Engineering Salaries Are on the Rise in the United States

From Luehrmann’s paper:

Or should his education also include learning to use the computer (1) to get information in the social sciences from a large database inquiry system, or (2) to simulate an ecological system, or (3) to solve problems by using algorithms, or (4) to acquire laboratory data and analyze it, or (5) to represent textual information for editing and analysis, or (6) to represent musical information for analysis, or (7) to create and process graphical information? These uses of computers in education cause students to become masters of computing, not merely its subjects.

Although Luehrmann may have intended that to mean programming (he was contrasting with computer-aided instruction, which is roughly the equivalent of the Khan Academy today), a lot of what he lists there fits under “using technology” rather than “programming”, because the tools have gotten more user-friendly and a lot of the tasks he lists have pre-written tools for them.

Computers, more so than anything else we have available, provide the means for developing our ability to think logically.  We’ve always used toy problems in mathematics to develop reasoning skills, but computers serve to take that one step further, since a computer’s logic definitively uncovers the flaws in one’s reasoning, especially where boundary conditions are involved.  That is the upside of programming.   A computer is the ultimate task master, exposing simple errors of reason that are difficult to spot otherwise.

Computers are not quite so effective when shades of gray are involved, but even there I think it would be possible to use a computer simulation package like Extend effectively with non-majors, especially to help them grasp the nature of statistics (Extend is a simulation tool which uses a drag and drop approach for building discrete event and/or continuous simulation models to whatever level of complexity one can handle, and for the power user, provides the means for building one’s own model blocks).  If I was still in the throes of my academic career, I would be greatly tempted to build a general course around Extend or something like it, since that would address both the logical nature of programming and using computers for decision making under conditions of uncertainty (and which just might enable more students to cross the barrier statistics poses for many of them).

One could go further in the direction Professor Talbert suggests, and say that introductory complexity theory and basic boolean logic are relevant things to be learning as an undergraduate requirement.  This has the nice effect that the primary course material is not dependent on a particular computer language or software package that may be outdated in four years for an undergraduate student and will definitely be outdated in eight years for a high school student.

Most programming that people do involves basically two constructs – loops and conditional statements.  In addition to these constructs, for object-oriented programmers, the information encapsulation of the class structure is also important.  The reason I think a lot of people have been and still are able to just ‘start programming’ and make careers of it (i.e. without an educational background in CS), is because they’re able to recognize that the truly new concepts that one needs to learn aren’t that many (I count three).  The rest is learning syntax, which you will have to do again and again no matter what you learn in undergrad because of the rate at which the ‘language du jour’ changes.

I have a friend in my university who was an undergraduate engineer.  She was explaining to me how programming was used in her classwork but also explained that she never learned loops. This shows a perverse prioritization of learning, highlighted by the above article’s mention of emphasizing ‘technology’ over an understanding of the logic that makes computer programs work.  My young friend’s knowledge of the syntax of some of the popular engineering softwares such as Matlab, may be helpful to her to put on her resume but the utility will disappear as soon as Matlab releases another update or if the company she works for uses say, Maple, instead of Matlab.

What my university would have done well to teach her instead would be three basic concepts (loops, conditionals, classes) and enough practice in a programming language that would allow her to apply these concepts again and again and to nest them as needed. As an engineer you would have thought that someone would teach her these basic things but instead the curriculum is geared to learning some ‘sexy’ features of a language that may or may not be useful the day after tomorrow.