Re: normalisation vs denormalisation

Sorry for the tone of my original response. OF course I am not a proponent of NEVER having any column or table denormalized. This just doest make sense. I have however heard this argument many too many times to let it slip by. If you don't do a good design up front you're heading for disaster. The easiest way to do a good design is to do a normalized design. I find that 99.9% of all percieved performance problems can be handled by better SQL or a slicker Pro*C routine without resorting to messing up the database.

It happens to be very frustrating to see entire projects fail because the people involved find it easy to fall back on this idea that somehow denormalization will solve their problems, instead of thinking a little about how to improve some response time...

Randy :)

--
    ..uu.                                     ----------------------
  .?$" '?i     .                              I  Randy DeWoolfson  I
 .T^M  ._at_"    d9    .     f   ,.un.  b,    i  I--------------------I
 "  Z :#"    M `8   U    <  .dP"``"# `M   _at_"  I  randyd_at_cais.com   I
    &H?`    Xl _R   $5.  $  ?*    _at_   'P,#"   I--------------------I
  ,d#^*L   :RP'~$b  f`$L:M  Xf  .f'    dH`    I        ,\//.       I
    &  'M ,P    `E  M   "$  Mux~      n!`     I        |o o|       I
   dk   `h"       ' j     " y"       *~       I====oOO==(_)==Ooo===I