On Thu, Nov 6, 2008 at 00:17, Roger Hui <[EMAIL PROTECTED]> wrote: > Re: The implementation should reward conciseness. > > In APL it was (and is) well-known that the inverse of > a permutation p can be computed as (i.#p) p}p > and is more efficient than /:p . But watch this:
How does (p i.(i.#p)) compare to these? Ambrus ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
