From: Spike Jones (spike66@ibm.net)
Date: Wed Sep 13 2000 - 00:05:11 MDT
Emlyn O'Regan wrote:
> Geez Spike, what is it with excel & engineers? Some years back I used to
> work in a company with lots of engineers, supporting small systems.
> Engineers, I discovered, love excel. At some point, the really keen ones
> discover Excel macros, and build systems, which grow until they become both
> mission critical and entirely broken. That was a, ah, challenging job.
A big roger on that. Excel is an incredibly powerful tool, but it does
invite bad programming habits and abuse. Macros tend to grow in
unintended ways. I have many horror stories, but will give you just
a couple of the good ones.
A number of years ago I was working in a conceptual design group.
We had a very sophisticated fortran package that does top level
design, but it was really more than we needed, and the guy who
ran the thing had other stuff to do than fool with us. Turns out he
was busted for a security violation, Wen Ho Lee-like, only in this
case he really was so enamored with his job, he just kept taking
datafiles home until he had a ton of classified stuff in his home.
Then he got caught. They investigated him for months and decided
he was honest, just somewhat misguided. No criminal investigation,
but he was fired, which was a tremendous loss to the company.
I ended up developing a spreadsheet that did what his successor
couldnt quite manage in a reasonable time. It was a solid rocket
modeller that took into account nozzle shapes, loads, atmospheric
conditions, fuel, variable drag, yakkity yak and bla bla. It has
44 inputs, and will calculate how much, how far, etc. It kept
growing ever more sophisticated until it got to the point where it
consistently produced answers very close to the real software.
Would you believe it, they are *still* using that spreadsheet to
this day? It is faster and more convenient than software that
was designed right. I have an even better story, but it must wait.
spike
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:30:57 MST