I rather write garage code than using slop

you are viewing a single comment's thread
view the rest of the comments
[–] 29 points 2 weeks ago (13 children)

I suggest you also spend some time learning about data structures and algorithms. That's one of the things that separate university trained computer scientists from many self-taught or college trained programmers. Spending some time learning about how the hardware works is also worthwhile. Just stick these on your todo list, they aren't necessary to understand to program, but the quality of what you make will improve if you understand these.

There's a lot of other ways you can increase your depth or breadth of knowledge, but IMO those are the two biggest.

And for software design in general, the biggest thing I'd emphasise is modularization. Try to break up your program into different pieces that fit together. Like for a game engine, you could have the renderer framework as one module, input handling as another, and game logic as a third. Then, if you want to make a different game, you just swap out the game logic module and you've got a different game using the same renderer and input modules. This is essentially what libraries are.

  • source
  • hideshow 13 child comments
  • [–] 2 points 2 weeks ago* (last edited 2 weeks ago) (4 children)

    algorithms and data structures are overrated IMO. No average programmer is manually writing sorting algorithms or reversing linked lists. And if you do need something like that, you can just google it

    edit: People have raised some VERY good points about why DSA is actually really helpful, and I guess I have to agree. I think this comes from my disappointment upon entering the professional programming scene and realizing that readability, maintainability, organization and structure are 10x more important when it comes to collaborating with others on production-grade codebases.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 3 points 2 weeks ago

    Even if you aren't implementing them directly, it can help to just understand how a library you're using implements them. Most data structures have scenarios where they are the best option and other scenarios where they are the worst option. Which one you pick can be the difference between an operation taking under a second vs minutes or hours. Hell, even if you pick the correct structures overall but layer them poorly, you can see a difference like that.

    It's like adding an understanding of engineering to woodworking skills. Sure, a carpenter could probably build a house frame on their own, but it takes engineering knowledge to minimize the amount of wood used while still being sure that it won't fall over if the wind gusts to 100 km/h or it snows during winter.

  • source
  • parent
  • load more comments (3 replies)
  • load more comments (11 replies)