Re: rm RULE based optimizer != GOOD IDEA

  • From: Mladen Gogala <mgogala@xxxxxxxxxxxxxxxxxxxx>
  • To: lex.de.haan@xxxxxxxxxxxxxx
  • Date: Fri, 22 Apr 2005 13:57:26 -0400

Lex de Haan wrote:

>you don't have the faintest idea about the size and complexity of Oracle's
>regression tests,
>and the frequency they run with ... bug-free software is an utopia.
>kind regards,
>
>  
>
No, I do not have the faintest idea about the extent of regression 
testing. There are missing libraries,
patches shipped with old versions of the product (they actually 
downgrade the product, oracm on 9.2.0.6
for Linux). As a laymen, I can only look in awe at the mighty Oracle 
Corp. and characterize their valiant
efforts as awe inspiring but obviously insufficient. Nobody is 
complaining about bugs, the vast majority
of  us makes quite a handsome amount of money off them, what we are 
complaining about is the size of
the bugs. Some of those bugs are larger then an airplane carrier and yet 
they pass unnoticed. I must say
that I ascribe these bugs to outsourcing to a tropical region, with far 
greater tolerance for bugs, spiders and
scorpions. I wonder when will we start seeing spectacled cobras embedded 
into code instead of bugs?
Bug free software may be utopia, but producing a square wheel is not 
what I'd call a "bug". There is a difference
between a bug and a fatal flaw. Patch quality of 9.2.0.6 patch for 
Oracle EE RDBMS 9.2.0.4 on Linux is
such that it makes the product completely unusable. That is not a 
patchset, it's a disaster of biblical proportions.
I surmise that the purpose of that particular patchset is to confuse the 
enemy and that it was conceived by Terry
Jones, Terry Gilliam, Michael Palin, Eric Idle and John Cleese on their 
new positions in the Oracle Corp.

-- 
Mladen Gogala
Oracle DBA
Ext. 121


--
//www.freelists.org/webpage/oracle-l

Other related posts: