Conversation
how does one reduce complexity of complex programming issues? sadcat
7
0
0
@kaia divide and conquer is the most basic approach
1
0
3
@kaia but basically you just have to spin the problems around a bunch until you find good ways to solve them
1
0
1
@kaia experience is what gives you the tools to do this well and fast. You will always produce turds you look back on in shame and next time you'll know better
0
0
5
@kaia do a "stupid" first solution, you often understand the issue better after that and can do something smarter
2
0
7
@kaia if you can stomach the math, look at how recurrence relations are solved, quite a few recurrence algorithms can be reduced to a single expression, other wise the general, d&conquer, dynamic programming, branch and bound etc
0
0
1

Yeah it's only by trying to solve the problem that you get to understand what it even is (sometimes)
@eal
@kaia

0
0
2
@eal @kaia it’s all fun and games until your cobbled up hack remains in the codebase for years (i’ll get back to it someday)
1
0
4
@cell @kaia @eal I finally got back to it (themes 3)
0
0
2
@kaia run time complexity or human understanding complexity?

anyway thats basically why Algorithm Theory is a huge part of Computer Science.

the approach depends on the type of issue. can it be optimized or solved? u can calculate bounds for complexity for a problem.

u also learn to convert issues into other similar issues. they might not sound the same but are equivalent. if u have solved one of them u have solved all of them.
0
0
1

@kaia You establish a SAFe project structure, thus making sure that the actual coding will be pushed back by meetings until the original problem has been solved in some other way.

0
1
1
@kaia You don't reduce complexity you break it in small pieces to be able to manage them more effectively.
0
0
0