Andrew Lentvorski wrote:
DEC could have shipped EV6 at least 12 months earlier. Had they, the
company would probably still be around. However, there was no
incentive to ship. The only reward was to be a notch higher on the
next project. That meant "don't ever make a mistake as your career
will never recover." Lots of "defect" catching and testing ensued.
The fact that the company was going down the drain never entered the
picture.
The other aspect of the problem is the notion that it is cheaper to
prevent defects than to fix them. It is true that for a given defect, it
is easier to prevent it than to fix it, but preventing defects falls
under the law of diminishing returns. There's tons of low lying fruit
that can be prevented, but trying to to get that last little bit tends
to be more expensive than anything else preceding it (much like
achieving 99% uptime tends to be way cheaper than achieving 100%
uptime). At some point, it becomes cheaper to just build the [EMAIL PROTECTED]@#
thing, try it out, and then fix what you observe as not working.
--Chris
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
