[Mud-Dev] Money supply in game economies (formerly Broken eco nomies)

J C Lawrence claw at kanga.nu
Thu Apr 5 23:42:33 New Zealand Standard Time 2001

On Thu, 5 Apr 2001 13:17:08 -0700 
Raph Koster <Koster> wrote:

> "Opportunity for exit."

Without restating your point, which I'm now going to do anyway,
perhaps the primary difference in the evolutionary pattern of RL
versus VR is that in VR individuals can, with rare exception,
excape, compleatly and utterly, and they know it going in.  RL
conversely has the tawdry quality of having your body both stuck in
it, and of it being largely unremovable.

On the social side of analysis we talk of the "social contract" and
of levels of "expectation" and "agreed upon behaviour".  These
things rest on a base which assumed that identity is singular and
that accountability, as a side effect of authentication, can always
be derived from a singular identity.  When identity is no longer
singular we reflexively attempt to build systems, both social and in
mechanical suports for social systems, which attempt to maintain the
illusion of singular identity and its ramifications on
authentication (seperation (or not) of character and identity, you
didn't do it, your character did, many aspects of role playing,
forms of consent, etc).

Its a nice, and more importantly, comfortingly familiar illusion.
Problem is, its easy to shatter and it really doesn't work all that
well in the first place.  The world is tough enough when it isn't
populated by Loki-like identity tricksters, even well meaning ones.

I'm not convinced there is or will be a clean solution.  The
advocated solutions to date have circled around various forms of
digital signatures and other public key crypto applications, which
while all very nice (and appealling to repressed techno geeks like
myself), also have very nice large human-factor holes to match.

  Yes, PGP and GPG are wonderful.  How many of you use them on a
  regular (more than once a week) basis?

The current commercially motivated direction (cf prior discussions
on credit card fraud) is to use bioometrics (I've had reason to poke
into this area).  My concern has been that the authentication might
be bidirectional, that not only would I be able to derive an action
back to a singular human, but that I would be able to take a
singular human and from its signature derive a list of actions, or
that I would be able to take two actions and would necessarily be
able to determine that they were commited by the same human were
that the case.  Very Bad Stuff.

However, with the hoped-for asymmetric setup (you can derive humans
but you can't derive actions etc) the social effects in VR are still
not trivial.  Identity is only partially singular, and more
particularly, it is not singular in the VR important ways.  There
remains no way to determine that two characters necessarily derive
from the same human without that human's consent (think
non-reversable hashes ala Unix crypt() or MD5), so you still don't
really know who, in a human sense, which really means in a sense of
singular identity, who is doing what.  Without some level of
complicity from Bubba, you can't prove that Bernie and Boffo are
actually two characters which are both played by Bubba.  Ergo, the
definitions of trust, and of risks and veracity analysis become
hopeless complex and subjective ("But his typing looked so
trustworthy!  He spelled well!", "Of course Bernie isn't Boffo,
they're such totally different people, and anyways, I'd know if
someone was trying to trick me like that!").

Oy vey.

J C Lawrence                                       claw at kanga.nu
---------(*)                          http://www.kanga.nu/~claw/
--=| A man is as sane as he is dangerous to his environment |=--
MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list