--- Geoffrey Young <[EMAIL PROTECTED]> wrote: > > > There are two usual rebuttals. > > the third being "just add it and let me decide" > > :)
Right in one! I'll go fix that diagnostic thing now. Unfortunately, I think I'll have to violate encapsulation :( As a side note: after reading everything Michael wrote, I remember an incident at the company he mentioned where the test suites died on failure (I *hated* that feature because it was mandatory). One day I was in a meeting with a vice president and he mentioned to a programmer that a particular "adjustmentment percentange" on a page had changed and the customers (a bunch of very powerful prima donnas) wanted it changed back. The programmer took a huge amount of time explaining why the number had changed, drawing diagrams on the board, explaining the new statistical formulas which were being used for the calculations and making it very, very clear why the *new* number was correct. The vice president's eyes glazed over and when the programmer was done, the veep said "yeah, but can you just change the fucking number back?" So the programmer did and the customers were happy. Cheers, Ovid -- Buy the book - http://www.oreilly.com/catalog/perlhks/ Perl and CGI - http://users.easystreet.com/ovid/cgi_course/ Personal blog - http://publius-ovidius.livejournal.com/ Tech blog - http://use.perl.org/~Ovid/journal/
