I've been working in the industry for a bit now and I lead a small team of developers (those are not always the same depending on the project I work on). When writing code I normally receive feedback that my code is very clean and understandable. I try to explain to people how important is in gen...
Getting told to do x and y is theory. Experiencing that one really doesn't understand one's own code is practice. Can you pass the bug to the person who wrote that code, such as "hey something is not working as expected. I see you wrote that, surely you can find out what goes wrong here faster than me"? The more time has passed between writing code and now, the better is the chance that this will work.
Well, it is, because if what you want is not aligned with the business reason, then it's furthering what you, personally, want instead of furthering the business.
I don't wanna start a massive debate.. but working with discipline is necessary to deliver the product the customer asked... no disciplines puts actually at risk project deliverable, I say this from experience. We had a employee who was the only one who knew his code and it was considered for commercial product. However it was so badly written we had to ditch the all code and start from scratch that feature. All I am doing in my question is separating the two points cause they're related but I am focusing on a specific one. This is not personal but just common sense...
"no disciplines puts actually at risk project deliverable, I say this from experience." Does your management agree with you that this is true for this specific project?
Some of this seems subjective to me. What is "clean" code and who determines whether code is "clean" or not? Is there a universal or widely and generally accepted standard for "clean" code. Who determines how many nested lops are too many and how many are "just the right amount of nested loops". Who determines if the programming logic is faulty? All I'm seeing in this question are your opinions.
@joeqwerty pretty sure if I ask you (for example) to implement a matrix vector multiplication (which has known complexity) but you overcomplicate it that's an obvious problem... if you give me code where a function to perform A actually does B that's also a problem. If your code has 0 documentation/comments that's also a problem. I am not aiming for perfection. Just few basics where everyone would agree would make the code better and more maintainable.
This is engineering, not science. Yes, easier-to-read code is better. So is more performant code. So is getting something working quickly. And it does take experience to balance these and do any one of them well, If you're trying to teach, understand when to do so, and how to make it constructive criticism. If the code works and isn't unreasonable, consider approving it and opening a technical debt item to improve clarity/documentation, then prioritize that alongside other tasks. "Make it work, make it good, make it great. In that order."
@joeqwerty There is a definition of Clean Code, but I am also not sure if the OP knows it. I'd recommend Uncle Bob's series (available e.g. at YouTube). In short: Write code in a way you do not have to comment it.
@FlorianAlbrecht I'll disagree with "Write code in a way you do not have to comment it.". Code tells you what the computer will do. Comments tell you what you intended the computer to do. They are orthogonal concepts. (Note that the quality of the comments is a different subject. But I'll refer you to my answer here )
@PeterM Well, you may disagree, but to my understanding, this is what Robert C. Martin tries to achieve with his concept of Clean Code (very, very simplified). This is what I referred to, no more, no less.
See also many of the questions listed as "Related" (If you're on a large enough screen, they display in a column to the right of the question.) This comes close to duplicating some of them.
"Quality, Service, Price. Pick any two." (Or more accurately, accept that they do need to be balanced against each other.)
Discussion on question by user8469759…
Discussion on question by user8469759…