Re: Data structures for cplmsi bijections

In article
<a7a4117d-2b8d-430a-ba53-633dd8d5656b@[EMAIL PROTECTED]
>,
David B. Benson <dbenson@[EMAIL PROTECTED]
> wrote:
>
>Could be somewhere, but I am a retired professor of computer science

Heh!  (The explains why it reads like a spec a professor would write.)


>To make it problem more interesting, assume that lookups are in
>monotonically increasing runs.  Now the efficiency becomes a slightly
>more interesting questions.

Hmmm.  A couple clarifications?

  - We have to take the function as a blackbox?

  - What level of precision are we looking for in the inversions?
    Floating point equality?

  - How does the computational cost of the function call compare
    to the overhead of data structure etc?