Re: AIs and such...

In message <199707172025.QAA24262@smtp1.sympatico.ca> Allan Goodall
writes:
> At 11:03 AM 7/17/97 +0000, you wrote:
> 
> >Any ideas as to how to tie this philosophical argument with the game 
> >mileu or the rules?
> 
> Funny enough, I was thinking that it was about time to pull the thread
back
> to FT. I started the thread as an attempt to justify what we see in FT
based
> on an extrapolation of AI technology. To my mind, the idea of small
ships
> and fighters having horrifically high casualty rates is no problem:
they are
> merely run by computers and just a loss of hardware.

I don't see any problem with saying that these expendable ships are
piloted by humans. One doesn't really design any military force on
the assumption that you are going to fight nice, fair, even point
battles that end when the last ship on the losing side is destroyed.
One rather hopes that one will fight an opponent that will run away
when confronted with your superior force, or which will be 
overwhelmed by your advantages in material, technology and/or 
tactics.

(Or in Drek Trek, you built exploration/research/diplomatic vessals,
that merely happen to be heavily armed.)

However, there may well come a time when the piddling little escorts
that you intended for bullying weedy merchants into doing what they 
are told have to be expended in a humungous bloody fleet action, 
which is hard on the crew, but they drew the short straw that anyone 
else might have picked. Fighters, likewise, you hope you have 
superiority, or that you attack piddling little one PDAF ships, but
when you have to do-or-die, you send them in, death or glory.

> The idea for the thread was to try and come to some group consensus on
a use
> of AI in the FT universe. This would give anyone wanting to write
scenarios
> or fiction a common ground. I'm not sure that special rules need to be
> brought in, except perhaps for disallowing certain classes of ship
from
> having to make morale checks. 

Something I admire about GZG rules in general, although it breaks
down a little at the Stargrunt level, is the "conservation of game 
effect" that says that a stand of infantry is just a stand of
infantry, or a battery is a battery, and you really don't care what
the trivial details are, like what calibre rifle the infantry 
carry.

I've been following the thread and, while there isn't a very strong
consensus, it seems that a fair POV would be that these if some
fancy A.I. is going to replicate the function of the vast number of
vastly-interconnected neurons in the brain, you're going to get a
biomechanical machine that goes >squidge< and blacks-out just like
we do, that isn't programmed but learns slowly, just like we do, 
thinks much like we do, imperfectly, and which might well turn out 
to be treacherous, psychotic, sociopathic or meglamanac... just 
like we might (particularly if we have all our emotions suppressed). 
I'm really starting to wonder what people think the advantage of 
this is. Wouldn't a human just be cheaper? You can't invite the A.I. 
commander of an escort onto your flagship for luncheon either. Or
marry your offspring to him/her.

I feel like we might as well say that the presence/absence of an
A.I. on any given ship is just a trivial detail.

At the day's end, the fleet will be commanded by a human... me or 
thee. I guess if you want to simulate super-intelligent machine
commanded forces you could subtract d6% for the points value of
each force, square each forces new strength, subtract one from the
other and take the square root. This will give you the points value
remaining. Fun.

> Actually, if there are drone fighters and drone escort ships, I could
see
> drone tanks in DS2 and SG2. That would be kind of cool. I'd imagine it
would
> be very low and sleek to present a tiny profile. The main gun would be
> mounted on a swivel mount on top with maybe some armour protection.
There
> would be no need for a big turret. That would be a cool figure for GZG
to
> produce.

Scotia make nice SF tanks with guns-on-stalks in 1:300.

-- 
David Brewer