From: joemonk64@gmail.com   
      
   > My idea of efficiency is compiled C code, in real   
   > memory, running at native speed, and with all   
   > its data cached, so quibbling about I/O is a   
   > non-issue.   
      
   Then you will never have efficiency.   
      
   There is no way for a program to know, ahead of time, what data it needs and   
   doesn't. Reading a million record dataset for the 999,999th record just so it   
   can be in "cache" is not efficient. You have to read the entire dataset.   
      
   IBM does this today with hiperspaces. Entire datasets are kept in cache. But   
   there are limits.   
      
   Joe   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)