Humble disagreement on straightforward recursive fibonacci implementation being "divide and conquer"
Rulikkk opened this issue · comments
I think that binary search is a better representative of "divide and conquer" approach.
Even more, I suggest removing straightforward recursive fibonacci implementation from that group — it seems to belong to group "never do like this" (unless you have some cool recursion optimisation in your language).
The reason is, instead of "dividing and conquering", the problem is actually multiplied to two problems on each step (and one problem is equivalent to another one from previous step).
Maybe code like this would be closer to demo "divide and conquer", using fibonacci and recursion.
Thanks for raising your concern @Rulikkk. I see your points, maybe in the GitHub repo is harder to follow the sequence. Let me share my answers about your comments:
I suggest removing straightforward recursive Fibonacci implementation from that group — it seems to belong to group "never do like this" (unless you have some cool recursion optimization in your language).
This performance issue is on purpose. It's actually the nexus for the next topic Dynamic Programming and the new fib implementation.
I think that binary search is a better representative of the "divide and conquer" approach.
I agree there are many candidates for divide and conquer and we cover others as well.
Maybe code like this would be closer to demo "divide and conquer", using Fibonacci and recursion.
In the same stack overflow thread, you can see that use one of the implementations called memoization, very similar to the one we used on this repo
Let me know if you still have questions or suggestions
I'm cool. I've mostly referred to this section of README: https://github.com/amejiarosario/dsa.js#algorithms
If that's not the core part—it's OK.
Ah, that's a good point. In the readme, I should put another example since there's not much context. Thanks for pointing that out!