Hacker News new | past | comments | ask | show | jobs | submit login

Yeah I screwed that up, what I meant to refer to is this:

    >It is not contradictory however, to say that the worst-case
    >running time of insertion sort is [Big Omega](n^2),
    >since there exists an input that causes the algorithm
    >to take [Big Omega](n^2) time.
    >Introduction to Algorithms, 2nd Edition, p 46



Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: