hello frank!! im about 3/4s of my way through getting my computer science degree right now, do you have any advice? or maybe memories from when you were in university?
Well, it's been a long time since I took any computer science classes, so I'm not the best person to ask. However, I did enjoy my CS classes, although in retrospect the ones that were the most useful were probably the (relatively few) ones that weren't programming classes. (If you get the opportunity, I recommend it if you can handle the workload. Computer science requires more practical thinking than the usual liberal arts degree, and even "practical" means something different in the context of programming than it does in the context of, say, philosophy.)
Here is some (rather generalized, no doubt obvious) CS advice, most of which I can guarantee I was given at one point or another:
Never leave out blank lines in your code. Even if the code works fine with them. It just makes it more difficult to read and maintain later, since you have to mentally fill in those blanks. I got this piece of advice from my freshman year calculus professor, as a way of explaining why he was insisting on putting extra space around all his mathematical symbols. It applies just as well to programming.
Always keep in mind the machine that is running your program. If you're doing something too slow, or using too much memory, you might have to start from scratch and figure out how to do it less inefficiently. Write every line of code as though you could be running it millions of times.
If your code is running in a loop, every one of those loops must complete in a finite number of steps. Start each loop iteration on a clean stack. If you're operating on an array, the variable should always contain a valid array.
In almost all cases, seek out the most elegant solution to a problem. Again, don't give up efficiency to reduce mental effort. Keeping the mental model of the machine in mind helps here, because you won't end up writing something as slow as possible on purpose, then end up speeding it up by adding some low-level optimization later, which adds to your mental load and increases the maintenance effort. If you can, start with the most straightforward and elegant way of solving a problem, and if it's too inefficient (or otherwise unsatisfactory in some other way), just start over with something else.
