Re: Anyone help develop an algorythm?

["Marin David Condic, 561.796.8997, M/S 731-93" <condicma@PWFL.COM>]

Joel VanLaven <jvl@OCSYSTEMS.COM>writes:
>Note that selection sort (still O(N^2)) is much better than bubble sort
>and is "just as simple".  It would also be easy to apply to your
>situation (as the primary sort).  Basically, if you are really using
>bubble sort, almost anything at all would be better.  Of course it might
>be that you are not using what I call a buuble sort.
>
    Of course, an alternate strategy is to always start development
    using the Slowsort(1) algorithm. Moving to almost anything
    else (including bubble sort) gets you an immediate improvement in
    performance and much customer satisfaction ;-))

    (1) Slowsort is an algorithm developed by myself & Bob Zaret in
    which you generate a random permutation of the list, then check to
    see if the list is sorted. If not, do it again... Behavior is
    O(N!). It was written up in Ada Letters many moons ago.

    MDC

Marin David Condic, Senior Computer Engineer    ATT:        561.796.8997
Pratt & Whitney, GESP                           Fax:        561.796.4669
West Palm Beach, FL                             Internet:   CONDICMA@PWFL.COM
===============================================================================
    "A verbal contract isn't worth the paper it's written on."

        --  Samuel Goldwyn
===============================================================================