Skip to main content
Create Date

Apostle of scientific theory and practice — Professor Kai-Yew Lum 

Tune in below for detailed responses from NCNU Professor Kai-Yew Lum (林繼耀) — a Singaporean scholar-engineer and home-baking enthusiast — on his approach to teaching, learning with software, the military and civilian applications of aerospace engineering, and the best course to internationalization for different school systems.

The transcript, which has been edited for clarity, is sprinkled with a healthy dose of life advice and bread-making metaphors and is credited with “KY” — the same signature the esteemed academic uses to sign off his emails.


ZC: Zoe Cheng of NCNU’s Secretariat Office, supplying the questions

KY: Professor Kai-Yew Lum, NCNU Dept. of Electrical Engineering


  • Ph.D. in aerospace engineering from the University of Michigan, Ann Arbor
  • Formerly with the Singaporean Ministry of Defense
  • Area of specialization: Automatic controls, specifically algorithms for feedback control
  • Research applications: Industrial processes, aeronautics, spacecraft, robotics, design optimization
  • Technical focus: Guidance and navigation of autonomous vehicles, aerodynamic control
  • University teaching: Mathematics, controls, and robotics
  • Current interests: Machine learning in robotics, the science of making bread
  • Practical philosophy:
  1. First theory, then practical application
  2. Mathematics is best taught with software
  3. Engineering is built on language and mathematics


You are a Ph.D. graduate from the University of Michigan in Ann Arbor, one of the top ten most renowned academic units that specialize in aerospace engineering.  Your academic career also starts from there. You are now a great member of the NCNU faculty. Please tell us how you came from the start point until today.


Right. I was born in Singapore, and I received my early education in Singapore until high school. And later I received my undergraduate training in engineering in France, at Institut National Polytechnique de Grenoble. I obtained my master’s and Ph.D. degrees in aerospace engineering at the University of Michigan Ann Arbor, United States.

So, before I joined National Chi Nan University, I was actually an engineer first, then later, a researcher for the Ministry of Defense in Singapore. That part of my career ended, or rather, I chose to end it because I had been dealing with research problems but not really people. So I thought I want to have the next phase of my career to have more impact on people instead of projects and problems. So I thought teaching would be a good way of making an impact, especially on young people and students.

And it is also a chance to share my knowledge, both professionally and academically, with students. So, when there was an opportunity at National Chi Nan University that sought people with international experience, I jumped on it.


How is your previous profession related to your current academic teaching?


Right, in a way they are both related even though I was doing research and engineering. In this process there is always the need to share knowledge among one another, among the team. There is always the need to keep abreast of developments in the field and I think that is something that I can bring to a teaching career. I bring my practical experience as an engineer to my teaching. I let my students know how things are done beyond just theory and textbooks. And I also bring, hopefully, a perspective that is more or less global in a sense. Regarding the field, regarding science and technology.


Can you share with us your valuable experiences from working at the Ministry of Defense in Singapore? Such as satellite technology collaboration with some institutes at the national level. Are those fields working together?


Yeah, they are all directly related. First of all, my field of specialization — both as an undergraduate and Ph.D. student and later in my career in defense technology — is in the area of control, automatic control. So you find the application of automatic control in, say, aeronautics like flight control. You find that in space technology; incidentally my Ph.D. thesis was about spacecraft control. You also find that in process control, which is an industrial application. And you also find that in robotics, which is what we find a lot in industries today. The use of robotics has become a major component of training in electrical engineering as well.

So, they are related in that sense. And the projects that I did when I was with the Ministry of Defense in Singapore, some of them are classified but a lot of them are fundamental, in terms of developing algorithms in the area of control. The goal of control is to gather information from a system and to decide on the actions needed to make the system behave in a way that you want it to behave. 

First, we collect information, process information, with sensors or other types of equipment. After processing that, we design an algorithm that will compute the correct type of action or the appropriate type of action that we want to apply to the system. And this is basically the kind of projects I’ve encountered, like guidance of vehicles, where unmanned vehicles need to be guided from point A to point B, avoiding obstacles and things like that. And this is also part of the fundamental control courses at the university level.

So, instead of just talking about equations, straight out of textbooks, I think what I can do is to offer my insight on how those theories are applied through real-life examples. And also for my research students, I assign them projects that are related to that.

From my experience, for example in using a camera to track something, I know that the camera has to point at something while, say, the vehicle is moving. Or to use a camera to guide an unmanned aircraft to land on the ground. Now these are the projects I have assigned to my students. And as you can imagine, they are related to military applications, and also civil applications.

So in a way, I think even though I’m here with the electrical engineering department because control is itself a discipline, it actually is multi-disciplinary, cross-disciplinary rather, meaning it can be applied to both electrical engineering problems as well as to other engineering problems.


You intend to make more impact on people instead of doing research at the lab. What prompted you to make such a great shift in career paths?


I think most researchers come to a point in their career where they review the work that they have done; with my coworkers we have authored more than 80 or 90 papers. But when I look at it, 80 or 90 papers that I co-authored is just a drop in the ocean of the entire field. Alright, and I just think it doesn’t make any impact, some of it may have some value to the field, some of it have been actually applied in to real system in my previous work. But most of it doesn’t have an impact, direct impact on a person. On a young person, on a student.

I had the opportunity to teach when I was in Singapore,  I was attached to the National University of Singapore. So I was a lecturer there for a short time, but the opportunity to teach, to impart knowledge over a long duration of time... I’m not saying that’s made a big impact and changed somebody’s life. But to be part and parcel of the process of knowledge acquisition or their process of personal growth, that is what I wanted to be part of.


So how do you teach or apply this knowledge to real life, to work?


In two ways, starting with the way I teach. In fact I teach four courses right now. Two of them are mathematics, freshman calculus and engineering mathematics for second-year students. And I offer two other courses which are electives, one in digital control for undergraduate seniors, and for graduate students, I offer a course in robotics.

For a student to do well, I think there are two areas he or she has to be strong in, and that is language and mathematics. Of course, by language I mean any language. Chinese, English, French, any other language. Language is the universal platform for the formulation of knowledge, for the expression of knowledge, and for the exchange of knowledge.

So a student has an advantage if he or she has a good command of language, you just take any good and well-written textbook and just look through the textbook. You’ll find that there are many more plain sentences than there are equations. So a student who is good in language is able to read more information from text. But that doesn’t mean that mathematics is less important.

Mathematics is just much more dense than plain language. And mathematics is a tool to describe physics, physical phenomena; it is also the tool, the common tool for logical deduction. And I think a lot of students have difficulties in engineering subjects, such as circuitry, electronics, robotics, simply because they find difficulty in speaking the mathematical language that is used to describe, to design, and to operate an engineering product.

So the way I teach in all my classes, I try to emphasize the usefulness of mathematics by making it as intuitive as possible, as well as practical as possible. This is what I, myself, have realized both through my own education as well as through my work. Many times I would have found that if the mathematical analysis is not done well, if an error is made, then the end product, the engineering product will have some problem. Either it doesn’t perform as much as well as you want it to, or it fails, right?

So the way I teach mathematics is I’d like students to appreciate the usefulness, how useful the formulas are, what the meanings are, not just simply bunches of equations that they need to remember the steps for and stuff like that.

In this day and age, it is not so relevant anymore for students to remember the exact steps, or work through a long calculation by hand. Because we can use software to do that. So I teach mathematics using software called computer algebra system, or CAS, and there are 2 CASs that I find most interesting, particularly useful for university education. And one of them is Geogebra.

Geogebra is simple to use: it has a nice interface, it has good graphics, 2D and 3D, it allows a student to visualize a function to appreciate what it describes, and also to appreciate, for example, how changing certain parameters changes that representation. So Geogebra is good for basic mathematics like calculus, differentiation, integration of common functions, it’s good for basic algebra. In fact, in some countries it’s used in high school education.

So I use that in my freshman calculus course, to let students appreciate both the usefulness as well as the beauty of mathematics. Because in my field, control itself is a branch of applied mathematics. And this mathematics, once well-established and  analyzed, is turned into algorithms by writing the mathematics in the computer program, letting the computer do the calculation and perform the action that are dictated by those mathematics. So that’s why I see mathematics are so important in the training of a student.

The other CAS that I find very useful not only for teaching but also for research is called Maxima. Now compared to Geogebra, Maxima doesn’t have a very nice interface, the graphics is minimal; it’s not really meant for graphics. But it is a lot more sophisticated than Geogebra, in terms of the kind of mathematical operations you can perform, and it has a much more expansive library of mathematical functions.

So I find Maxima very useful for higher-level courses such as robotics, so my students in robotics use this software to calculate very high dimensional calculations, for example, for an industrial robot. This calculation would have been almost impossible and extremely tedious, if you do it by hand.


But I first try to address the fundamentals using technology, using software. And that is where I emphasize practical work in my courses. Besides the fundamental math classes, all my classes and electives have a practical component that students are required to apply the theory that they have learned in class to a practical problem.

Now, the practical problem can take the form of software simulations, simple simulations by computer, but also hardware implementation. For example, in robotics I require my students to first, using software and using the mathematics that they have learned, design the movement of a six-degrees-of-freedom robotic arm. It is quite a complicated movement to perform a given task and I have a video of the robot tracing a perfect circle using a laser pointer. After they have tried that in simulation and verified that their algorithm worked, then they transfer the algorithm into the real robot.

So we have a robot in the lab, they have to put it into the robot and verify that it really works in real life. And I always sense the excitement in the students, now even though this is something like a 3- to 4-week project, a mini project, but I always sense the excitement in them when they can finally press the go button and see their hard work come to life. Even if it’s just a matter of simple trajectory, they are very happy to see it work.

So why do I emphasize practical work? First of all, the obvious reason is because in my view, they are trained future engineers, so throughout their training they should have a glimpse of how to apply knowledge to the real world, what the process and applications are, what form of application it can take.

But secondly and most importantly, I think practical work gives value to the knowledge, gives value to the theory, because I think in the mind of the student, this practical work connects the abstract with reality, it connects things they see in textbooks, on lecture notes, to something that works on the table or in the field. And I think that is a very important part of engineering and university education.


Does this process have anything to do with the robot behind you?


This robot named Now enjoys a high degree of freedom, it’s a robot, except it’s not bolted to the table, it can move around. This is a platform for learning many high-level algorithm design; for example, it can be used to teach students to program, say, voice recognition, so there are built-in libraries like building blocks, and students can take those building blocks and construct a certain behavior like recognizing a voice, like recognizing a face, and other things like motion control, so it might recognize something and walk towards that object or person.

This robot is easy to program, in fact. It might look daunting to someone who sees it for the first time. But it comes with a very nice interface, a development platform that allows even undergraduate students to program this robot to exhibit a certain behavior. So, behavior is a result of a set of algorithm or a set of logic.

So I use this robot as a tool for undergraduate students who volunteer on their own, not as a part of any course; we’re all doing this just because we are interested. There is no credit to be earned, but a few students take this up as their challenge and they want to make something work. For example I assigned a project and challenged a student to do it, so a bunch of year one students now want to do it, so hopefully they will continue working on it until they graduate and achieve some real results.

So the project involved the robot recognizing certain objects and putting those objects into boxes that they should belong. So it has to tidy the room, pick up something that’s lying on the floor, and put it back where it should go. So, it is also a fun way to learn the concept of machine learning, what we call artificial intelligence, but AI after all, it’s a control cube. It’s a much higher level of control, which means it is information, feedback, and action. And this is again related to the courses I teach, and the research I conduct.


Could you please give us some examples?


So I’m personally interested in robotics, so I’m put in charge of this robot. So yeah, but the students have barely started, so there isn’t any actual project that has been developed yet. Well I’ve written some codings but those are for school events, welcoming guests and things like that. But what I can share with you is how I lead students to gain practical experience after they have learned their basics in robotics.

I have a six-degrees-of-freedom robot in my teaching lab, and this robot can be controlled remotely by software. And the point of robotics is to be able to coordinate the motion of the 6 joints so that the tool, or what we call the end effector, here performs a certain task. So in my course I require my students to first use software. My students first use software to design a certain set of motions.

Using simulation software, we have a model, an exact model of the real robot, all the motions are exact and what the students have to do is design a set of motions of the joints, of the individual joints, but they all have to be coordinated. For example, the tool of the robot moves from one end to the other, like it has to trace a straight line at a certain altitude from target 1 to target 2; this is one of the basic exercises that they have to do.

As their final project, they need to program the robot to trace a circle. By tracing of the circle using a laser pointer, they are graded as to how well the circle is traced, whether they go off the circumference, or if the motion is not smooth.

Even though this is a simple task, but I hope students realize the challenges and also the techniques for dealing with these kinds of problems. And they can be easily encountered in the industry, in industry settings such as precision machinery.

Now, for my research students, I also require them to do both theory and experiment. You know some students, they prefer to just do paperwork because it’s easier, it’s clean, they don’t want to get their hands dirty. And some students just prefer hands-on work, and they said “don’t give me the difficult theory part, I just want to screw screws together” and stuff like that. But my students, they have to do both.

They have to first do the theoretical analysis and later apply that to an experiment. For example, we have an aircraft here that is called a quadrotor; it has 4 rotors, so it can move up and down and sideways, and the objective is for the aircraft to track the target on the ground, and then land on it automatically, without human intervention. So we have mounted a camera under the aircraft. We have put mini-computers on the aircraft; the algorithm captures the target and steers, constantly steers the aircraft towards the target, and maintains the aircraft above the target as it slowly descends.

So the theoretical techniques are called filtering; that means information processing, data communication. But the practical part of this is system integration, programming. Because to make that work, it has to take the form of a computer algorithm. So in this kind of research, not only do I want students to connect theory and practice, I also want them to develop system-whole thinking because the type of problems, especially the control problems that we deal with, will always involve a system with many components.

Information is being passed back and forth among the components, and the action of 1 component can affect the rest. So by developing system-whole thinking, I hope the students will not simply focus on one aspect of the problem, but be very aware of how the entire system is interacting within itself. Even though their thesis work is on a particular aspect, they have to think about the entire system.

So there are other things like using a camera mounted on a gimbal that is steered by a mini-computer on a cart, and the algorithm tracks again the target and after that it steers the camera always pointing at the target, even though the vehicle is moving back and forth.

You can recognize the military aspect of this. This is how guided missiles work. But this is not a guided missile, this is a very friendly little car, so through this students learn how to apply control, how to apply signal processing, information, and even image processing to a problem like this. In my other research, which I think also is representative of the kind of applied research that I do, is work on wing flutter.

Here we have the model of an aircraft wing, and this model is being tested in a wing tunnel. And what happens to aircraft wings like this, is that the interaction between airflow and the elastic structure will result in sustained vibration. And this can happen in the real aircraft, in flight. When the aircraft flies under certain conditions, this phenomenon can be triggered and it can be quite dangerous. Technically, it’s called flutter, wing flutter.

So we are trying to develop a way to control the flutter by means of a little flap here. So this is a flap that can move, and the objective is to capture the motion of the wing through sensors mounted here and then design a control algorithm to move the flap in the way known to suppress that vibration. If you do not design the control well, moving the flap could exacerbate instead of damping the vibration, making it worse.

So, in an applied control research like this, my approach which is the same approach for many researchers, the first thing is to fully understand the physics behind the phenomenon. Building mathematical description that is true. Analyzing data that you collect about the system, and then to apply the method, the control method that matches the type of problem. Now, there should be a match between method and problem, because if the method doesn’t match the problem, however beautiful it is, however popular the method is, it’s not going to work.

Sometimes we see researchers match the problem to the method, but that’s not the point. You should match the method to the problem, instead of the other way around. So in this kind of research, that paradigm of description, building the model, applying the correct method and finally doing an experiment, this is very important to me. Any research like this must end with an experiment because the experiment validates the method and it also brings out maybe aspects of how the method or how the mathematics deviate from physics. And if that deviation is important or how to mitigate that kind of differences.


So those who want to follow you must first excel at language and mathematics?


Well, we cannot require them to be excellent but they must be willing to learn. It’s not possible to demand students to be excellent in both language and mathematics, but first what I require of them is the willingness to learn and to recognize the importance of these two areas and be teachable. Because that’s my job, I’m here to teach. So, a teachable student can always overcome some of those difficulties.

A student has to be willing to be taught and know that the things they are being taught are useful. It has to be seen to be useful. And that is why I, as I said before, I rely on software to teach. One of the reason I think students find mathematics difficult or boring is that they are overwhelmed by the complexity, by maybe sometimes the tedium of working through tens and hundreds of steps of mathematics. And software can take away the tedium.

Let them focus on the concept and on the fundamental. Softwares like Geogebra and Maxima don’t replace knowledge, because to use it the users still need to know what mathematical operation he or she has to perform or the software has to perform. So that the right commands can be entered into the software and what kind of result should be expected and how to interpret the result. But the tedious part of doing the calculation is done by the software. Whereas the user, the student, focuses on the concept, focuses on the meaning and focuses on how to interpret the result.

So, I recognize that not every student is strong in mathematics or language, but through my interaction with them I try to make them find mathematics fun. Mathematics can be fun, alright, if you take away the tedious part.

And also language. I used to teach a course in English for technical writing for graduate students. Now the course is being taught by another professor. In that, I also try to let students appreciate how interesting language itself is.

The pattern that you can find between different words, how they come about, how a certain word comes about. Why do people say this? Why is it said the way it is said? I think that putting a fun perspective on every subject is in fact, the important paradigm of teaching, instead of making it very dull.


So how about the matter of bridging technology and humanity? The electrical engineering department emphasizes interdisciplinary training. Students need to learn how to present their ideas, to explain how their end product can benefit regular folks. Any thoughts on this?


Well, what you said is very important. Even though they are engineering students, they have to be able to present their ideas. They have to be able to present their findings and also be able to ask questions. Questions that are relevant that, will bring out the essence of the problem. So, I think, first and foremost importance should be given to human communications. Students have to develop that skill, and I hope there are more courses in the university that addresses that instead of the linguistic or literary aspects of a language, for learning how to apply and use a language effectively.

I also emphasize that they have to learn to communicate in English, because information in science and technology is almost always expressed or communicated in English. So by improving their command of English, they can gain more information.

Secondly, even in Taiwan, the industry needs engineers and staff to have a certain level of English proficiency. So how I teach them is: you can only talk to me in English or you can only write to me in English. I allow them to talk to me in Chinese, but I require them to write to me in English.

And if they like, in conversation they can mix English and Chinese; I welcome it when they mix English words, especially technical terms. I say “don’t bother to translate into Chinese, just use the technical terms in English” and I do have a few students who voluntarily speak to me in English, even for casual conversation. So they learn at different paces. And I allow that, I accept that. I don’t impose 1 model for all.

And I recognized that I’m not an English teacher. I’m a teacher in electrical engineering. So I try to suppress the impulse to correct their English. Right, because that is the sure thing that will put them off. And they will write less and less. Instead I just let them express in the ways they can think of, with a few grammatical mistakes and wrong choice of words and all that. The strategy that I developed is to reply them almost using their sentences but in a correct way. Without explicitly telling them“this is the correct way to write this” or “say this.”

If they ask me a question, I reply to them using almost the same sentence structure but I use the correct words, I use the correct grammar. And hopefully they picked that up without me telling them. And they do pick it up. After a few times they do pick it up, like “oh this is the way the professor writes, so maybe I should follow that.” And I think this is a very effective way of informal guidance instead of formal teaching.

But for students to learn that way, it requires one thing: Frequent use. It only works with frequency. If the frequency is low, if they communicate very rarely, very seldom, then the benefit is minimal. They don’t get it, because it takes repetition, it takes reinforcement. I do sometimes explicitly point out that this is not the right word, this is the right word. But I would say that probably just takes up 10% of our communication. I am not an English teacher, I am not correcting them every time.


Well that’s very valuable for the university, your teaching via software. You’ve also demonstrated your way to teach, to encourage them to speak, to communicate in English. And I know there are also other faculty members here, who tried to enhance their international connections or internationalization or connect students to overseas academic development. We are pursuing internationalization, keen to recruit a more diversified and international faculty.


I think you hit the nail on its head. It’s diversity. I think, for any university, especially ours, to make progress in this new age of education is to embrace diversity. I hope in doing my job, I can be a small part of that diversification. As to your question of internationalization, this can take place at various levels. At the basic level, through faculty members like myself, we can tell students about our experience outside, we can bring in best practices from around the world, and that is important.

An important aspect of my background was shaped by how Singapore is a very small country. Being very small has its advantages, meaning that you are exposed to the whole world. When you’re small you tend not to close up, because there is nothing much to close up on. And in that way, we are very much exposed to the international way of doing things. And through that, our key phrase is best practice. That means you try to gain and learn from experience from all over the world and not be too hung up on your own approaches. So I think for a university, that is true also.  And that can happen not only through a diversified faculty, but also a diversified student body.

I think a university cannot said to be internationalized or equipped with a global outlook unless it has a diversified student body. And by that I mean students from different cultural and language backgrounds. Because even if we get students from say Hong Kong and Macau, or even Malaysia, if their background is rooted in the Chinese language, it’s still not diversified in a certain extent. And if we want to diversify, meaning that we are able to attract a diversified body of students, then one thing is necessary. Which means that our courses has to be accessible to international students, and that means it has to be taught in English.

And not only that, also the way the administration works, the paperwork, the method of registration for international students from different backgrounds. And for that to be accessible, the university administration itself has to be internationalized.

Focusing on individual courses being taught in English, that is the easy part. That is in fact, the easiest part. Okay, pick up a textbook, teach the course in English. All our professors are capable of doing that. But the entire environment and the entire mode and mentality of operations have to be international, too. And I am not saying we have to work like the British or the Americans, in fact there was a saying about the universities in Singapore. They are not western, they are western-style universities. Well, I even want to say they are international-style universities.


Is this your core value?


This core value, I think we develop it by looking at how people work overseas. The style is international. I was educated in France and when I was there, the style of instructions was very much focused on French students only. And I see the parallel here, in Taiwan. When I went to America, it is not American style. It is an international style. Style that can attract, and that is accessible to students from all over the world. I hope we can put our heads together and put the effort in to work towards that goal for our university.


NCNU is situated at the initial stage of internationalization.


We are at the initial stage, a little late, but better late than never. Operations count. To coin a rather mundane word — the business model. The business model of the university like ours today has to be a global model. And I think help is on our side, not just from a few of us international faculties, but because we have examples in front of us. We can look at how other universities in our neighboring countries, how they internationalize themselves. In fact, the two main universities in Singapore became international only in the 1990s, which is not that early. And you can trace the steps they took to get there.

I think it’s important for us not to reinvent the wheel. To look at how people did it before us and there are plenty of examples. Even schools in Malaysia, Vietnam, Japan, and Korea. How they did it, some are more successful, some are not successful. And these are learning models for us. So, starting late has an advantage. You can learn from the others’ mistakes, as well as the others’ cleverness. And I always go back to this thing that I call best practice.

We may have our own creativity here and there. Maybe unique to the Taiwanese context. That will help too. But there are plenty of best practices to learn from around the world.


And your life beyond your academic activities? Because, Taiwan is home already? Second home?


Um, the campus is my second home? I only live on campus in Taiwan with my wife. Otherwise my home is in Singapore, but I don’t go back that often. Outside of school and research, I’m a Christian. So I read the Bible; incidentally I’m on Sabbatical leave, so I’ve taken the opportunity to sort of relearn guitar after a long, long time. And recently I have fallen in love with bread making. So I’ve been making bread at home. And with my wife, who is an artist. When she allows it, I get involved in her art. Maybe sometimes it’s just opening my mouth, because I cannot paint. But I like art.

So, yeah, it’s not a jet-setting, star-spangled life. Quiet and pleasant life here...


You sound very versatile. Is it difficult to find counterparts or colleagues who share the same interests?


Well it is always difficult to find someone who would share all your interests. That’s impossible. Yeah I do find some colleagues who share part of my interest and not the other part. I wouldn’t say I’m versatile, I’m not 100% good in everything I do, but I like challenges. I don’t want to settle in a comfort zone and that’s the same mindset for taking up bread-making. I like challenges, and I believe that nothing is too difficult if you put your mind to it. And once you start you will reach a certain level, it may be high, it may be lower than expected, but it’s always something, right? So that’s why I pick things up this way, with this mentality, instead of doing the same things that I’m comfortable with.


Can we also talk a bit about different higher-education institution systems?


I’m not sure if you are familiar with the history of Singapore’s education. In the Singaporean context, because we inherited a lot of management styles from the British, the ways we operated were more or less British. But I wouldn’t call that global or internationalized. Until the 90s, the focus finally shifted to not just producing graduates for the Singaporean industries, but to producing graduates that are suitable for the entire world.

So there were ramped-up efforts in attracting international faculties and a huge body of international students. So in terms of that, I think no universities in Taiwan can match the internationalization of Singaporean universities simply for one criterion, and that is the proportion of international faculty and students among the entire population.

I would say in the faculty population, more than 70% are international. And in the student population, I think the official figure is more than 35%. So you can see how diversified it is. Once you diversify the population, your style of working, the way you conduct classes, the way you interact with other universities, with the research world, that will also change. I think our university in particular, and as well as most Taiwanese universities, have barely started internationalization.

We are still in the traditional mode of operations. It’s not just a matter of language, not just about conducting classes in English. I think we have a long way to go before becoming truly global. One of the criteria in the Times Higher Education World University Rankings — global outlook — accounts for 7.5% of the total score. The criterion is stated simply in terms of one’s ability to attract faculties and students from all over the world.


So what do you think are the most important criteria for retaining or appealing to international faculty?


Whether it’s for faculty or for student, a school that has international appeal is a school that makes the person feel that he or she doesn’t have to fit into a mold that is alien, unfamiliar. Of course, language is one of the key things, because most faculties members are from other parts of the world, they have the command of the English language. So if the language of instruction and language of administration is English, then the new faculty members wouldn’t have to learn to fit into a new mold.

Secondly, the mindsets of most global universities are independent. They might be guided by government policy, but they operate independently. They have their independent approach to, for example, curriculum design. They have their independent approach to rewarding faculties, rewarding students, or they have their independent approach to establishing international collaboration either in research or in curriculum.

So I think what might be preventing more Taiwanese universities from internationalizing is that we are more bound by a single set of policies, mandated by the government, than universities you might find elsewhere. Take, for example, schools that have their mindset on internationalization often maintain a certain percentage of an international faculty staff. Because international staff population is high, that forces the administrators to operate in a way that encompasses practices that all international staff can recognize, “oh, this is the standard international way of administrating university.”

Whereas, in a Taiwanese university like ours, we operate based on mainly Taiwanese laws. Whether its by university or government laws, most of those laws are very different compared to internationally recognized practices. So in a way, for a faculty coming in, he or she has to understand, what is the rationale for this law. Why does the government or why does the university endorse this or that policy. Which to them, might be a little bit surprising. It’s not always very different, but these are small little things that might be surprising to newcomers to Taiwan.


Do you mind giving us an example?


One example is the student quota policy here. I understand the rationale behind it, as it would be bad for the university, bad for the program if we recruit too many students and there are not enough professors to teach those students; the teaching quality might suffer. But I don’t think we are at that point yet, our student-to-staff ratio is still low compared to other places.

But what puzzles me the most is that this quota is hard and fast. That means you cannot have any flexibility in the quota, depending on maybe this year’s intake, depending on opportunities available, depending on the willingness of staff, level of energy; there can be some level of flexibility but we are very hard and fast to reject any alternatives.

Sometimes it has a negative effect, because if students can only apply to one school, they will apply to the one that has higher quota. So when you announce your small quota, it’s self-defeating. You are scaring people away. So I find that is rather rigid in a way, because to me, a university is after all a business of education. It’s a business. You have to operate it, look at the bottom line, invest in it for the longer future. And it takes effort, investment can be painful, tiring.

But if you do not invest in it, you will remain bounded by law, by government policies that are dictated by an upper limit instead of a lower limit. So that accelerates a tendency what we call “a race to the bottom.” We just keep going, you know, downward. Avoiding that upper limit. I think it's self-defeating. 

That is the weakness of the system here. Fighting over the limited number of local students doesn't create more students, alright? Unless the birthrate just dramatically increases, that will not work. So if all these quota systems is just for universities to have a fair share of the number of students, it promotes mediocrity. In the sense that, "okay that's my share" whatever happens you got to give me my share.

If our university has a reason to continue and exist, we have to develop a way to recruit from outside the local population. We have to improve our international appeal, to students who don't speak Chinese, to students who want an education, who are not seeking to be in the top or famous schools, but seeking an unique and affordable education. And I think we can offer some unique experience here. But the first thing is to boost our overall appeal. The unique experience is something like icing on the cake. But first, the cake has to rise.

The challenge is actually the substance of the entire program, the entire business model, has to up take an international outlook. I recognize it's difficult, even for Singapore. I told you the schools right from the start, they were already teaching in English. Took them 15 years, and language was never an obstacle. So we have double jeopardy. We have language as an obstacle, but we also have the administration style, the business model, which so far is still not in the right place yet.

I think that is a double challenge that we are facing. 


So, private universities enjoy more freedom to run their businesses with different models, compared to national universities?


The very height of Singapore's efforts to internationalize its education system, and that was not limited to universities, was in the year 2000. In fact, Singapore's secondary education is also much internationalized.  At that point, the government decided that the universities should be autonomous. Very few things are autonomous in Singapore, but the universities were autonomous from then on, giving them a greater flexibility of how they want to design their business, their curriculum.

And if you go to Singapore and visit the main universities, National University of Singapore, Nanyang Technological University, and Singapore Management University, you'll find that they are like universities from 3 different places.

They are very different, even in a place smaller than Taipei, these universities operate totally differently. The curriculum has a lot of differences, and the students they attract are also very different.


It is oriented differently.


Yeah, oriented differently, and that was by design. Because in the 1990s, the Singapore government decided that these universities should be given their own models, and they are now very different in terms of student population. It's interesting. As I said, we shouldn’t despair because we have many examples in front of us. And what is important for us is to learn from them; maybe visiting is not enough, but to really study the trade from them.

I think most faculty members are parents and it's easy for them to realize, just to put our feet in the shoes of the youngsters and think, why would a young person of 19 or 20 years of age choose this place to study versus that place to study? They don't know much about research, they barely even know the subject.

Part of academic branding is to use your research. But the real point is that whether by spending 4 or 6 years here, he or she becomes employable. And I think that is a big yardstick for students. If the degree he or she obtains here is of value, to their home country. So we have to understand the national context. Does a degree from NCNU earn them a good job there?

Of course, it's a yes it will earn you a good job. It earns you a good job only when there are people before you. People before you who have graduated from the same university and are doing well on the job. So, I think we have to focus on how ready our graduates are once they leave school at the end of their 4 years of undergraduate studies. How ready they are for the job market in an international context.

And I think that reveals a big flaw, because in the Taiwanese model, undergraduates don't go to work when they leave school, at least for engineering. They go on to study for their master's. But to an international student, he or she needs to be ready. So our program will have to make everybody ready, whether he or she is a Taiwanese student who’s preparing for studying at the master's level or an international student who's going home to seek a job. So we have to ready them. And our current program doesn't ready them. It's just purely academic, you know, expecting that they'll go on for masters’. And that is why we don't have an appeal in that sense.


Industrial-academic collaborations are perhaps a way out.


Yes, that is a must. And we have to invest a lot more both financially and teaching-wise, in practical training. That is, every course or every module has to have a significant portion of practical training. And practical training means the students will be operating machinery that is being used in the current industry, learning where the buttons are, what the software is. I think that is very important, because we can say with confidence to potential students and say, this is why you should choose us.

We have to think of their future. Even if two universities offer the same prospects in terms of careers, why would a student choose A and not B? Well I think a lot of that has to do with still wanting a good learning experience. They want to know if they will have a good time during the 4 years here. And that is very important.

Although it's cliché, but you have to somehow treat your students as customers. Not as much as pleasing, but you have to let him or her have a good experience while here. Then they remember the school, the teachers who taught them, the good memories. The classes they have taken, where they have sat in the class and stuff like that. And it's only with this legacy, and if the students or the graduates do well later on, will the reputation of the school improve.

I still remember fondly most of the schools I have been to. And I think it is important to build a strong alumni. An alumni network that remembers the school well, who have fond memories of the school. Yeah, they still have some professors that they don't like, that they criticize, but we have to keep an overall appeal for our alumni who remember this as where they have spent the best part of their youth. That is the mindset we have to migrate towards instead of just saying, "okay you are a student here, you must do this, this is our law," right? It's really moving from the concept of civil service to public service. Providing service to the public population.

It's about management, because there might be a misconception that academic freedom takes precedence over anything else, but I would say 100% academic freedom means it’s not a school. It's just individual teachers renting the room, do their own classes. Because a university is still a collective, and people have to be team players. But to both academic and administrative staff, the management needs to project a vision. What is the vision for this organization? Most of us would spend a good part of our career here, so what is the vision for the next 10, 20 years. That has to be pronounced and that has to be upheld. Renewed, if necessary.

And the reward system is also part of the management. Right now we are rewarded according to the national standard. And not according to the organizational vision. Take, for example, if develop our curriculum by prioritizing our teaching over our research. But the reward system here is to first reward professors who’ve done a lot of research, never mind he or she teaches poorly. And therefore, there is a lack of an internal authority that governs the university. So the individual member will not look towards the university as where they belong; they look towards external authorities as the one they have to answer to. And most of the time, these two are not compatible.

And right now, for NCNU, we are at a crossroads, because the Taiwanese student population is dwindling, and there is no sign of it picking up. So the time for action is now.

I was just looking at a university in Malaysia. It is not a university that's very well known, but it looks like a very international university with a lot of appeal; all the courses are in English, and that is quite rare in Malaysia. And they have curricular collaboration with some UK universities. Now, the standing of that university is not very high. But they survive. And it seems like they have quite a certain appeal. When did they start doing that? Year 2000. So it's been 19 years they've working towards this goal.

So as I’ve said before, it's not something that we can achieve overnight. I'm just hoping that we recognize this and if we all think and agree that NCNU has a reason to operate the way it operates now, all the activities as they are unfolding, then we must really take action. We must really strive to see ourselves, what we'd be like in 20 years. There has to be some vision in education.

Those who are interested in learning more about the professor and his field can contact Professor Lum at

Share To: