In the third episode of our "Books are people too" interview series, we have, as our guest, an extremely useful book for our young scientists and researchers. The name of the book is: "Advice to a young scientist". It was written by Sir Peter Medawar, a British scientist, who is regarded as the father of transplantation because of his work on organ and tissue transplants. Dr. Medawar was awarded a Nobel prize in 1960 for his ground-breaking work in the field. Advice to a young scientist (AYS) was written in 1979 and it has aged extremely gracefully to almost a wise sage or an oracle one should consult every so often.
FM: Welcome to our interview series. If you ask a kid to tell you what a scientist his, he or she would immediately respond with someone in a white coat blowing away stuff. Is that what you think as well?
AYS: Haha, not just a kid but most people in the society would associate a similar image with a scientist in their mind. However, I feel that we must always realize that scientists are people too! They are affected by the same nobilities and vices as any one else. The term "scientist" was coined in 1844 to describe someone who works in science.
FM: And how would you define science?
AYS: To me, science is nothing more than an exploratory activity whose purpose is to come to a better understanding of the natural world. And this activity is called research! However, there are other science based or scientific activites in addition to research as well such as scientific administration, scientific journalism, teaching science, scientific advisory, supervision, industrial process management, etc.
FM: So, according to you, anyone engaged in research is a scientist, No?
AYS: Yes! And we must never forget that there may be many practitioners of science and the scientific method who may not be officially labeled as scientists but are doing good quality science, for example, a pool cleaner who checks the hydrogen ion concentration of the water, keeps an eye on the bacterial and fungal flora and studies their effects on humans!
FM: So instead of asking what qualifies for good science, let me ask you what makes a "bad" scientist?
AYS: It is pretty simple, just like any other profession, if you are into science for the wrong reasons you will be terrible at it!
FM: What is or are the right reason(s) to get into science: curiosity, love of details, finicking?
AYS: Ummm... I think it is more than that. Good scientists I have seen have a sort of an exploratory impulsion or a restless endeavor to get to the truth of things! They feel uneasy if they can't understand or explain something they see and can't stop themselves getting to the heart of it!
FM: How can one find out if one is made to be a scientist?
AYS: I think the existence of inbuilt tendencies to be a scientist are over-glorified! Don't get me wrong -- they are important but skills required for science can be and must be learned as well. If someone thinks that their thinking can make them a good scientist without learning the skills required for doing science and experiments, they are, in my opinion, not made for science. Manual work such as doing experiments or, more directly, programming in your case should never be considered "below" science as these also help you think and be a better scientist! If someone thinks that being a scientist entails having lesser mortals such as graduate students fulfill his or her whims, they wouldn't typically make good scientists. The novice who tries his hand at research and finds himself indifferent to or bored by it should leave science without any sense of self-reproach or misdirection.
FM: So how can a young scientist pick a field or research topic?
AYS: It is pretty simple to say but maybe a little difficult to follow: if you want to make important discoveries, you must study important problems. Dull or piffling problems yield dull or piffling answers. By important problems, I mean to refer to problems whose answers would matter even if your work is not necessarily fashionable or sexy. It is perfectly okay to fall in line with a larger field of research, for example, molecular genetics or cellular immunology or machine learning but it is not advisable to fall in with a prevailing fashion for a new histochemical procedure or a new classification approach and just stick to it!
FM: What is your opinion about Ph.D. programs?
AYS: Human life can continue without a Ph.D. However, a Ph.D. is a license or passport for immigrating into any academic institution and it does have its merits. But it is important for a student to think why one wants to go for a Ph.D., what they want to work on, who they want their mentor to be and what would he do after a Ph.D. One of the major consequences of modern Ph.D. programs is that once one gets into a Ph.D. program, one is then expected to continue in a scientific or academic career because one doesn't know anything else to do! Therefore, it is imperative to really think hard before committing to such a career choice. After a Ph.D., a scientist must not continue working in the same direction for the rest of his life and should be willing to explore different things before settling on something they truly like.
FM: And about post-doctoral research?
AYS: Oh yeah! I feel that post-doctoral research is very important as it gives a budding scientist a chance to work on something they want to do, possibly in a different or related field, after having gone through their doctorate degree. As a consequence, post-docs are typically more motivated and make better researchers.
FM: Better than a thousand days of diligent study is one day with a great teacher. How can a scientist pick a mentor or a Ph.D. advisor?
AYS: The easiest, and not necessarily recommended, way is to pick the closest person at hand: the head or senior faculty in the department from which you graduated who may be looking for a disciple or an extra pair of hands. However, if a student understands or is proud of the work being done in the department then it is perfectly okay to do so but after much deliberation. The best way these days is to simply explore different scientists working in the field that a student likes and try to work with them! Don't just work somewhere without aligning your chosen field of research with what goes on there!
FM: How much should a scientist read to be successful?
AYS: I feel that most young scientists give too much emphasis on equipping oneself. They focus too much on reading the latest work and trying to know each and every new technique that is published to satisfy their sense of "being equipped to do great research". However, there is practically no limit to it. Too much reading also has psychological consequences and can be a hurdle to conducting actual or original research. It is like reading romantic novels all the time instead of trying to actually trying to talk to the girl sitting next to you! By far the best way to become proficient in research is to get on with it -- pick an important problem and work passionately in the pursuit of finding answers to it. I recommend to NOT learn new skills or mastering new disciplines until the pressure is upon to do so in this pursuit of finding answers; thereupon they can be mastered pretty quickly as there is plenty of motivation at hand. However, I must also say that reconstructing someone else's work or understanding a paper is important for two reasons: it helps you identify new and interesting problems or short-comings and it gives you confidence. Do it but don't over-do it!
FM: You refer to the art of research as "art of the soluble". What does that mean?
AYS: The art of research is that of making a problem "soluble" by finding out ways of getting at it -- how to easily attack a problem. One easy way I have found of attacking tough problems is to begin by quantifying things related to the problem such as how strong is the reaction of a patient to a transplanted tissue? Or, in your case of working on neural networks, how important is a particular neuron in a neural network? Typically, if you can quantify an effect you can make the problem soluble.
(to be cont'd)
FM: Welcome to our interview series. If you ask a kid to tell you what a scientist his, he or she would immediately respond with someone in a white coat blowing away stuff. Is that what you think as well?
AYS: Haha, not just a kid but most people in the society would associate a similar image with a scientist in their mind. However, I feel that we must always realize that scientists are people too! They are affected by the same nobilities and vices as any one else. The term "scientist" was coined in 1844 to describe someone who works in science.
FM: And how would you define science?
AYS: To me, science is nothing more than an exploratory activity whose purpose is to come to a better understanding of the natural world. And this activity is called research! However, there are other science based or scientific activites in addition to research as well such as scientific administration, scientific journalism, teaching science, scientific advisory, supervision, industrial process management, etc.
FM: So, according to you, anyone engaged in research is a scientist, No?
AYS: Yes! And we must never forget that there may be many practitioners of science and the scientific method who may not be officially labeled as scientists but are doing good quality science, for example, a pool cleaner who checks the hydrogen ion concentration of the water, keeps an eye on the bacterial and fungal flora and studies their effects on humans!
FM: So instead of asking what qualifies for good science, let me ask you what makes a "bad" scientist?
AYS: It is pretty simple, just like any other profession, if you are into science for the wrong reasons you will be terrible at it!
FM: What is or are the right reason(s) to get into science: curiosity, love of details, finicking?
AYS: Ummm... I think it is more than that. Good scientists I have seen have a sort of an exploratory impulsion or a restless endeavor to get to the truth of things! They feel uneasy if they can't understand or explain something they see and can't stop themselves getting to the heart of it!
FM: How can one find out if one is made to be a scientist?
AYS: I think the existence of inbuilt tendencies to be a scientist are over-glorified! Don't get me wrong -- they are important but skills required for science can be and must be learned as well. If someone thinks that their thinking can make them a good scientist without learning the skills required for doing science and experiments, they are, in my opinion, not made for science. Manual work such as doing experiments or, more directly, programming in your case should never be considered "below" science as these also help you think and be a better scientist! If someone thinks that being a scientist entails having lesser mortals such as graduate students fulfill his or her whims, they wouldn't typically make good scientists. The novice who tries his hand at research and finds himself indifferent to or bored by it should leave science without any sense of self-reproach or misdirection.
FM: So how can a young scientist pick a field or research topic?
AYS: It is pretty simple to say but maybe a little difficult to follow: if you want to make important discoveries, you must study important problems. Dull or piffling problems yield dull or piffling answers. By important problems, I mean to refer to problems whose answers would matter even if your work is not necessarily fashionable or sexy. It is perfectly okay to fall in line with a larger field of research, for example, molecular genetics or cellular immunology or machine learning but it is not advisable to fall in with a prevailing fashion for a new histochemical procedure or a new classification approach and just stick to it!
FM: What is your opinion about Ph.D. programs?
AYS: Human life can continue without a Ph.D. However, a Ph.D. is a license or passport for immigrating into any academic institution and it does have its merits. But it is important for a student to think why one wants to go for a Ph.D., what they want to work on, who they want their mentor to be and what would he do after a Ph.D. One of the major consequences of modern Ph.D. programs is that once one gets into a Ph.D. program, one is then expected to continue in a scientific or academic career because one doesn't know anything else to do! Therefore, it is imperative to really think hard before committing to such a career choice. After a Ph.D., a scientist must not continue working in the same direction for the rest of his life and should be willing to explore different things before settling on something they truly like.
FM: And about post-doctoral research?
AYS: Oh yeah! I feel that post-doctoral research is very important as it gives a budding scientist a chance to work on something they want to do, possibly in a different or related field, after having gone through their doctorate degree. As a consequence, post-docs are typically more motivated and make better researchers.
FM: Better than a thousand days of diligent study is one day with a great teacher. How can a scientist pick a mentor or a Ph.D. advisor?
AYS: The easiest, and not necessarily recommended, way is to pick the closest person at hand: the head or senior faculty in the department from which you graduated who may be looking for a disciple or an extra pair of hands. However, if a student understands or is proud of the work being done in the department then it is perfectly okay to do so but after much deliberation. The best way these days is to simply explore different scientists working in the field that a student likes and try to work with them! Don't just work somewhere without aligning your chosen field of research with what goes on there!
FM: How much should a scientist read to be successful?
AYS: I feel that most young scientists give too much emphasis on equipping oneself. They focus too much on reading the latest work and trying to know each and every new technique that is published to satisfy their sense of "being equipped to do great research". However, there is practically no limit to it. Too much reading also has psychological consequences and can be a hurdle to conducting actual or original research. It is like reading romantic novels all the time instead of trying to actually trying to talk to the girl sitting next to you! By far the best way to become proficient in research is to get on with it -- pick an important problem and work passionately in the pursuit of finding answers to it. I recommend to NOT learn new skills or mastering new disciplines until the pressure is upon to do so in this pursuit of finding answers; thereupon they can be mastered pretty quickly as there is plenty of motivation at hand. However, I must also say that reconstructing someone else's work or understanding a paper is important for two reasons: it helps you identify new and interesting problems or short-comings and it gives you confidence. Do it but don't over-do it!
FM: You refer to the art of research as "art of the soluble". What does that mean?
AYS: The art of research is that of making a problem "soluble" by finding out ways of getting at it -- how to easily attack a problem. One easy way I have found of attacking tough problems is to begin by quantifying things related to the problem such as how strong is the reaction of a patient to a transplanted tissue? Or, in your case of working on neural networks, how important is a particular neuron in a neural network? Typically, if you can quantify an effect you can make the problem soluble.
(to be cont'd)
