Discussion:
[MG] morality in descriptive sense
Michal Štěpánek
2017-03-31 16:41:31 UTC
Permalink
Hi!

Let us assume...(still talking about "prediction of goodness")

1. Good result is honest positive democratic feedback. (morality in
descriptive sense)
2. Good system creates good results.
3. Good processes makes good system.
4. Discussion-decision-implementation-effects is elementary process of a
system.

but the 1. is problematic.
It may work fine at municipal level.
It will fail at some moments imho, foreign policy might be a good example.
So I started searching for other definitions of "good result".
I have not found for foreign policy, but for birth rate. Let us accept 1
but we should also add an condition imho and say that starvation must be
zero to accept the feedback. If it is not zero, I would say that the
processes ( those like in 4) influencing birth rate we not good. I mean
bad result, no matter what democracy says.

You may see it as a small flaw, but I think it is something bigger. For
most common human flaws we do need to compensate and we will need to find
other definitions of "good results".
This means different different equations for different areas.
The compensation must happen at the field of science. Such compensations
may be impossible in running political system.

Please do not confuse me with Scott. As I understand it Scott is designing
one system that compensates all flaws. I design mechanisms for evaluation
of systems. And I try to pair problematiques with systems.

Cheers
m.
--
michalstepanek.github.io
Scott Raney
2017-04-02 23:55:36 UTC
Permalink
Post by Michal Štěpánek
Let us assume...(still talking about "prediction of goodness")
1. Good result is honest positive democratic feedback. (morality in
descriptive sense)
That's backing off a lot on your first proposals (which seemed to set
a goal that required some sort of objective measure), but I agree that
this is sufficient.
Post by Michal Štěpánek
2. Good system creates good results.
I would propose this is *by definition*, and again not by any sort of
post-hoc analysis of the actual outcome. The latter I'd argue is
impossible to achieve (expectations change, externalities affect the
results, etc.)
Post by Michal Štěpánek
3. Good processes makes good system.
A little more tenuous, but OK.
Post by Michal Štěpánek
4. Discussion-decision-implementation-effects is elementary process of a
system.
Agreed.
Post by Michal Štěpánek
but the 1. is problematic.
It may work fine at municipal level.
It will fail at some moments imho, foreign policy might be a good example.
You probably know my solution to this particular problem: Eliminate
the need for "foreign policy" by ensuring that *all* decisions are
made using the same system.
Post by Michal Štěpánek
So I started searching for other definitions of "good result".
I have not found for foreign policy, but for birth rate. Let us accept 1 but
we should also add an condition imho and say that starvation must be zero to
accept the feedback. If it is not zero, I would say that the processes (
those like in 4) influencing birth rate we not good. I mean bad result, no
matter what democracy says.
Again, I'd recommend staying away from "birth rate" or population as a
domain for measuring "goodness". I've done a *lot* of research and
thinking about these issues and have come to the conclusion that it's
impossible to make specific predictions about how social engineering
in this area will work. Although my gut tells me that overall a slow
reduction is the best plan, at this point I've resigned myself to
defer to the need for individual freedom to "automagically" solve this
problem (e.g., studies show that including birth control in health
care systems substantially reduces birth rates).
Post by Michal Štěpánek
You may see it as a small flaw, but I think it is something bigger. For most
common human flaws we do need to compensate and we will need to find other
definitions of "good results".
This means different different equations for different areas.
The compensation must happen at the field of science. Such compensations may
be impossible in running political system.
Agreed, which is why I've decided to punt on this whole issue ;-)
Post by Michal Štěpánek
Please do not confuse me with Scott. As I understand it Scott is designing
one system that compensates all flaws. I design mechanisms for evaluation of
systems. And I try to pair problematiques with systems.
Sounds fragmentary and therefore probably not viable: Better a system
that imperfectly solves all problems than one that lets key problems
fall through the cracks because there isn't a "system" design that
solves it really well. Cherrypicking problems is only going to cause
you (and maybe the rest of us) grief in the long run.
Regards,
Scott

PS: Getting close to public testing on the Matchism System app! Warm
up those mice because I hope to do the first alpha-test release in a
week or two and hope that you'll all be able to participate.
Post by Michal Štěpánek
Cheers
m.
--
michalstepanek.github.io
Michal Štěpánek
2017-04-04 13:59:46 UTC
Permalink
(snip)
Post by Scott Raney
Post by Michal Štěpánek
2. Good system creates good results.
I would propose this is *by definition*, and again not by any sort of
post-hoc analysis of the actual outcome. The latter I'd argue is
impossible to achieve (expectations change, externalities affect the
results, etc.)
Thank you :)
So, do you want to say, that if people mess it by themselves it is not so
bad. To some degree yes, but imho...
Decision makers should expect "expectation change and externalities". I do
not care that thay cannot know, they should get better at guessing.
I still say ...good system creates good results.
Post by Scott Raney
Post by Michal Štěpánek
3. Good processes makes good system.
A little more tenuous, but OK.
Post by Michal Štěpánek
4. Discussion-decision-implementation-effects is elementary process of a
system.
Agreed.
Post by Michal Štěpánek
but the 1. is problematic.
It may work fine at municipal level.
It will fail at some moments imho, foreign policy might be a good
example.
You probably know my solution to this particular problem: Eliminate
the need for "foreign policy" by ensuring that *all* decisions are
made using the same system.
I would say this is one of your weak spots.
Post by Scott Raney
Post by Michal Štěpánek
So I started searching for other definitions of "good result".
I have not found for foreign policy, but for birth rate. Let us accept 1
but
Post by Michal Štěpánek
we should also add an condition imho and say that starvation must be
zero to
Post by Michal Štěpánek
accept the feedback. If it is not zero, I would say that the processes (
those like in 4) influencing birth rate we not good. I mean bad result,
no
Post by Michal Štěpánek
matter what democracy says.
Again, I'd recommend staying away from "birth rate" or population as a
domain for measuring "goodness". I've done a *lot* of research and
thinking about these issues and have come to the conclusion that it's
impossible to make specific predictions about how social engineering
in this area will work. Although my gut tells me that overall a slow
reduction is the best plan, at this point I've resigned myself to
defer to the need for individual freedom to "automagically" solve this
problem (e.g., studies show that including birth control in health
care systems substantially reduces birth rates).
Well, I also hope we are clever enough to know how many children we should
have, but what if not?
I better dare to have the same gut as you, and I hope I will be
scientifically "ridiculed" if I am wrong.
Post by Scott Raney
Post by Michal Štěpánek
You may see it as a small flaw, but I think it is something bigger. For
most
Post by Michal Štěpánek
common human flaws we do need to compensate and we will need to find
other
Post by Michal Štěpánek
definitions of "good results".
This means different different equations for different areas.
The compensation must happen at the field of science. Such compensations
may
Post by Michal Štěpánek
be impossible in running political system.
Agreed, which is why I've decided to punt on this whole issue ;-)
Post by Michal Štěpánek
Please do not confuse me with Scott. As I understand it Scott is
designing
Post by Michal Štěpánek
one system that compensates all flaws. I design mechanisms for
evaluation of
Post by Michal Štěpánek
systems. And I try to pair problematiques with systems.
Sounds fragmentary and therefore probably not viable: Better a system
that imperfectly solves all problems than one that lets key problems
fall through the cracks because there isn't a "system" design that
solves it really well. Cherrypicking problems is only going to cause
you (and maybe the rest of us) grief in the long run.
I wouldn't say that it is cherrypicking, democratic feedback is the basic
mechanism for selection of a system. Sometimes is the feedback flawed and
that is what I am writing about - some other criterions than feedback need
to be invented too.

As for the "fragmentary". There might be a lot of systems (as in current
states), but there is always at least one that solves them best, even if it
is system that gets 20%( it may be still the best if no other gets over 20%
in the problematique). I do not see the cracks.

Cheers
m.
Post by Scott Raney
Regards,
Scott
PS: Getting close to public testing on the Matchism System app! Warm
up those mice because I hope to do the first alpha-test release in a
week or two and hope that you'll all be able to participate.
Post by Michal Štěpánek
Cheers
m.
--
michalstepanek.github.io
_______________________________________________
Start : a mailing list of the Metagovernment project
http://www.metagovernment.org/
Manage subscription: http://metagovernment.org/mailman/listinfo/start_
metagovernment.org
--
michalstepanek.github.io
Scott Raney
2017-04-04 15:55:33 UTC
Permalink
On Tue, Apr 4, 2017 at 7:59 AM, Michal Štěpánek <***@gmail.com> wrote:

(another big snip)
Post by Michal Štěpánek
Post by Scott Raney
You probably know my solution to this particular problem: Eliminate
the need for "foreign policy" by ensuring that *all* decisions are
made using the same system.
I would say this is one of your weak spots.
I would agree to characterizing it as "ambitious", perhaps even overly
so. But it definitely isn't a "weak spot", and indeed I'd argue that
the one-government plan is the *only* reliable solution to a great
many problems.
Post by Michal Štěpánek
Well, I also hope we are clever enough to know how many children we should
have, but what if not?
Speaking of "overly ambitious"! As part of my research on this for the
Matchism Manifesto I came across this enlightening statistic: In the
US *half* of all pregnancies are unplanned. And while you might be
inclined to toss off "abortion" as the automatic solution to this
problem, I'd advise against being so cavalier about what in most cases
is a gut-wrenching decision.
Post by Michal Štěpánek
I better dare to have the same gut as you, and I hope I will be
scientifically "ridiculed" if I am wrong.
It's not the scientists or scientifically-minded you have to worry about here...
Post by Michal Štěpánek
I wouldn't say that it is cherrypicking, democratic feedback is the basic
mechanism for selection of a system. Sometimes is the feedback flawed and
that is what I am writing about - some other criterions than feedback need
to be invented too.
No, I meant cherrypicking your problem domains (albeit reproduction
rates being one of the harder cases). Whatever metrics for "goodness"
you propose should be applicable to *all* public policy issues, not
just a few specific areas.
Post by Michal Štěpánek
As for the "fragmentary". There might be a lot of systems (as in current
states), but there is always at least one that solves them best, even if it
is system that gets 20%( it may be still the best if no other gets over 20%
in the problematique). I do not see the cracks.
So each state crossed with each problem domain can have a unique
"most-good" system for it? You may not see the "cracks" but it looks
like a jigsaw puzzle with a lot of pieces missing to me...
Regards,
Scott
Michal Štěpánek
2017-04-20 16:52:07 UTC
Permalink
(snip all, but topic remains)

Hi,

is society good at morality? I would like to know if there is any research,
when public is asked an obvious moral question and is statistically failing
to answer it "right", just for the reason of some fatal moral flaw in genes
or culture.
Something like asking: "Do you like people to suffer?" and there would be
to many percents of "yes".

I would say that poeple are good at morality even if they know little about
ethics. What do you think? Do you agree?

Cheers
m.
Scott Raney
2017-04-20 18:26:49 UTC
Permalink
Post by Michal Štěpánek
is society good at morality?
I would say "society" doesn't do morality at all: It's an individual
endeavor. It's worth noting, however, that the propensity to make
moral judgments is not uniformly distributed in the population (yeah,
it's authoritarians that mostly feel compelled to do this).
Post by Michal Štěpánek
I would like to know if there is any research,
when public is asked an obvious moral question and is statistically failing
to answer it "right", just for the reason of some fatal moral flaw in genes
or culture.
There is no "right" answer when it comes to moral judgments: It
totally depends on the environment the individual was raised in and on
the context in which the judgments are made. I think this is made
pretty clear in the Matchism manifesto
(http://www.matchism.org/internal-moral-codes/) and I've yet to see
anyone make a coherent argument otherwise.
Post by Michal Štěpánek
Something like asking: "Do you like people to suffer?" and there would be to
many percents of "yes".
And many would also say "I don't care one way or the other". But I
think mental illness is a separate issue here: The vast majority of
people do have a functioning morality system. Unfortunately it is very
difficult to determine how it is structured for any given individual,
and nearly impossible to change it past a certain age unless the
individual happens to be able to reason above Kohlberg stage 4 (and
most people can't).
Post by Michal Štěpánek
I would say that poeple are good at morality even if they know little about
ethics. What do you think? Do you agree?
There are two separate issues here: Whether an individual's moral code
is already functioning at t + 1, and whether they can function above
Kohlberg stage 4 and therefore change their moral code if not. The
answer for both of those questions is "no" for most people, although
there's at least hope for the latter group. Maybe a more fruitful path
for you would be to determine if people can be educated to function
above stage 4. My guess is probably not, but I wouldn't rule out the
possibility, especially if training started very young (at least
before age 7).
Regards,
Scott
Post by Michal Štěpánek
Cheers
m.
_______________________________________________
Start : a mailing list of the Metagovernment project
http://www.metagovernment.org/
http://metagovernment.org/mailman/listinfo/start_metagovernment.org
Michal Štěpánek
2017-04-21 12:11:27 UTC
Permalink
Post by Scott Raney
Post by Michal Štěpánek
is society good at morality?
ok, aggregated morality
Post by Scott Raney
It's an individual
endeavor. It's worth noting, however, that the propensity to make
moral judgments is not uniformly distributed in the population (yeah,
it's authoritarians that mostly feel compelled to do this).
good
Post by Scott Raney
Post by Michal Štěpánek
I would like to know if there is any research,
when public is asked an obvious moral question and is statistically
failing
Post by Michal Štěpánek
to answer it "right", just for the reason of some fatal moral flaw in
genes
Post by Michal Štěpánek
or culture.
true, you may use your judgement here, no problem
Post by Scott Raney
It
totally depends on the environment the individual was raised in and on
the context in which the judgments are made. I think this is made
pretty clear in the Matchism manifesto
(http://www.matchism.org/internal-moral-codes/) and I've yet to see
anyone make a coherent argument otherwise.
Sure it depends, but some conditions are more common as we are all humans
living on Earth.
Post by Scott Raney
Post by Michal Štěpánek
Something like asking: "Do you like people to suffer?" and there would
be to
Post by Michal Štěpánek
many percents of "yes".
And many would also say "I don't care one way or the other". But I
think mental illness is a separate issue here: The vast majority of
people do have a functioning morality system.
good, that is partly I was asking about.
Post by Scott Raney
Unfortunately it is very
difficult to determine how it is structured for any given individual,
no need to know the structure, results of their algorythm (thin evaluative
concept) are enough for me.
Post by Scott Raney
and nearly impossible to change it past a certain age unless the
individual happens to be able to reason above Kohlberg stage 4 (and
most people can't).
ok
Post by Scott Raney
Post by Michal Štěpánek
I would say that poeple are good at morality even if they know little
about
Post by Michal Štěpánek
ethics. What do you think? Do you agree?
There are two separate issues here: Whether an individual's moral code
is already functioning at t + 1, and whether they can function above
Kohlberg stage 4 and therefore change their moral code if not. The
answer for both of those questions is "no" for most people, although
there's at least hope for the latter group. Maybe a more fruitful path
for you would be to determine if people can be educated to function
above stage 4. My guess is probably not, but I wouldn't rule out the
possibility, especially if training started very young (at least
before age 7).
Ok, individually their do big mistakes. But are there mistakes (use your
opinion here) in their aggregate ex post judgement (numerical aggregation
not a mob). And are these mistakes statistically significant or is it safe
to use the aggregation as approximation of normativity?

Regards
m.
Post by Scott Raney
Regards,
Scott
Post by Michal Štěpánek
Cheers
m.
_______________________________________________
Start : a mailing list of the Metagovernment project
http://www.metagovernment.org/
http://metagovernment.org/mailman/listinfo/start_metagovernment.org
_______________________________________________
Start : a mailing list of the Metagovernment project
http://www.metagovernment.org/
Manage subscription: http://metagovernment.org/mailman/listinfo/start_
metagovernment.org
--
michalstepanek.github.io
Scott Raney
2017-04-22 16:15:49 UTC
Permalink
On Fri, Apr 21, 2017 at 6:11 AM, Michal Štěpánek <***@gmail.com> wrote:

(snip)
Post by Michal Štěpánek
Post by Scott Raney
Post by Michal Štěpánek
I would say that poeple are good at morality even if they know little about
ethics. What do you think? Do you agree?
There are two separate issues here: Whether an individual's moral code
is already functioning at t + 1, and whether they can function above
Kohlberg stage 4 and therefore change their moral code if not. The
answer for both of those questions is "no" for most people, although
there's at least hope for the latter group. Maybe a more fruitful path
for you would be to determine if people can be educated to function
above stage 4. My guess is probably not, but I wouldn't rule out the
possibility, especially if training started very young (at least
before age 7).
Ok, individually their do big mistakes. But are there mistakes (use your
opinion here) in their aggregate ex post judgement (numerical aggregation
not a mob). And are these mistakes statistically significant or is it safe
to use the aggregation as approximation of normativity?
IMHO definitely not, at least not if what you're measuring are
outcomes rather than making some sort of objective measurements (i.e.,
put every individual in the population in an fMRI scanner and see what
they *really* think/feel). If the latter were translated directly into
public policy decisions you'd end up with the Matchism-prescribed
society (albeit because of the aggregation step it would be a kind of
majority-rules kind of place which some people on this list would find
unacceptable, at least on a philosophical basis if not on a practical
one).

But the problem with our current decisionmaking and government systems
is that not only are they poor at performing the measurements you
propose (i.e., most people don't even vote, and even those that do
often make logical errors even when they intend to vote their morals,
not to mention the issue of them not even being offered the choices
that would truly match their morals or their rational analysis), but
also that most decisionmaking is done outside of even that poor
measurement system (i.e., by SDAPs in the judicial and executive
branches).

Again, I'm not sure where you're going with this, but I'm pretty sure
you lack the technology you need to get there. Proxyfor.me might be a
part of what you need, but even it only relies on sampling instead of
aggregation and it also has *overriding* morals with rational analysis
whenever possible as one of its design goals. A system that tries to
use moral codes directly for decisionmaking I can only think would
quickly degenerate into a true mob-rule society (e.g, 50% of the
population would become activated authoritarian, the very definition
of a "mob").
Regards,
Scott
Post by Michal Štěpánek
Regards
m.
Loading...