I think he makes a good point. I was thinking about posting an "Ask HN" about what would be considered the fundamentals of programming/software development, but will ask it here instead. Does anyone know of a decent list of to start with? Basic algorithms and data structures would be on there, off the top of my head.
I mean, past pairs, linked lists, [dynamically resizing] arrays, and binary trees, it gets either more complex or more esoteric. Are quadtrees basic everywhere? Is knowing the way a red-black tree works, by heart, useful? Does it really help to know exactly how hash-table based dictionaries work, or is it sufficient to have a vague understanding that a function transforms identification numbers into "bucket" slots? This isn't rhetorical, I just don't know.
I doubly don't know for algorithms. To me it seems the algorithms themselves are never fundamental, it's the techniques and thinking behind them that is fundamental. Nobody cares about binary search, we care about "divide and conquer"-- you can rebuild binary search from that.
Yeah, learn the basics of divide and conquer, greedy, and dynamic programming algorithms and you will be doing well. Of course to learn that you also need to learn how to apply those to basic data structures such as graphs, skip lists, and various n-ary trees.
The algorithms that the MIT lectures go over in the introduction to algorithms course should be a good stepping stone to knowing enough to get you by in an interview for a junior programming job.
I guess I should amend this to say that if this is the first time going through the material you might just want to pay cursory attention to the big-oh notation proofs and just look at the overall speed of the algorithm. Ignore the proofs and look at the result. Though in a job scenario they will want to know the overall running time of the algorithm that you created so it is beneficial to know. (IE quick-sort is generally a good algorithm and will run in O(n logn) on the average but there is a degenerate case in which it will run O(n^2) namely a sorted list.
