Re: Expected accuracy of planner statistics

["John D. Burger" <john@xxxxxxxxx>]

> src/backend/commands/analyze.c, around line 1930 as of CVS HEAD:
>
>             /*----------
>              * Estimate the number of distinct values using the  
> estimator
>              * proposed by Haas and Stokes in IBM Research Report  
> RJ 10025:

Thanks for the pointer, Tom.  I shouldn't have been surprised to find  
such a nice comment pointing me at the literature.

- John D. Burger
  MITRE