This week I was reading the results of the annual Developer Survey from the excellent tech forum Stack Overflow (available here).
I was struck by a chart about educational background, excerpted below.
The top line gathers three different streams of technology education into one: Computer Science, Computer Engineering, and Software Engineering. There are profound similarities between the three, which is why they are grouped, yet also significant differences, in their approaches, assumptions and end goals.
What struck me was the realization that, despite my strong formation in the Computer Science stream, over time I lean more and more to the Software Engineering stream now.
I was also struck, to a lesser extent, on how low the Math/Stats undergrad was in the chart. Anecdotally, at least, so many of my bosses over the years have been from a Math background, but maybe it was the path of an earlier generation.
But back to CompSci vs Software Engineering.
I finished both a Bachelor's degree and a Master's degree in Computer Science. Looking back, I would say that the education I received from this stream tended toward the more theoretical. It drilled deeply into the mathematical, physical and digital underpinnings of computer technology. It explored complexity and efficiency (at least in my Algorithms-centered specialization). It tackled hard problems with a goal of pushing the limits of our knowledge.
To my CompSci eyes, the Software Engineering stream is less focused on pushing the bounds of knowledge and more about formalizing and productizing what is learned and discovered, making it marketable and maintainable.
Does that sound crass? That's not my intention. I now find Software Engineering more compelling and interesting, peronally, than CompSci.
Now, after more than a dozen years in a variety of IT jobs, I find that I think more like a Software Engineer than a Computer Scientist. Is this a natural evolution given the reality of building software for a living rather than researching in an academic setting? No doubt that is one influence.
Looking back to my CompSci Master's thesis, in which I studied the effects of some theories as applied in a Software Engineering context, I can see that this new mindset is not so new. It has hints stretching back to those days, rather than just being an adjustment to make it easier to do my day-to-day work.
Let me give an illustration of some of the differences I see between Software Engineering and Computer Science.
Imagine we want to tackle creating a Sudoku board generator. Sudoku is a number-puzzle in which a 9x9 grid of numbers must be filled in such that the digits 1 to 9 appear exactly once in each row, in each column, and in each of 9 sub-squares.
My CompSci side would be interested in writing the algorithm at the heart of the problem. It would focus on questions of data structures, of loops and recursions, of back-tracking when a conflict is generated. And then that side of me would want to make it as efficient as possible. It would perform a Big-O analysis. It might consider hardware improvements. It might draw on the latest human-computer interaction theory in designing user input strategies.
My Software Engineering side would not reject any of that as unimportant. But it would approach the problem with the long-term testability and maintainability of the design and architecture in mind. It would create abstractions of the constraints or rules behind a Sudoku board. It would encapsulate data and actions into appropriate sub-modules. It would make decisions about the structure of the application as much as the structure of the algorithms under the hood.
It's a question of the priorities and emphases of each discipline.
If you are looking into which educational program is right for you, get the real and specific differences in each stream from the schools you are considering. For example, the excellent University of Waterloo describes what each flavor means to them here.
I was struck by a chart about educational background, excerpted below.
The top line gathers three different streams of technology education into one: Computer Science, Computer Engineering, and Software Engineering. There are profound similarities between the three, which is why they are grouped, yet also significant differences, in their approaches, assumptions and end goals.
What struck me was the realization that, despite my strong formation in the Computer Science stream, over time I lean more and more to the Software Engineering stream now.
I was also struck, to a lesser extent, on how low the Math/Stats undergrad was in the chart. Anecdotally, at least, so many of my bosses over the years have been from a Math background, but maybe it was the path of an earlier generation.
But back to CompSci vs Software Engineering.
I finished both a Bachelor's degree and a Master's degree in Computer Science. Looking back, I would say that the education I received from this stream tended toward the more theoretical. It drilled deeply into the mathematical, physical and digital underpinnings of computer technology. It explored complexity and efficiency (at least in my Algorithms-centered specialization). It tackled hard problems with a goal of pushing the limits of our knowledge.
To my CompSci eyes, the Software Engineering stream is less focused on pushing the bounds of knowledge and more about formalizing and productizing what is learned and discovered, making it marketable and maintainable.
Does that sound crass? That's not my intention. I now find Software Engineering more compelling and interesting, peronally, than CompSci.
Now, after more than a dozen years in a variety of IT jobs, I find that I think more like a Software Engineer than a Computer Scientist. Is this a natural evolution given the reality of building software for a living rather than researching in an academic setting? No doubt that is one influence.
Looking back to my CompSci Master's thesis, in which I studied the effects of some theories as applied in a Software Engineering context, I can see that this new mindset is not so new. It has hints stretching back to those days, rather than just being an adjustment to make it easier to do my day-to-day work.
Let me give an illustration of some of the differences I see between Software Engineering and Computer Science.
Imagine we want to tackle creating a Sudoku board generator. Sudoku is a number-puzzle in which a 9x9 grid of numbers must be filled in such that the digits 1 to 9 appear exactly once in each row, in each column, and in each of 9 sub-squares.
My CompSci side would be interested in writing the algorithm at the heart of the problem. It would focus on questions of data structures, of loops and recursions, of back-tracking when a conflict is generated. And then that side of me would want to make it as efficient as possible. It would perform a Big-O analysis. It might consider hardware improvements. It might draw on the latest human-computer interaction theory in designing user input strategies.
My Software Engineering side would not reject any of that as unimportant. But it would approach the problem with the long-term testability and maintainability of the design and architecture in mind. It would create abstractions of the constraints or rules behind a Sudoku board. It would encapsulate data and actions into appropriate sub-modules. It would make decisions about the structure of the application as much as the structure of the algorithms under the hood.
It's a question of the priorities and emphases of each discipline.
If you are looking into which educational program is right for you, get the real and specific differences in each stream from the schools you are considering. For example, the excellent University of Waterloo describes what each flavor means to them here.