> > Whoops. I started to rant there.
> >
>
> I don't know about the rant Hal, but I think you even drove Dennis Miller
> (and me) to the dictionary. Come on, "orthogonal axis of design space"!?
Well, at least Dennis Miller can learn a new way of saying "linearly
independent variables".
Engineering is the essence of compromise. Speed/space, etc. There's a lot of
variables people must understand to design "world class" information systems.
If you don't understand what can be varied without impacting the other, what is
closely coupled, and what rules determine the coupling, you simply can't make
good compromises. You certainly will have a hard time maintaining it.
Hill climbing algorithm driven design is too easily caught in a local minima if
you don't have some representation of the abstract space that represents the
compromise of the design. Without at least some visualization, you don't see
that there's a lower valley across the mountains from your current design
state. It's designing without eyes.
The alternative "simulated annealing" design methodology throws in a bunch of
heat in the form of a design committee and hopes to jiggle the design to a
lower state. It usually results in the burning of the cake, sadly.
I can hyperlink to an online dictionary, I suppose. <g>Hal
___
I've heard so much about boats coming in
but I see no ships, so I guess I should swim
http://www.hellblazer.com
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
