03.06.08

Ailing Education

Posted in College, Computer Science, School, Thoughts at 5:50 pm by Nick

In recent times it has been very hard for me to attend class and get things done. This is due mainly to a troubling and complex family and living situation. Thus, I have decided to withdraw from two of my classes and work harder towards not marring up my academic record then I have already. Along with this, I am going to be moving during spring break.

However, I must say I am absolutely unimpressed (and disappointed) with some parts of my education. I know it may feel like an attack, but it should probably be heard. It is no news to the Computer Science program at UNI that enrollment is down and there are many people who switch majors. After talking to -many- CS majors (mostly freshman), I have decided it has little to do with the course content. If I could choose, there would be many things I would change based on my personal experience and feedback.

The first thing I would change would be our introductory courses. I know this is talked about and debated a lot. But what we have taught recently, Ada, C/C++, and Java, is a terrible choice. It is no reason our majors are running away! None of these languages are terribly user friendly for beginners. To exacerbate the problems, we thrust our students in without ever telling them what it will be used for, how it will be used, or why we use those particular languages. We don’t mention other alternatives. Why is this? Why do we not teach introductory classes with dynamic typing, like PHP or Ruby. I have never heard more complaints then about students who have no idea what the different between certain data types are. In C/C++ there are all kinds of exceptions, like using char’s as small integers, unsigned and signed, and C-style strings vs. C++ string data type. How are the students supposed to figure that all out? Ada is a language designed initially for embedded systems by the Department of Defense. That line alone is enough to scare freshman. There is nothing simple in these languages. So why do we teach them in our introductory courses? Our degree program already requires an extra Comp Sci general class. Why not save these industrial languages for them?

Students need to learn the simplest things first. These languages were not designed to be simple. Yes, if you know how to use them they are not as complicated as they first seem. But there is no reason to think a freshman in computer science, with no previous programming background, will understand the concept of passing the address of data to a function. I know several good programmers who still do not completely understand pointers. To teach the simplest thing possible, we need the language with the least amount of baggage. Something you can get using quickly, and without a lot of quirks. In C++, you need includes, a main function, knowledge of data types and operators before you can do anything. In dynamic typed interpreted language, you need basic operator knowledge, and that is all.

And yet I fear we cannot change our introductory languages because we have our two full professors teaching our introductory courses. Never mind the fact that the C/C++ students learn C, not C++. C++ comes later and is merely brushed on. I understand better then most the overlap between the languages, but there are some huge differences. I think students in the C++ section should be taught about C-style strings (that is, character arrays) but they should be taught about the more commonly used (and far more useful) STL string data type.

I strongly believe our faculty grew up when computer science was brand new. They all have fantastic and interesting stories. But this age is different. Our upcoming CS students are different. The times have changed. Computer science is no longer ‘new’, it is now ‘young’. The students will not spend all of their nights up late, huddled over their computer attempting to fix a broken program. That type of dedication is gone. It is the zeal of the new and unexplored. While there is plenty to explore in CS still, freshman will no longer be the ones doing it. We have a new computer science generation, and our program needs to realize this. We will continue to see low enrollment and a high drop rate if something doesn’t change. You cannot scare away all of the students by making them feel dumb and inferior because they cannot remember complicated syntax and expect your academic program to live.

Our introductory courses need to teach different things. Never, in any class I have taken at UNI, has a professor started out a class by mentioning why we should learn what we are learning. They have never mentioned it’s use or importance. In our introductory courses we jumped right in, delving into logical algebra and programming methods, but no one said what we needed to learn the algebra or methods for. I was lucky enough to know how to apply my skills already. Most students don’t. They have no drive because they are merely frustrated by everything they are learning. And who wants to walk up to a professor and admit they have no idea how to compute square roots using for loops? No one. And that is a big problem. I think we should promote a different way of learning. Students find it extremely hard to take program specifications and write a program from scratch. It is hard, and advanced programmers do it. Not new ones. We should develop programs for our freshman to modify, so they can see why code works and how their changes will reflect it. We should promote programming in small steps. If you write a program from scratch, it won’t work the first time. It never does. Maybe because it was late and you initialized your data wrong, or perhaps you thought through your packet filling the wrong way. Whatever the cause, if a student writes it from scratch and it doesn’t work, it is a moral failure. They aren’t used to it. They don’t understand that even the best programmers don’t write bug-free code.

Our students are not motivated. As I listen to classes, people, and friends talk about computer science, they are NEVER talking about what cool project they are working on. They don’t talk about things they are looking into, nor are they talking about things they aspire to make. They talk about how they failed to complete the homework because they had no idea where to start. They don’t have the time to work on something they are actually interested in because they spent four hours trying to fix a program that was probably never going to work the way they wrote it in the first place.

There are two things that bother me the most:

One, the fact that our major is long and requires so much math. Yes, I understand math is great. Yes, I understand we have a science degree and it will not be short because it requires training. But our degree provides no flexibility. If you want to reasonably complete it, you will work your ass off or take a more sane five year plan. I’m still unsure what is with the emphasis on math. I understand it -was- the foundation of our program. And I know it can help understanding, or expand thinking. But seriously, the last math class I technically completed was Geometry. My family life went to hell during Advanced Algebra and Trig, so I never completed the second semester, and I got a C in the first. I am not ashamed to admit this. It proves you do not -NEED- math to succeed in computer science. In fact, apart from simple arithmetic, math is largely unneeded unless you are working on math related problems. Discrete structures is a great class and very applicable. But you don’t need calculus to do it. I got an A- in it, merely missing the A by a percent or two. As a software engineer I’m pretty sure you will rarely deal with real math problems. So what is the math portion of our degree really for? Why not dump it and strongly encourage a mathematics minors? If you do the courses just right, our major is only one or two classes away from a math minor. But, of course, there is no real time to take those classes to get it if you want to graduate in four years.

And two, you never have time to work on projects that matter to you unless you are completely uninvolved on campus and jobless. Most (not all) professors are not flexible enough to allow freedom in their homework. Rarely can I think of times when I could pick homework that was interesting to me. I have gotten more then most, that is for sure. In our introductory classes, students may not know what interests them programming wise. But they certainly will not be interested in mundane examples. Not all students will want to write stuff on their own. But we should encourage them to pick homework that will relate to the topic at hand and help them do it. Everyone knows and can enjoy tic-tac-toe, and it is not hard to program, even for beginners. It demonstrates many programming techniques, such as input/ouput, array manipulation, or potentially AI or even OOD. Yet never has gaming been brought up in any of my classes, except for AI where game playing is part of the curriculum. In fact, in many of my classes, gaming is frowned upon. Never mind the fact that gaming is perhaps the easiest way to relate the material to the current students, and demonstrate any technique, be it lists, trees, arrays, classes, or OOD.

This all bugs me to the point of wanting to switch majors. I won’t, but it seems like a damn good idea. Economics is simple. They take the core of what you need and give it in eighteen hours. Then you choose eighteen more hours of education, putting what you want to learn in your hands. This gives you room in your degree to deal with bad classes, or explore things you are interested in. While I’m not saying computer science can be simplified to thirty six credit hours, I think the BS could provide far more flexibility then it does. Since we do not offer Software Engineering, nor anything else as a major, there is very little point in requiring students to pick an emphasis beyond Foundation courses. Let students take what they are interested in. They will look for jobs they are interested in, and then they will know more of what they need to know for their jobs. I have no real special interest in Information Science, it just seemed like the lesser of all of the evils that are our choices. I wanted to learn about databases, but none of the other courses seem interesting. I think the foundation courses are a good idea, but they should be expanded. Perhaps take a primary course from each section (Info Storage and Ret, Software Engineering, Networking) and put it in the foundations, then remove the emphasis requirements. This would require students to fulfill a requirement and pursue whatever path they would like without punishing them if they decide to change. Maybe students want experience in Real Time Embedded Systems, but have no interest in testing or project management. Currently, students will probably not pursue their desire to learn about real time embedded systems because they will be devoting their emphasis to something they only have one real interest in.

Note: This IS a first draft and WILL be revised.

Permalink