the bricks of the pyramids. Our researchers want to build a solid, comprehensive, complete, coherent understanding of our universe.
As a pyramid needs to be built one bricks at a time, the only purpose of our research is to help our fellow colleagues go further. So, now we can answer the previous questions, if we want to make "new" things, we need to think "how we can elevate our colleagues to tackle the bigger problems by providing them useful tools, knowledge and techniques?"
Once we have that, then the goal of research suddenly changes from "finding the truth of nature" to "help other researchers achieve bigger goals". It's just like GitHub, if you can provide more useful software / packages, then you will get more attentions. It cannot help you find out the "true nature" of computer, but it surely can help you find a good position in tech.
Scientific community is the same, the only way you do what you do is that you believe what you do will matter to others someday. And the more useful your findings are, the more famous you will become.
Now we know that: firstly, we need to discover "new" things; secondly, "new" things mean that the things which are useful for the scientific community. Now let's dig deep into the question, and try to come up with an algorithm to help us find research topics. For that, we need to ask our final question:
What questions can we ask when we read a paper?
Most of our time in research, we spend on reading papers. If you don't read papers, you are dead.
Your ideas may come from your supervisor, colleagues, god, or maybe your mother, but they are temporary and need a bit of luck. We want a permanent and workable solution for everyone.
Ironically, the solution to our problem is: Ask as many questions as you can.
If you ask enough questions, the new research topics will emerge, since it originates from published scientific papers, means that this question will be interested by the relative community, so congratulation!
In below, I will provide several questions which I think could be useful in reading papers.
Notice: if you ask a question, you need to answer it. If you can't, that usually means you find a new opportunity, or maybe you are not smart enough (: .
This is the list of questions / thoughts that you can ask yourself / think about during reading a scientific paper:
1. What is the system/question/idea in this paper? Do I know something about that?
2. What is the authors' angle? What questions do the authors want to know?
3. Have they provided sufficient information in order to answer the question? Is there a better way?
1. For an experimental paper, theoretical guy may ask: Can I confirm it theoretically? Is this property / phenomenen universal on other similar systems? (such as bimetallic surfaces, perovskite)
4. If this is right, that means ...... (your predictions)
5. Can I use the method / technique provided in the paper to solve XXX ? (Usually you can borrow ideas from another field and then apply it to your own field)
6. Are there any counter-examples of this phenomena? Find them and figure out why that happens. (This is usually a very good strategy)
In conclusion, we talk about what is a good research topic, what is "new" and how to find "new" things through reading papers, here are the three most important things that you might want to remember:
• The "new" things in scientific research usually means something that you can do which has the ability to make life of other's easier / provide them more insights and tools for their own research. You don't have to find the answer, you need to find the next brick that can help your colleagues achieve higher goals. (Just like building a pyramid)
• The best way to find important and useful ideas is through reading papers and asking lots of questions.
• The list of questions that may be helpful for you. If you have more useful questions, please comment in below, and I will gladly add it into the list.
Thanks for your time, take care! Keeping doing good research!
See you in next article!
Since I'm a PhD student, I need to do some research in order to get my degree. These research topics should not only come from my supervisor, also should come from me. But it is really hard to come up with original ideas like relativity (Albert Einstein), quantum mechanics (Schrodinger, Heisenberg, Dirac, Pauli, Born, Bohr etc.) and chemical bonds (Linus Pauling).
So the question that puzzles every researcher is: How do we tackle this major issue? I think the best way is to think of this question as a research topic. Now let's analyze it.
The question that we want to answer is:
How to come up with new and important ideas in research?
First we need to clarify the question:
What is new and important ideas in research?
It is really hard to define what is "important". It is very subjective. But it is relatively easy to define what "new" means, since there are only several things that could be new in scientific research (ordered by importance):
1. New scientific concepts (e.g. "second-order phase transition")
2. New experimental techniques (e.g. "STM")
3. New phenomenon (e.g. "super-conductive graphene")
4. New explanation (e.g. Feynman's explanation for liquid Helium)
5. New experimental / theoretical data
Usually, if we did something, and it hasn't been done before (or has been done a long time ago), then these data is valuable for level 5 "New experimental / theoretical data", it can be published, but only in a mediocre journal.
If a researcher want to publish in high-profile journals such as Nature, Science, JACS, Angew Chem, etc. (Since I'm a chemist, so I only list journals in chemistry)., they need more than just provide the data. They need to extract some new concepts, new theories, and new predictions from that, it would be much better if they found something that can shatter the old believes (e.g. T. D. Lee & Z. N. Yang).
But we cannot make graphene out of a tape everyday, so we must find a way to generate something "new", this comes to our second question:
What is the purpose of scientific research?
The purpose of scientific research is solely for the advancing of science itself.
If we agree on this, then make "new" discovery becomes less hard. Why? Because if we follow this purpose, then the "new" things means that "no one has ever done it before".
If we treat the whole scientific area as a pyramid, then different ideas and discoveries are the bricks of the pyramids. Our researchers want to build a solid, comprehensive, complete, coherent understanding of our universe.
As a pyramid needs to be built one bricks at a time, the only purpose of our research is to help our fellow colleagues go further. So, now we can answer the previous questions, if we want to make "new" things, we need to think "how we can elevate our colleagues to tackle the bigger problems by providing them useful tools, knowledge and techniques?"
Once we have that, then the goal of research suddenly changes from "finding the truth of nature" to "help other researchers achieve bigger goals". It's just like GitHub, if you can provide more useful software / packages, then you will get more attentions. It cannot help you find out the "true nature" of computer, but it surely can help you find a good position in tech.
Scientific community is the same, the only way you do what you do is that you believe what you do will matter to others someday. And the more useful your findings are, the more famous you will become.
Now we know that: firstly, we need to discover "new" things; secondly, "new" things mean that the things which are useful for the scientific community. Now let's dig deep into the question, and try to come up with an algorithm to help us find research topics. For that, we need to ask our final question:
What questions can we ask when we read a paper?
Most of our time in research, we spend on reading papers. If you don't read papers, you are dead.
Your ideas may come from your supervisor, colleagues, god, or maybe your mother, but they are temporary and need a bit of luck. We want a permanent and workable solution for everyone.
Ironically, the solution to our problem is: Ask as many questions as you can.
If you ask enough questions, the new research topics will emerge, since it originates from published scientific papers, means that this question will be interested by the relative community, so congratulation!
In below, I will provide several questions which I think could be useful in reading papers.
Notice: if you ask a question, you need to answer it. If you can't, that usually means you find a new opportunity, or maybe you are not smart enough (: .
This is the list of questions / thoughts that you can ask yourself / think about during reading a scientific paper:
1. What is the system/question/idea in this paper? Do I know something about that?
2. What is the authors' angle? What questions do the authors want to know?
3. Have they provided sufficient information in order to answer the question? Is there a better way?
1. For an experimental paper, theoretical guy may ask: Can I confirm it theoretically? Is this property / phenomenen universal on other similar systems? (such as bimetallic surfaces, perovskite)
4. If this is right, that means ...... (your predictions)
5. Can I use the method / technique provided in the paper to solve XXX ? (Usually you can borrow ideas from another field and then apply it to your own field)
6. Are there any counter-examples of this phenomena? Find them and figure out why that happens. (This is usually a very good strategy)
In conclusion, we talk about what is a good research topic, what is "new" and how to find "new" things through reading papers, here are the three most important things that you might want to remember:
• The "new" things in scientific research usually means something that you can do which has the ability to make life of other's easier / provide them more insights and tools for their own research. You don't have to find the answer, you need to find the next brick that can help your colleagues achieve higher goals. (Just like building a pyramid)
• The best way to find important and useful ideas is through reading papers and asking lots of questions.
• The list of questions that may be helpful for you. If you have more useful questions, please comment in below, and I will gladly add it into the list.
Thanks for your time, take care! Keeping doing good research!
See you in next article!
Comments
Post a Comment