Aaron Konstam wrote:
I would hate to put some rationality into what is otherwise a great
discussion, but have any of you tried to create a system of any size
with useful (that is useful to every user) error messages. It is pretty
tough.
This is one of the reasons as Fred Brooks comments in his book "The
Mythical Man-Month" it takes 9 times as long to create a software system
that can be used by people in general than a system that only your use.
Wonderful book.
I recall one time when I created a debugger which could be used
over a network to debug massively parallel processors (MIMD
loosely coupled). I put in the code for the protocol I designed
with no back out, no recovery, just presume all messages get
there, etc. I got it working in about a week. My company wanted
to ship it!
I did a quick estimate, and came up with two months to make the
thing actually usable. I had a very Very VERY hard time getting
management to understand why I couldn't just package it up and
ship it.
I'm sure some of them thought I wasted two months fooling around.
Mike
--
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
Oppose globalization and One World Governments like the UN.
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
