[Talk] "Real Software Engineering" by Glenn Vanderburg

Summary

Abstract of this talk by Glenn Vanderburg himself:

I think this is my best talk. In it, I explain why I think software development really is an engineering discipline (even though formal software engineering methods have mostly failed) and provide a way of viewing current accepted software development techniques in an engineering context. Some software engineering professors show this talk to their students. I’ve given it at several conferences, and there are multiple videos available—each with slightly different material, as I’ve continued to learn and refine my thinking on this topic over the years.

He gives a very nice definition of engineering by showing examples of how engineers solve problems in different fields of engineering.

He defines engineering as a spectrum; and by examples, shows despite the differences and variations in methodologies of different fields, they are similar enough to call all of them "engineering".

And then he tells why "Software Engineering" is actually a real thing and it's a field of engineering, and not art (and I also recommend to read Don Knuth's ACM Turing award's lecture in which he talked at length about the meaning of "art").

Why I enjoyed?

He presents an interesting perspective on engineering in general.

I want to emphasize that he's talking about methodology, not terminology!

Thoughts

As a side note, while I was watching this talk, it reminded me of this great observation from Doug McIlroyref:

Design and build software, even operating systems, to be tried early, ideally within weeks. Don't hesitate to throw away the clumsy parts and rebuild them.

My favorite part: "even operating systems"!

I believe in experimentation, write some code, and throw it away, and start fresh with a new understanding of the problem. And do it over and over again. This kind of feedback helps a lot to get a better understanding of the problem.

In the end, I should say that I first got curious about Vanderburg when I saw a quote in a talk by Kevlin Henney called 1968:

A capsule definition of engineering, independent of any discipline, as you're likely to find: the set of practices and techniques that have been determined to work reliably through experience.

I like this definition.

Happy watching!