Moebius Stripper, always able to supply a nice rant about how integrated calculators in math classes are producing students unprepared for college math, supplies a good set of links to articles about the connections between Texas Instruments and the math textbook industry.

For some reason, these articles got me thinking about a related question – how much computer usage in school is a good thing, and how much distracts from real learning? That’s a huge can of worms, so how about just the question of the correct balance of computer usage in my department’s classes?

It obviously varies from class to class. But I think about my programming class as a good example of where *some* computer usage is certainly necessary. One *can* learn to program without a computer (ask Babbage and Lovelace…) but it doesn’t seem necessary to argue that students should actually get to write, compile, and run their programs on a computer.

But how about the classroom experience? Previously, my experience with introductory programming classes was that they were taught in a chalk-and-blackboard setting, perhaps with the insttructor being able to project their own laptop screen at the front of the room, and students got practice writing programs on theiir own time, or perhaps in a supplementary lab section outside of class. I was very excited that in my current position I can have all of my students actually work at a computer during class with the concepts we are learning. It seemed, on the surface, to be much more effective to introduce a programming concept and then immediately allow students to work with that concept, instead of having to discover a day or week later when they went to a lab that they hadn’t quite understood all of the details.

Now, so far, this is no different from saying that students in a math class should get to actually work some problems and get immediate feedback in class – not really analogous to the calculator concerns. But having now taught a programming class in a fully computerized classroom, I’m reflecting on whether there were advantages to the old technique whereby students did more paper-and-pencil work. I wonder if it allowed students to initially focus more on the concepts being introduced, rather than the syntax, because they were not immediately faced with compilers and syntax errors. When exams are written with paper and pencil instead of on the computer, it is more evident that you are testing problem-solving skills, because one can’t expect character-perfect code of any signficant length written out longhand.

My concern about calculators in the math classroom is that it prevents students from being familiar with basic concepts and manipulations that they ought to be able to do on their own – that they replace their deep understanding of the math with a memorization of what the right buttons to push are. In a programming classroom, does constant access to a compiler, an editor wth syntax highlighting, and collections of past code from which to copy and paste end up encouraging students to take a trial-and-error approach to programming, instead of forcing them to evaluate based on their own understanding what the most effective appproach to solving the problem will be?

In the one programming class I took—about six years ago—I remember being a little intimidated but fond nevertheless of pseudo-coding. We had computers in the class, so there was that; but we were also required to at least provide a token effort to pseudo-code our functions.

While there isn’t a direct analogue of pseudo-code in math, working problems by hand and mocking up code share an intimacy with the problem rather than an intimacy with the solution. It is an understanding of problems that allows us to achieve—a solution has only immediate and local applicability, but a given class of problem can be found in nearly countless contexts.