I will suggest a less harsh approach to the topic than the author...
Traditionally, physicists and mathematicians develop the technology, engineers apply the technology by designing things for commercial applications, and technicians try to build the design and make corrections based on experience.
Consider a large building: architects, structural and architectural engineers, and trades people.
In software, there used to be a formal design phase. The computer hardware and operating system was not a given (sometimes you were designing the computer at the same time). So the software engineer was the job of figuring out the optimal way to marry the hardware and the software to the requirements. And, how to get it done on a budget. Formal reviews were required to make sure everything was going to work together.
Now, the computing system is well-defined, very little design (if any) is done before coding begins, and the “engineering” aspect of software is not evident.
In my opinion, the general confusion on HN around this topic stems from the fact that “software engineers” are programmers, who are in turn technicians. The distinction now may have more to do with number of years of experience and credentials than with job function. I know this is a generalization, but I have been in the engineering business for almost 40 years, so I have lived through the transition.
In my opinion, the jump to writing software before design is a mistake. Yes, you get some code in a hurry, and it’s a good way to develop a rapid prototype to talk about. But is it optimal? This is where good software engineering comes in.
The article makes some valid points. But I'm confused by some of the claims:
The author might be better described as an electronics engineer. Electrical engineers design systems with voltages and currents that would burn up the breadboard. Think industrial equipment, power stations, etc. Hundreds if not thousands of volts and often thousands of amps.
Breadboards are Ok for low voltage, low frequency circuits. You can't breadboard a radio frequency circuit or even a modern day computer circuit the "parasitics" will stop circuits from working properly.
Traditionally, physicists and mathematicians develop the technology, engineers apply the technology by designing things for commercial applications, and technicians try to build the design and make corrections based on experience.
Consider a large building: architects, structural and architectural engineers, and trades people.
In software, there used to be a formal design phase. The computer hardware and operating system was not a given (sometimes you were designing the computer at the same time). So the software engineer was the job of figuring out the optimal way to marry the hardware and the software to the requirements. And, how to get it done on a budget. Formal reviews were required to make sure everything was going to work together.
Now, the computing system is well-defined, very little design (if any) is done before coding begins, and the “engineering” aspect of software is not evident.
In my opinion, the general confusion on HN around this topic stems from the fact that “software engineers” are programmers, who are in turn technicians. The distinction now may have more to do with number of years of experience and credentials than with job function. I know this is a generalization, but I have been in the engineering business for almost 40 years, so I have lived through the transition.
In my opinion, the jump to writing software before design is a mistake. Yes, you get some code in a hurry, and it’s a good way to develop a rapid prototype to talk about. But is it optimal? This is where good software engineering comes in.