I guess it's both better and worse...

I wasn't trying to be mean by the way.

While this works, I'd generally avoid naming my own function "sum" since it overwrites another function with the same name in Python, even if only in the scope of the "frog_contest" function in this case.

Thank you for pointing it out.

Python: random tests are broken. According to the kata description returned value is supposed to be AN INTEGER, but they expect me to return values in a float form ending with ".5".

You're a spammer, Harry -.-'

Excuse me, I'll ask just to be perfectly sure - was it supposed to be a joke? I mean, I know it should work, but...

Same problem in C#. Description asks for the answers starting with small letters but accepts only the solutionwith capital ones.

After the first day it'd lose 10 percent of 10 which is 1 so we're left with 9, right? But after the next day we're gonna lose 10% of what's left (10% of 9 which is 0.9) instead of another 1. So the amount lost in a day gets lower in time.

[C++ version] Description says "The input should be modified, not returned" yet the test cases checks the returned values, so... I basically have to return them anyway. Maybe I can't understand something or it just makes sense in some other language.