[MUD-Dev] Containing automation?

[Timothy O'Neill Dang]

On Wed, 21 Jul 1999, Katrina McClelan wrote:

> If the only real cost of manufacture is the time spent typing it, then the
> economic model is weak.  The real problem here is that players are coming
> on to play swordsmen and wizards and such.  Some people are content
> running shops, but most are wanting adventure.

Still, the problem with the ability to automate production is that, well,
you've automated production. If a bot is running someone's factory while
that player is asleep, then that player is playing a different game. They
have miniscule opportunity cost being engaged in this important economic
activity. If this becomes the case, we might as well implement trade-war
style dynamics where instead of writing a bot part of the game is a
spreadsheet on allocating labor and capital to various production tasks.

We want to avoid this trade-war style game in general just because of the
focus of our research. Economic situations which can be modelled well in a
trade-war game we can already research in our laboratory; what we can't
research is individual choices in a rich long-lived environment with
freeform goals.

So we want the game players to be acting with their characters as
individuals. If these individuals gather together to form a business,
that's great. If, on the other hand, an individual acquires a completely
obedient full-time employee at zero cost (a bot), that throws us off.

> They go to (work|class) every day and lead normal lives, so they come
> to a MUD for adventure.  If you make specializing in sword making get
> in the way of learning how to use the weapons, you won't find many
> people using up their aptitude to make lots of coins off sword
> production.

There are many reasons one might want to get into production rather than
merely adventuring, perhaps role-playing a craftsperson, perhaps non-RP
interaction with others, perhaps exploring the game physics of smithing,
perhaps exploring the game economy, perhaps simply to be self-sufficient
and make one's own tools.

> a free market economy will stabalize itself if you implement it correctly.

Often it will stabilize, but it might stabilize to an undesireable
equilibrium (for us as researchers, this isn't a problem, for game
designers it can be) or stabilize via undesirable dynamics (that's our
concern).

If automating production is possible, it will be very hard for manual
players to be involved in production. Manual and automated players will
face all the same constraints *except* that automators have no opportunity
cost. Even if the automator and manual player were both interested solely
in production, the automator will have all that extra time to sell their
wares.

One obvious equilibrium in this case is that the first few folks to
implement sucessful bots to produce good X produce most of good X. They
make little off the sale of each unit, but the quantity they produce makes
it worthwhile. Someone trying to get into this market manually will make
the same small amount on each unit, so that it's not worth their time to
work on it. Another botter could move in on the market, but doing so would 
punish themselves as much as the poeple already in the market, so why
bother?

> Such fluctuations are normal for a free market economy, and should
> defeat a bot that can't react to the changing market fast enough.

I'm assuming that the folks running bots do touch base from time to time 
(a couple times a week, say) to verify market conditions and tweak
parameters in their bots. A market which fluctuates too fast for this
would be a very unstable one indeed.

Also keep in mind that some of the dynamism of the real world economy
comes from the ability for us to change conditions of the economic
environment. In real life, if oil gets too scarce or expensive to extract,
not only can we shift to natural gas (switching resources could be
realistic in a MUD), we can also develop technology to extract oil
more cheaply. This sort of internal technological change is something I've
rarely seen in a MUD, for a lot of good reasons.

But without changes such as this, the economic environment stays very
stable except when the admins throw something new in. Under such
conditions, one would expect market conditions to be reasonably
predictable.

This lack of change-from-the-inside is actually one of our big stumbling
blocks in desiring to do research on MUDs. They still look very valuable
for microeconomic research, but the economic growth issues are still
highly artificial, so macroeconomic research will have to wait for the
state of the art to advance.

------------------------------
Timothy O'Neill Dang/Cretog8
timothy at nmia.com
H: 505-843-6966
W: 505-244-8803
One monkey don't stop no show




_______________________________________________
MUD-Dev maillist  -  MUD-Dev at kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev