Archived from groups: rec.games.roguelike.development (More info?)
Disclaimer about ethical/moral position:
I am an INTP/INTJ atheist, with liberal-left politics (Eco: -7 , Soc:
-3.5 on the Political Compass), and with my own strange "religious"
beliefs that simultaneously eliminate Creator(s) and time-travel paradoxes.
Note: AFAICT, "ethics" is the correct word to use here, though various
websites disagree.
General Idea:
=============
Due to reading a document (which I now can't find) about the problems
with the traditional "good" vs. "evil" systems in some games (e.g.
_Black and White_), I wish to ask the group about a possibility I have
not yet seen anywhere else: ethics inference. The "good"/"evil" is
inferred by the EIE (Ethical Inference Engine). Inference is used in
the technical sense of finding a general pattern from specific examples.
The idea is not to have many restrictions (if any) on what you are
allowed to do (e.g. eating when already full, robbing corpses, killing
civilians, etc.), and not to have penalties for breaking moral rules,
but to let the player act how they like and have the game infer whether
they are "good" or "evil". For balance, both good and evil must have
benefits and drawbacks.
The EIE's inference of "good" and "evil" could be used to give a player
holy or unholy powers (e.g. bonuses for some types of spells, effects of
intrinsicly holy/unholy objects/places, reactions of some beings, or
even the plot of the game (heaven/hell).
One possibility is that "evil" should give short-term profit and
long-term pain, with "good" the opposite, rather than each being equally
easy -- this could be seen as pushing a moral agenda (maybe you want
to). If others' treatment of you is determined by how you treat them,
"evil" may quite naturally end up as a very difficult pit to get out of.
Thoughts On A Specific Game:
============================
This whole idea was partly inspired by a game idea that is kicking
around in my head where the player is a Kigdatsi (a
genetically-engineered highly intelligent dragon-like creature) from a
parallel universe waking up in our universe badly injured in a zoo (or
similar). It could be made fairly clear in the intro text that humans
are to blame for your current situation, which would give you ample
reason to be annoyed. (Has anyone seen anything like this before?)
You (the player) could co-operate with most humans, only kill to defend,
and minimise their suffering, *or* you could slaughter everyone (even
easy early on, as zoo visitors are not normally armed :->), care only
for yourself, and let people die painfully (you are large enough to
swallow people whole). Although you would recover rather faster if you
eat lots of humans, they will try harder to kill you (by sending the
army in), and your own species could despise you for it. The game would
need to do long-term monitoring of how you treat humans (and animals?)
and change their behaviour to fit.
Inferences From The Specific Game:
==================================
In fact, a linear "good"-"evil" system is not the best way to express
ethics. Some topics (e.g. euthanasia, religion, judgement of others,
homosexuality (but how would that appear in a RL?)) will be judged
differently by game-makers and game-players, so we need a more objective
measurement system, maybe one of:
http://en.wikipedia.org/wiki/Political_spectrum
(These are party policies, and not moral opinions, though. Economics
might be a hard one to infer :-> .) However, you will end up with
dozens of moral axes if you try to be too complete, so some
simplification will be necessary.
Some axes that could be inferred (many of these overlap):
* authoritarian -- anti-authoritarian (?)
* pro-euthanasia -- anti-euthanasia (?)
* painless killing -- painful killing
* merciful -- sadistic
* friendly -- aggressive
* run -- fight
* offensive -- defensive
* avoid danger -- destroy danger
* terrorising (scaring people for fun) -- non-terrorising (avoid
scaring people)
* honest -- tricky (e.g. feigning unconsciousness)
* survivalist -- bloodthirsty (e.g. is there any *point* invading a
kindergarten (Brit: "primary school") other than to kill lost of
innocent children?)
A useful system (for the above game at least) could be to have the
colour of beings indicating their mental state(s) e.g.:
Normal, Friendly, Aggressive, Scared, Terrified, Unconscious
as well as physical state e.g.:
Uninjured, Injured, Dying, Dead (or should that be a symbol?).
There are plenty of colours.
Slightly relatedly: a game based on the _His Dark Materials_ trilogy
(_Northern Lights_, _The Subtle Knife_, _The Amber Spyglass_) or just a
very twisted game, could have:
* angelic == evil,
* demonic == good.
--
Simon Richard Clarkstone: s.r.cl?rkst?n?@durham.ac.uk/s?m?n.cl?rkst?n?@
hotmail.com ### "I have a spelling chequer / it came with my PC /
it plainly marks for my revue / Mistake's I cannot sea" ...
by: John Brophy (at: http://www.cfwf.ca/farmj/fjjun96/)
Disclaimer about ethical/moral position:
I am an INTP/INTJ atheist, with liberal-left politics (Eco: -7 , Soc:
-3.5 on the Political Compass), and with my own strange "religious"
beliefs that simultaneously eliminate Creator(s) and time-travel paradoxes.
Note: AFAICT, "ethics" is the correct word to use here, though various
websites disagree.
General Idea:
=============
Due to reading a document (which I now can't find) about the problems
with the traditional "good" vs. "evil" systems in some games (e.g.
_Black and White_), I wish to ask the group about a possibility I have
not yet seen anywhere else: ethics inference. The "good"/"evil" is
inferred by the EIE (Ethical Inference Engine). Inference is used in
the technical sense of finding a general pattern from specific examples.
The idea is not to have many restrictions (if any) on what you are
allowed to do (e.g. eating when already full, robbing corpses, killing
civilians, etc.), and not to have penalties for breaking moral rules,
but to let the player act how they like and have the game infer whether
they are "good" or "evil". For balance, both good and evil must have
benefits and drawbacks.
The EIE's inference of "good" and "evil" could be used to give a player
holy or unholy powers (e.g. bonuses for some types of spells, effects of
intrinsicly holy/unholy objects/places, reactions of some beings, or
even the plot of the game (heaven/hell).
One possibility is that "evil" should give short-term profit and
long-term pain, with "good" the opposite, rather than each being equally
easy -- this could be seen as pushing a moral agenda (maybe you want
to). If others' treatment of you is determined by how you treat them,
"evil" may quite naturally end up as a very difficult pit to get out of.
Thoughts On A Specific Game:
============================
This whole idea was partly inspired by a game idea that is kicking
around in my head where the player is a Kigdatsi (a
genetically-engineered highly intelligent dragon-like creature) from a
parallel universe waking up in our universe badly injured in a zoo (or
similar). It could be made fairly clear in the intro text that humans
are to blame for your current situation, which would give you ample
reason to be annoyed. (Has anyone seen anything like this before?)
You (the player) could co-operate with most humans, only kill to defend,
and minimise their suffering, *or* you could slaughter everyone (even
easy early on, as zoo visitors are not normally armed :->), care only
for yourself, and let people die painfully (you are large enough to
swallow people whole). Although you would recover rather faster if you
eat lots of humans, they will try harder to kill you (by sending the
army in), and your own species could despise you for it. The game would
need to do long-term monitoring of how you treat humans (and animals?)
and change their behaviour to fit.
Inferences From The Specific Game:
==================================
In fact, a linear "good"-"evil" system is not the best way to express
ethics. Some topics (e.g. euthanasia, religion, judgement of others,
homosexuality (but how would that appear in a RL?)) will be judged
differently by game-makers and game-players, so we need a more objective
measurement system, maybe one of:
http://en.wikipedia.org/wiki/Political_spectrum
(These are party policies, and not moral opinions, though. Economics
might be a hard one to infer :-> .) However, you will end up with
dozens of moral axes if you try to be too complete, so some
simplification will be necessary.
Some axes that could be inferred (many of these overlap):
* authoritarian -- anti-authoritarian (?)
* pro-euthanasia -- anti-euthanasia (?)
* painless killing -- painful killing
* merciful -- sadistic
* friendly -- aggressive
* run -- fight
* offensive -- defensive
* avoid danger -- destroy danger
* terrorising (scaring people for fun) -- non-terrorising (avoid
scaring people)
* honest -- tricky (e.g. feigning unconsciousness)
* survivalist -- bloodthirsty (e.g. is there any *point* invading a
kindergarten (Brit: "primary school") other than to kill lost of
innocent children?)
A useful system (for the above game at least) could be to have the
colour of beings indicating their mental state(s) e.g.:
Normal, Friendly, Aggressive, Scared, Terrified, Unconscious
as well as physical state e.g.:
Uninjured, Injured, Dying, Dead (or should that be a symbol?).
There are plenty of colours.
Slightly relatedly: a game based on the _His Dark Materials_ trilogy
(_Northern Lights_, _The Subtle Knife_, _The Amber Spyglass_) or just a
very twisted game, could have:
* angelic == evil,
* demonic == good.
--
Simon Richard Clarkstone: s.r.cl?rkst?n?@durham.ac.uk/s?m?n.cl?rkst?n?@
hotmail.com ### "I have a spelling chequer / it came with my PC /
it plainly marks for my revue / Mistake's I cannot sea" ...
by: John Brophy (at: http://www.cfwf.ca/farmj/fjjun96/)
yt E+++ T-- R+ P+++ D+ G+ F:*band !RL RLA-
Facebook
Twitter
Instagram