[
Lists Home |
Date Index |
Thread Index
]
In the small, I agree, Robert, but when the system
gets very large and has strict performance metrics,
tough issues of storing and sorting very very large
datasets, mining the data, and so on, it really is
hard. There are dozens of committees grinding out
specifications and standards that can be included
into a contract with a single sentence.
The CALS vets will understand: we really have to
build the bloody thing this time and it is much
more real time than last time. The time of
programmer experimentation and Starbucks programming
over a latte is over. This is back to the cube,
long hours, kiss your wife and kids goodbye until
the deed is done stuff. The good old days....
Guys like deRose who really know this down to
the nits can be very helpful.
len
From: Robert Koberg [mailto:rob@koberg.com]
Steven J. DeRose wrote:
>So assembling a system that works well from the tools at hand is
> far from trivial.
I don't know if that is true or at least not as bad as you make it out
to be. If you use the 80/20 rule (as is generally understood today), I
think it is trivial - for simple folk like me. It comes down to the best
way to deal with the current reality. If you want to cover everything
then you will never get there (an archer shoots an arrow which has to go
half way first always).
|
