[MUD-Dev] D&D vs. MMORPG "complexity"

[Timothy Dang]

On Mon, 19 May 2003, Dave Rickey wrote:

> You seem to be defending the traditional view of irrationality
> being only the result of inadequate information, so any
> irrationality in Camelot *must* be the result of inadequate
> information, and therefore "built in" (since we controlled most of
> the information availability).  This is an easy position to defend
> and a difficult one to falsify, since information for all economic
> actors at all times can never be truly complete in practice.  Am I
> correct (both in my assessment of your argument, and of the
> essential core strength of it)?

> My response is that the very act of assuming only "rational"
> participants in economic activities is an extreme case, since many
> very visible economic transactions in both virtual and real
> systems show irrationality, and in fact people are clearly not
> rational in many of their decisions (economic or not).  The
> assumption has been useful in allowing the theories to be reduced
> to manageable equations, but it should not be surprising if it
> turns out to be a source of systematic error.

> If, in fact, there is such a systematic error, the MMOG's would
> seem to be a good environment for finding it.  The number of
> participants is limited, the interaction with outside influences
> is comparatively minor, the ability to manipulate the core
> scarcities and inherent demands effectively limitless, and the
> information available to an observer far closer to complete, than
> in any real system.

I don't want to get into specific arguments about Camelot's economy,
but this discussion of rationality or lack thereof bothers me a
bit. I've long though about using ORPG's as platforms for economic
research, but I'm not sure they would be particularly helpful in
studying what economists typically mean by "rational" behavior.

Economists sometimes define rationality differently depending on the
context. I don't know of any case where it's been defined so that
making a poor decision due to limited information is considered
"irrational", but the definitions usually used when there is limited
information are different than those used when there's complete
information.

In any case, it's usually quite hard to falsify rationality. It's
easiest when you're looking at decisions which only involve
different amounts of money. When looking at decisions which involve
stuff, rather than money, it's really tricky to demonstrate
irrationality because economists are all trained to accept people's
preferences as primitivives which don't require justification. If
someone is willing to pay $2.5M to have their car painted an unusual
shade of green, a traditional economist wouldn't consider that
irrational.

The most common thing which can be observed in some experiments
which economists mostly consider "irrational" is intransitive
preferences.  Preferences are transitive if a person who prefers A
to B, and prefers B to C, then they prefer A to C. They're
intransitive if that reasoning doesn't work for them. There have
been experiments which show violations of transitivity, but
economists still argue a lot about whether those experiments show
anything important enough to merit re-working the basic models. So
far, nobody has come up with an alternative which seems worth
using. (As a personal aside, I consider the evidence for rationality
is a bit stronger than that against. However, I think that's because
it's difficult to falsify, not because it is correct.)

Anyway, I guess the point of all this is that I'm not sure what
definitions of rational & irrational are being used. I'm afraid that
common-speech ideas of the terms are being used, instead of the
definitions economists use. If that's the case, then one could see
things which are considered to violate common economic models, but
really don't.

------------------------------
Timothy O'Neill Dang / Cretog8
520-884-7261
One monkey don't stop no show.
_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev