Course Introduction 2.71 Optics, Fall 2004

GEORGE BARBASTATHIS: So 2.71/2.710 is an introductory optics course. The idea is that the students do not know anything or very little about optics to begin with. And so the goal is that by the end of the course, they will be able to not only understand the very basics of light propagation, but also they will be able to understand the systems that utilize light for things like imaging or telecommunications or a number of different engineering applications, and they will also be able to design systems of their own-- perhaps not very complicated systems, but definitely systems that can do something useful. So by definition, then, it means that it is oriented for students who are not necessarily into the physics or electrical engineering disciplines, where, normally, you get a lot of background in optics even early on, and the course has a lot of this orientation of starting with the very basics and then building upon them. So the general pedagogy is guided by that. A lot of the basic tools have to be taught from scratch. But also I think that helps because the students actually pick up a lot of their material. For example, that Fourier transforms, which is a mathematical technique, they actually pick it up in the context of optics, and that really-- from students and their advisors' reactions, I think this is something that they actually appreciate. And in terms of judging performance, obviously, you have to take that into account, that they are actually building the background there. But by the end of the class, they actually complete a project, at least the graduate students do, and they typically do very, very well. So my impression is that from that point of view, the course succeeds in its goals. So I came to MIT about five years ago, and before that, a colleague who is a senior lecturer, his name is Steve [? Anton, ?] he taught the version of this course three times, and so this was really the first time that our department offered an optics course, and it was wildly successful. He got very good attendance and very good reviews. But Steve actually runs a company outside MIT. He's not a full-time employee. So MIT decided this was a good direction for the department to go. So when I came here, one of my missions was actually to create that sequence. And substantially, why mechanical engineering needs that course is because optics is fundamental in many of the core mechanical engineering disciplines-- for example, measurement, instrumentation, controls. Imaging is a very important part, very often, of mechanical engineering for things such as robotics. And of course, these are all interdisciplinary subjects, but to a very large degree, they belong to MechE. Now, in addition to that, the course, of course, is open to all MIT these students. In fact, I do get students from electrical engineering and physics occasionally but also from other disciplines. I've had students from [INAUDIBLE] engineering, from AeroAstro, even architecture, actually, in my class. And of course, typically, these are graduate students for whom some aspect of optics typical imaging becomes important in their research. And then the course serves them very well because it gives them the tools and the fundamentals to use it, but it does not require they have a physics background that is typically to be expected in an optics class. So I have a slightly unusual schedule in the class. One of-- there's two lectures per week. One of them lasts for two hours, and the other lasts for one hour. So I tried to design the sequence of topics so that in the two-hour session, we cover a lot of the fundamental material and a lot of, so to speak, the heavy stuff, if I can use the term. And then in the one-hour session, we actually solve problems or we look into some more detail into applications, and so on. So the two-hour session sometimes ends up being a little bit difficult to carry out, both for me and the students, because it requires, you know, it requires concentrated attention for a long time. But I think it pays off because with a proper structure, typically, we break once or twice for a few minutes, and then I try to interject jokes, and so on, during the lecture, so the students sort of-- they recover their attention as opposed to falling asleep. And by the end of the two-hour lecture, they have received the material, which then they see again in a more practical form in the one hour, so I think it reinforces quite a bit. So the class has the typical requirements. The students have to do homework, they have to take exams. And in addition, the graduate students, they have to carry out the project. So let me start by saying why I do not require a project from the undergraduates. The reasons are purely logistical. I tend to get a lot of seniors and juniors from our department, and both of them, near the end of the semester when the project would have to be carried out, they're busy by other projects in other classes, so it would really become a huge burden on them. Graduate students typically do not have that limitation, so it is OK. What I find is that at the end, when the graduate students present their projects, then the graduates, they typically come into that session and attend, so they are definitely interested. And occasionally, I had one or two undergraduates who got involved in projects, but it doesn't happen so often. So what the project is about is actually very simple. Typically, I give the students a few papers from the literature, typically three or four papers. What they have to do is prepare a presentation, almost a conference-style presentation-- it is about 15, 20 minutes talk-- as if they were presenting their own work at the conference. But of course, it is not their own work. It is what they read in the papers I gave them. And typically, I assign them a mentor. It could be me, it could be the teaching assistant, or some of my research students from my research group. And then the mentors, I instruct them to try to go a little bit beyond the papers. Not too far, of course, because is a class project, not a research project. But typically, before the students present, they have to do some calculation or some additional design based on what they saw in the papers. And I think that works very well because it really forces them to understand a little bit deeper than the traditional lecture presentation. And that lasts for about a month, actually, and at the end, they give a presentation. Of course, this is a university, so we buy free pizza for them and all that stuff, and it's a lot of fun. Usually, the projects are-- the comments I get said the projects are the one thing that they enjoy a lot about the class.