Fwd: Re: starship-design: Hi, from a newbie

[Curtis Manges <clmanges@yahoo.com>]

--- "L. Parker" <lparker@cacaphony.net> wrote:
> 
> Peter et. al.
> 

> Which of course brings us to the ship/colonies other need: a 100
> percent
> uptime network. No failures, no shutdowns, no restarts, no lost
> packets, no
> bottlenecks -- zero. Can't have the ship not firing thrusters
> because the
> server's network card burped and restarted the network...
> 
> Probably you would actually want several networks, some less
> critical than
> others, but with access nodes between them that were software
> routable. Kind
> of like the Internet, but better.
> 
> Lee
> 

Maybe my mention of redundancy reflects old-fashioned thinking that
is still in use, but I see a redundant system set up such that, if
one of the units goes down, its partner (or maybe a separate,
independent monitoring system) yells for a replacement.
Redundancy is, you know, the current approach used in lots of
critical aircraft systems. Yes, it would be nice to have a system so
robust that it needs no backup, but until one is proven to exist, it
may be more effective all around to double up on the really important
stuff.
I do not like thinking in terms of having more time to fix something
just because it's bigger. Prudence dictates setting the system up to
avoid having to operate in crisis mode. Interestingly, this may
involve the planned obsolesence or even planned failure, as it were,
of certain parts. Remember the Wonderful One-Horse Shay? Every part
of it was equally perfect, and when it wore out, the whole thing
disintegrated simultaneously. You do not want that to happen in
space.
This is all a round-about way of giving Lee credit for the network
idea; sort of another way of approaching the issue of emergency
backups.

keep looking up,

Curtis


=====
get the facts! visit www.worldnews.com

__________________________________________________
Do You Yahoo!?
Send FREE video emails in Yahoo! Mail!
http://promo.yahoo.com/videomail/