Comment by asimpletune

I'd like to say that software engineering is a lot like playing jazz. It's really hard for the beginner to know where to start, and there're also endless sources for the "right" way to do things.

In truth however, like playing jazz, the only thing that really matters is results, and even those can be subjective. You can learn and practice scales all day long, but that doesn't really tell you how to make music.

I developed a style of software engineering that works really well for me. It's fast, accurate, and naturally extensible and easily refactorable. However, for various reasons, I've never been able to explain it junior (or even senior) engineers, when asked about why I code a certain way. At a certain point, it's not the material that matters, but the audience's ability to really get what's at the heart of the lesson.

Technically, you could say something that's indisputable accurate, like, "there're only 12 notes in an (western) octave, and you just mix them until they sound good", but that's obviously true to a degree that's fundamentally unhelpful. At the same time, you could say "A good way to think about how to use a scale is to focus less on the notes that you play and more on the ones you don't". This is better advice but it may altogether be unhelpful, because it doesn't really yet strike at the true heart of what holds people back.

So at a certain point, I don't really know if anyone can be taught something as fundamentally "artful" (I.e. a hard thing largely consisting of innumerable decisions that are larger matters of taste - which is a word that should not be confused with mere "opinion") as software engineering or jazz music. This is because teaching alone is just not enough. At a certain point people just have to learn for themselves, and obviously the material that's out there is helpful, but I'm not sure if anything can ever be explained so well as to remove the need for the student at a certain point to simply "feel" what sounding good sounds like, or what good software engineering feels like.

I'll add one last thing. Going back to what I was saying about not being able to explain to "junior (or even senior)" engineers. Were the same "lesson" to happen with someone who is very, very advanced, like a seasoned principal engineer who's built and delivered many things, time and time again, across many different engineering organizations and different technologies - someone like a jazz music great for example. Anything I would have to say on my approach to such a software engineer would be treated as obvious and boring, and they'd probably much rather talk about something else. I don't say this because I mean to imply that whatever I would have to say is wrong or incorrect, but rather at a certain level of advancement, you forget everything that you know and don't remember what it took to get there. There are a few who have a specific passion for teaching, but that's orthogonal to the subject.

I think it was Bill Evans who said something like "it takes years of study and practice to learn theory and technique, but it takes still a lifetime to forget it all". When you play like you've forgotten it all, that is when you achieve that certain sound in jazz music. Parenthetically, I'll add that doesn't mean you can't sound good being less advanced, but there's a certain sound that I'm trying to tie together with this metaphor that's parallel to great software engineering from knowledge, practice, and then the experience to forget it all and simply do.

I think that's fundamentally what's at the heart of the matter, not that it takes anyone any closer to getting there. You just have to do it, because we don't really know how to teach how to do really hard things in a way that produces reproducible results.