Didn't most of us get in to this mathsy line of work precisely because we didn't need to memorise a load of stuff and could just work things out from first principles as and when required?
I remember memorizing a load of stuff for my math degree. How did you write proofs without memorizing terms like "Method of Frobenius", "triangle inequality", "Zorn's lemma", "Bolzano–Weierstrass theorem", "Cantor diagonalization", and so on?
"Oh, you just have a loop on b dividing it by 2 using shift operators until it is zero."
"Wait a minute, you said you couldn't use loops."
"Did I? Ah, well."
Hackerrank/leetcode exercises are written the same way. So many times that a problem asks "Output the indexes of two numbers in the array such that their sum is K" and you write your code and the website says "INCORRECT! You said [3,6] but the right answer was [6,3]". Addition is commutative! The two are equal! And both right!
The real way to multiply without loops is to use a lookup table, or unroll the (fixed iteration) shift-and-add loop.
That page is both hilarious and sad at the same time. Hilarious because the second "solution" clearly has a loop, and sad because sites like those don't really help anyone. Some of the pages on that site are downright WTFs:
The shift-and-add algorithm for multiplication is usually implemented as a loop that iterates for the number of bits of the operand, so e.g. for an 8-bit x 8-bit multiplication, the loop runs 8 times. (An "early out" algorithm when one of the operands becomes 0 is also common, but let's not complicate things here.) It's trivial to unroll this loop into the 8 individual shift-and-add steps.
I don’t see what recursion/loops have to do with constructing output or dividing and conquering input.
For one thing, all loops can be trivially unrolled into a tail-recursive function that most languages will trivially transform back into a loop. Clearly these transforms have no bearing on the nature of the computation.
If asked such a question, I’d probably just use goto. (Bonus points if candidates use a goto on one part of one question I ask, since it saves a minute or two of whiteboard cut and paste, BTW).
> For one thing, all loops can be trivially unrolled into a tail-recursive function that most languages will trivially transform back into a loop.
That there's a transform between them doesn't make them the same thing, it makes them possible to use for each other in programming. You even seem aware that there's still a distinction -- you're just not sure why it matters.
> Clearly these transforms have no bearing on the nature of the computation.
Depending on the nature of the job, not understanding why you can transform between loops and recursion is big points off. (You can go back and forth because while not the same thing, they're duals, which means you'll be able to find similar structures in both for certain classes of problems.)
That's why you need to understand what induction vs co-induction is -- so you can prove things about the transforms between them, such as that it doesn't impact the result to change the algorithm in a particular manner.
Similarly, you're only talking about really simple inductive or recursive behaviors -- what about complex cases? Are all inductive structures transformable to co-inductive (or the other direction)? What does it do to computational complexity when you make the transform?
That there's an uninteresting kernel in the transform doesn't make the entire transform trivial -- it just means you're only used to working in the "well-behaved" portion of it. (And that there is such a kernel is itself an interesting fact about computation!)
> Clearly these transforms have no bearing on the nature of the computation.
Just to reiterate how off-base I find this comment: there's entire huge collaborations in mathematics and academic computer science exploring the nature of those transforms because understanding how they can be applied is essential for moving forward with things like mechanically/formally verified software.
I get that not everyone is going to know that distinction and its impact -- but the distinction absolutely matters. As you astutely point out, it's used by compilers routinely -- and they certainly need to be sure the transform is well-behaved in the cases they apply it!
You said that recursive and iterative computations are fundamentally different. I pointed out a class of recursive and iterative algorithms that are equivalent to each other. That class disproves your statement.
Most interview questions are ridiculous and don’t test for the knowledge a candidate should have. When I interview, I ask a system design problem that involves using an inverted index, a bst, understanding how to normalize data, and then being a little clever when merging some data. It tests fundamental concepts in the context of building a working solution to a meaningful problem.
I never understood why anyone would ask say DP questions in an interview. We don’t use that in 99.9% of software (I’ve never once in my career found a use case). What you’re really testing is whether or not the candidate had enough time to refresh that material. Worse still, it involves a very rigid solution pattern that’s over-specified to that class of problem and tells you nothing about what you want to know: can a candidate synthesize concepts to devise a solution to problems we actually encounter?
If anyone asks these questions without a particular scenario to actually deal with in the course of the position you're interviewing for, you are in for some bullshit at that company. This has become all too common at the major corporations sadly.
There was just a post on HN to a site that went over this exact problem. Essentially, you do two linear scans to calculate the product of every number before an index, and to calculate the product of every number after an index. Then, for each index, the answer is just multiplying those two numbers you found for that index.
I agree. I studied these sorts of questions through hackerrank, leetcode, etc. and became a better programmer. That said, I wouldn’t want to work somewhere where they asked me to find a “zero sum sub array” in an interview