►
From YouTube: W9 Labs WG: Rewards Research Panel and Labs Kickoff!
Description
This is the first week of the rewards research kick off! Starting with a panel discussion hosted by the governauts, and into the TEC lab kickoff for rewards research. This is expected to be a six week series in the labs.
🙏 Thank you for watching! Hit 👍 and subscribe 🚩 to support this work
🌱Join the Community🌱
on Discord https://discord.gg/uM4ZWDjNfK
or say hello on Telegram https://t.me/tecommons
Join the conversation https://forum.tecommons.org/
Follow us on Twitter: https://twitter.com/tecmns
Learn more http://tecommons.org/
A
Thank
you
all
for
joining,
and
I
would
just
like
to
pass
to
each
one
of
you
for
introducing
yourselves
and
your
projects,
and
you
have
five
minutes
each
to
this
to
this
introduction,
and
we
have
a
few
questions
that
we're
gonna
get
into,
but
also
everyone
that
is
present
is
very
welcome
to
submit
questions
in
the
chat
and
we
can
pull
it
from
there
and
and
ask
participants.
So
if
you
want
to
ask
a
question,
just
put
the
question
in
who
do
you?
Who
do
you
want
it
to
answer?
A
So
I
will
pass
to
griff
to
start
introduction
and
and
then
we'll
move
forward.
C
Sure
so
I'm
griff
hi
I've.
I
work
with
giveth
the
token
engineering
commons,
common
stack,
dab,
node
and
a
whole
bunch
of
other
projects
too,
but
those
are
where
I
spend
most
of
my
time
and
I
I've
been
looking
at
reward
systems
for
the
last.
I
don't
know
five
years
since
really
since
the
dow
days,
but
there
was
like
just
like
we
were
just
about
to
talk
about
how
we're
going
to
create
rewards
in
the
community,
but
that
didn't
last
long,
so
we
never
really
went
very
far
with
that.
C
But
then
we
carried
that
in
to
give
it
because
give
us
started
right
after
the
dow
and
we
were
looking
at.
How
do
we
reward
volunteers?
You
know,
and
the
whole
purpose
of
giveth
is
like
how
do
we
make
build
a
con?
The
purpose
of
giveth
at
the
time
was
was
kind
of
what
the
purpose
of
common
stack
is
now
it's
just
to
like
build
economies
around
causes
and
and
create
a
reward
system
for
the
work
that's
being
done,
because
so
much
good
work
is
being
done
by
charities.
Why
aren't
they
rewarded?
C
That's
just
silly
right
and,
and
so
rewarding
volunteers
was
an
important
part.
So
we
started
this
thing
called
reward
dial
and
we,
the
reward
dao
had
probably
four
or
five
iterations
trying
different
things,
and
while
I
was
doing
giveth
and
working
on
those
things,
I
also
run.
I
run
a
couple
burning
man,
camps,
decentral
and
doge
central,
and
there
are
lots
of
praise
experiments
in
in
decentral.
In
fact,
mitch
mitch
knows
all
of
them.
He
he
comes
to
my
camp
he's.
C
Our
head
chef
is
also
a
a
web
3
superstar,
but
we
we
do
praise
and
shame
which
was
interesting
right,
although
one
year
we
didn't
do
shame
but
the
what
we
learned
at
burning
man
or
what
I
I
felt
really
good
at
burning
man
where
we
would
praise
without
quantification.
It
was
just
straight
up
acknowledgement,
so
we'd
have
acknowledgement
for
praise
and
then
we'd
also
have,
but
there
would
be
no
quantification
and
everything
was
all
lovey-dovey
and
we'd
praise
people.
C
You
know
for
their
actions
and
then,
when
we
introduced
shame
we'd
only
shame
behaviors.
We
wouldn't
actually
shame
people
right
so
we'd
be
like.
Oh
there
I
found
a
hammer
out
on
the
ground
in
the
playa.
Everyone
knows
it's
supposed
to
be
in
the
tool.
Corral,
shame
shame,
shame
right.
It
was
really
fun
and
the
same
with
praise.
It's
like.
Oh
my
gosh.
You
know
mitch
made
such
an
awesome
meal.
C
How
about
we
all
praise
mitch,
mitch,
praise
praise
praise
and
it
was
really
fun,
and
this
acknowledgement
you
know
without
the
quantification
is-
is
half
of
the
battle.
It's
half
of
the
fun.
I
would
say,
with
these
reward
systems,
is
that
the
non-monetary
qualitative
acknowledgement,
but
then
hey
we're
gonna
build
web
35.
D
C
Kind
of
building
systems-
and
we
want
to
create
tokens
so
now
we
have
to
quantify
that
qualitative
love
and
that's
where
things
get
messy.
I
have
to
say
it's
like
this
is
how
the
sausage
is
made
right
and
then
we
have
to
figure
out
some
sort
of
way
to
quantify
that
qualitative
stuff.
So
we
we
do
it
with
with
the
tec.
C
The
tc
we've
done
some
really
cool
work
around
praise,
and
then
we
had
a
lot
of
research
around
it
that
I'm
sure
we'll
get
into
later
and
giveth
is
trying
to
giveth
and
tec
are
now
integrating
source
cred
and
for
the
prey
system
to
try
to
quantify
what
can
be
quantified
and
qualified
and
use
praise
for
the
qualitative
stuff.
That's
hard
to
quantify.
D
Hey
all
I'm
hamad,
also
known
as
metadreamer.
D
I've
been
working
in
the
dao
space
for
almost
two
years
now,
yeah
I've
been
super
interested
in
sort
of
better
dow,
tooling
and
mechanisms
in
which
we
can
help
to
like
make
decisions,
endow
and
get
people
rewarded.
D
I've
been
a
core
contributor
to
source
cred
for
about
a
year
or
just
over
a
year
now,
and
I'm
a
member
of
meta
cartel
medicare
dollar
ventures,
metagame
and
meta
factory
kind
of
bouncing
around
those
dials
and
you
know
finding
the
common
patterns,
and
you
know,
building
tooling
and
experiments
to
figure
out
how
to
better
like
measure
and
reward
value
creation
yeah.
D
A
D
Yeah
sure
so
the
goal
of
medicred
is
basically
to
start
to
capture
and
aggregate
all
this
like
information
on
value
being
created
and
work
being
done,
not
just
within
a
dao
but
between
daos,
because
a
lot
of
the
value
created
in
this
ecosystem
is
not
just
for
for
one
organization.
But
you
know
it
affects
the
whole
community
and
the
whole
ecosystem.
D
So
I
think
you
know
we
had
this
vision
of
like
the
dao
back
in
the
day,
but
I
think
that
vision
is
starting
to
like
manifest
now
and
not
as
like
one
singular
dial,
but
like
the
collective
sum
of
all
the
dows
in
the
community,
but
we
don't
have
as
much
connectivity
between
the
dows
themselves.
So
the
goal
of
meta
credit
is
to
be
a
layer
on
top
of
source
credit.
On
top
of
praise.
D
On
top
of
you
know,
all
these
different
rich
data
points
that
we
have
on
people
and
the
the
work
they've
done
in
daos
and
be
able
to
like
attach
this
to
our
identities
and
our
reputations,
and
you
know
explore
what
that
looks
like,
because
you
know
there's
a
lot
of
considerations
there
when
it
comes
to
like
privacy-
and
you
know
you
know,
there's
a
a
lot
of
moving
parts,
so
yeah
medicare
is
kind
of
like
a
shelling
point
to
to
get
people
who
are
thinking
about
these
problems
to
to
share
and
flow
information
resources
and
data
between
each
other.
D
And
hopefully
you
know
even
creating
markets
around
this.
D
So
we
can
start
to
flow
capital,
not
only
where
you
know
it's
like
at
the
intersection
of
monetary
transactions,
but
we
can
quantify
the
value
of
you,
know,
opinions
and
statements
and
facts
and
use
cryptography
for
not
only
cryptocurrencies,
but
you
know
crypto
data
and
identity
and
reputation
and
all
this-
and
I
think,
that's
really
like
the
next
major
step
into
you
know
taking
dowels
from
like
validation
like
phase
and
you
know,
experimentation
phase
into
like
you
know,
production
phase
and
you
know
really
allowing
them
to
realize
their
full
potential.
E
Thanks
and
good
to
meet
y'all
I
am
indeed
zach.
E
I
am
one
of
the
core
instigators,
I
guess
at
coordinate,
and
I
also
work
with
yearn,
I'm
a
contributor
at
urine
as
well,
and
then,
of
course,
like
many
of
us,
I'm
sure
have
a
lot
of
other
dows
that
I
kind
of
play
in
I'm
new
to
this
space.
I'm
almost
always
the
newest
person
in
in
crypto.
E
In
every
conversation,
I'm
in
I
started
last
year
and
and
before
that,
so
I
I
had
a
consultancy
that
we
started
about
eight
years
ago
called
converge
and
it
was
focused
on
how
to
build
human
networks
to
solve
social
environmental
impact,
so
basically
have
different
issues
like
landscape
conservation
or
income
inequality
or
migrant
rights,
and
bringing
together
government
and
corporate
and
non-profit
leaders
together
and
sort
of
building
them
into
now.
E
What
I
would
call
like
a
an
irl
dao,
and
so
by
doing
lots
of
those
different
projects
all
over
the
world,
learned
a
lot
about
coordination
and
how
humans,
like
actually
work
together
and
how
they
actually
balance
different
incentives,
because
almost
everyone,
that's
part
of
a
network,
has
their
day
job
that
they're
doing
too
and
so
they're,
always
balancing
all
these
competing
rewards.
E
And
so,
when
I
discovered
daos,
it
was
kind
of
a
big
moment,
and
I
got
obsessed
and
I
still
am
and
so
coordinate.
We
started
originally
kind
of
incubated
solving
a
problem
within
yearn,
which
was
at
urine.
They
had
this
grants
committee
and
they
would
get
petitions
for
grants
every
month
for
people
that
were
contributing.
You
know,
monitoring
the
discord
or
translating
docs
or
things
like
that.
E
What
they
kind
of
found
is
that
you
know
people
were
applying
for
grants
and,
and
it
was
kind
of
an
arbitrary
you
know,
process
they'd
just
be
like.
I
want
this
much
for
doing
this
and
then
some
people
didn't
apply
for
grants
that
added
a
lot
of
value
and
they
were
having
to
make
these
sort
of
apples
to
oranges.
Decisions
about
who
should
get
a
grant
and
why
and
then
they're
like.
E
Why
are
we
even
on
this
committee
like
who,
who
appointed
us
here
and
so
ultimately,
we
said
well
what
if
we
just
ask
ask
the
network
like
where
the
value
is
being
generated
and
how
the
grant
should
be
distributed,
and
so
we
created
coordinate,
which
essentially
a
system
where
each
person
in
in
a
circle
right
that's
contributing
in
whatever
sort
of
level
of
task
you're
talking
about,
but
people
get
together
and
can
distribute
value
amongst
themselves.
E
So
they
use
on
our
platform,
like
essentially
poker
chips
that
are
called
give
and
they
pass
them
to
each
other
and
then,
whatever
percentage
of
give
they
get.
That
is
the
percentage
of
the
the
budget
or
allocation
resource
that
they
get,
and
so
we've
been.
A
Thank
you
yeah
and
congratulations.
I
saw
that
coordinate
was
the
most
cited
reward
system
in
in
the
bitcoin
survey.
So
that's
pretty
cool.
E
Thank
you
I
I
am.
I
feel
compelled,
though,
to
like
asterisk
that
data
slightly.
I
think,
because
bank
list
is
such
a
huge
coordinate
circle,
but
I
think
it's
it's
not
entirely
representative
data,
but
still
it
was.
It
was
great
to
see.
F
Hi
y'all
I'm
andrew
penland.
I
work
in
the
mathematics
and
computer
science
department
at
western
carolina
university.
My
story
in
math
is
kind
of
important
to
know.
I
was
very
not
a
traditional
math
person
all
through
school
and
I
sort
of
came
in
cryptocurrency
through
an
interest
in
decentralized
governance
and
alternative
currencies.
F
I
have
experience
with
reward
systems
because,
as
a
professor
now
I
have
to
think
about
how
to
grade
the
class
and
even
before
I
got
interested
in
this
specific
topic,
I
thought
a
lot
about
what
was
fair
about
grading
and
what
was
important,
intrinsic
motivation
of
learning
versus
extrinsic
motivation
of
learning.
F
I
got
interested
in
reward
systems
and
cryptocurrency
completely
by
accident.
There
was
an
interesting
conversation
happening
in
the
t
e
c
one
day
and
I
just
kind
of
jumped
on.
But
what
was
happening
during
that
discussion
is
they
were
discussing
how
praise
was
being
mapped
to
something
called
impact
hours
and
then
how
those
impact
hours
would
be
mapped
to
the
tec
tokens.
F
So
I
just
kind
of
volunteered
to
think
about
that
question.
I
didn't
really
know
what
I
was
getting
myself
into
at
the
time.
It
was
a
lot
of
fun.
I
learned
a
lot
and
I
think
my
what
I
was
able
to
bring
to
it
was
precisely
what
was
mentioned
a
moment
ago
about
how
to
quantify
these
things
in
a
measurable
way.
F
Now
that
has
its
drawbacks.
But
I
read
a
lot
of
the
literature,
that's
known
in
cryptocurrency,
about
how
resource
distribution
is
handled
and
some
of
the
tools
that
are
used.
Things
like
the
genie
coefficient,
the
shannon
entropy
the
nakamoto
coefficient
and
I'm
not
trying
to
paper
bomb
you,
but
if
you're
curious
about
any
of
the
things
here
are
some
good
starting
places
and
it
turns
out.
F
Ironically,
just
as
we
were
wrapping
up
this-
and
I
just
found
this
paper
this
morning,
the
vitalik
has
written
a
blog
post
about
what
the
right
ways
to
measure
inequality
in
token
distributions
are,
and
his
blog
article
ironically,
is
called
against
overuse
of
the
genie
coefficient,
so
I've
skimmed
that
reading
it.
It's
very
interesting,
but
that's
my
perspective.
A
Thank
you,
yeah
thanks
everyone
for
the
introductions,
and
I
wanted
to
bring
you
all
together
here
and
also
have
andrew
present,
because
I
feel
like
a
lot
of
the
challenges
that
we
are
facing
is
how
to
quantify
things:
how
to
how
to
transfer
subjective
data
to
objective
data.
How
to
find
the
right
metrics
for
the
things
that
we're
measuring
and
are
those
metrics
are.
A
Can
this
matrix
tell
us
something
about
the
health
of
the
ecosystem
so
I'll
pass
around
this
first
question:
that
is,
for
everyone
again,
but
for
for
this
to
open
a
little
more
discussion
and
for
everyone
to
also
based
on
these
challenges,
to
write
on
the
on
the
chat,
any
questions
that
you
might
have.
So
what
are
the
the
biggest
questions
that
you
see
with
either
with
reward
systems
in
general
or
with
the
reward
system
that
you
work
with
so
I'll
pass
to
hamad
to
start.
D
Sure
yeah,
so
I
think
the
the
question
of
like
objective
you
know
truths
is
is
an
important
one
and
you
know
I
think
it's
easy
to
sort
of
come
to
like
a
consensus
on
what's
like,
objectively
valuable
or
should
be
rewarded
or
how
much
you
should
be
rewarded
within,
like
you
know,
smaller
groups
of
people,
maybe
up
to
like
you,
know,
15
20
max
after
that
it
starts
to
like
fall
apart
and
different
people
have
different
perceptions.
D
So
I
think,
in
terms
of
like
scaling
out
these
reward
systems
beyond
just
like
smaller
groups
of
individuals,
we
we
need
a
way
to
capture
like
inter
subjectivity,
so
you
know,
I
think
it's
it's
really
impossible
to
try
to
come
towards
like
global
consensus
on
some
sort
of
like
you
know,
let's,
let's
all
vote
on
how
what
should
be
rewarded
where
instead,
I
think
it's
much
better
to
like
come
to
consensus
on
a
process
and
then
like
just
deploy
that
process
and
then,
if
that
process
has
issues,
then
you
can
vote
or
you
know,
coordinate
and
come
to
consensus
on
how
to
change
that
process.
D
But
you
know
the
the
process
should
be
a
way
or
a
mechanism
in
which
we
can
quantify
individual
beliefs
and
statements
and
values
and
then
be
able
to
like
sum
across
multiple
of
them
in
either
like
subgroups
or
you
know,
even
in
a
nested
format,
so
coordinates
a
good
example
of
this.
Where
you,
you
know
we're
not
trying
to
come
up
with
like
an
allocation
from
the
top
down
level.
D
Instead,
we
let
people,
you
know
gift
each
other,
give
on
from
a
bottom-up
level,
and
then
you
know
the
the
sum
of
all
the
gifting
of
give
that
everyone
did
becomes
like
the
distribution,
and
I
think
that
pattern
of
like
pushing
the
decisions
out
to
the
edges
and
then
quantifying
them
and
then
lifting
them
back
up
by
you
know
running
algorithms
on
it.
D
I
think
that
that
is
a
generalizable
pattern
that
we
can
start
to
explore
more
and
that's
how
we'll
also
be
able
to
scale
beyond
certain
communities-
and
you
know,
flow
value
between
different
communities
without
needing
to
like
come
to
consensus
between
them
or
you
know,
have
to
fully
understand
who's.
D
Making
the
decision
in
this
other
community
and,
like
you
know
evaluating
whether
or
not
it's
accurate
or
not-
and
you
know
you
have
to
do
a
whole
bunch
of
due
diligence
before
you
decide
like
the
flow
funding
in
a
certain
community
or
direction
who's
going
to
be
managing
this.
Like
all
this
stuff
versus
you
know
letting
people
have
that
inner
subjectivity-
and
you
know
you
don't-
need
a
top-down
decision-making.
D
D
Sure
stock.
E
Thanks
so
the
biggest
challenge
I
see
with
reward
systems
is
that
there
are
always
people
in
them
and
that,
like
it's
kind
of
good,
but
it's
like
people,
it's
like
it
was
just
like
it
was
perfect
like
we
designed
it
so
well
right
and
then
you
put
people
in
and
they
are
not.
E
Each
person
themselves
is
like
a
super
complex
system
full
of
like
really
irrational
choices
constantly,
and
then
you
put
the
more
people
you
kind
of
put
in
you
know
the
more
complex
it
gets
and
the
harder
it
is
to
sort
of
design.
Something
with
that.
That's
going
to
be
totally
fair
and
totally
work,
because
people
will
take
all
sorts
of
unexpected
turns.
E
E
Looking
at
something
and
saying
and
deciding
what's
fair,
you
know,
and
you
know
what
kind
of
work
should
be
rewarded,
and
then
you
also
have
this
other
perspective
that
it's
like
well,
the
kind
of
work
that's
being
rewarded
is
what
what
the
system
wants
to
happen,
and
so
balancing
those
two.
I
would
say
it's
like
how
do
you
provide
minimum
viable
structure
for
people
to
start
interacting
and
sort
of
figuring
out,
what's
valuable
for
that
project
or
that
community
on
their
own?
E
And
then
you
know
how
to
what
is
the
feedback
loop
so
that
there's
some
kind
of
self-reflection
mechanism
say
like
wow,
we're
really
valuing
all
the
people
that
make
memes
like?
Is
that
really
what
we
value?
Is
that
what
we
want
to
be
rewarding
or
do
we
want
to
be
rewarding
other
stuff
too?
And
of
course
we
just
want
to
reward
the
means.
It's
usually
where
it
comes
out
by
secret
nodding.
Is
that
so
yeah?
E
I
think
the
the
challenge
with
road
systems
like
how,
how
small
a
structure
minimum
viable
structure
can
you
do
while
it's
still
helpful,
so
that
it's
not
just
total
chaos
and
then,
of
course,
like
the
more
pushing
to
the
edges
that
you
can
get.
You
know
push
the
decision
about,
what's
valuable
to
the
people
that
are
making
that
value,
and
that's,
of
course,
like
the
big
problem.
E
C
In
the
end,
you
know,
groups
of
people
have
a
tough
time
making
decisions
and
trying
to
have
like
a
list
of
people
and
everyone
deciding
on
who
gets
what
is
just
a
chaos?
So
really
you
it's
best
to
have
each
individual
person
like
make
a
ranking
or
or
give
their
own
quantification
as
an
as
an
individual.
But
then
you
run
into
this
challenge
of
oh
crap.
How
do
we
get
people
to
actually
participate?
You
know.
C
Do
we
pay
people
to
participate
like
all
the
the
people
who
are
loud
and
extroverted
are
going
to
show
up?
You
know
and
they're,
probably
going
to
reward
other
loud
and
extroverted
people,
and
so
that
now
the
introverts
that
are
quietly
doing
their
great
work?
You
know
that
don't
really
want
to
go
into
these
awkward
situations
where
they
have
to
quantify
people
and
stuff
they
get
not
they
get.
They
get
shafted,
you
know,
and
so
like.
C
C
C
And
I
also
think
that
there's
just
it's
really
hard
to
want
to
you
know,
there's
a
lot
of
value
in
division
of
labor,
sometimes
right
and
so
you're,
trying
to
you're
kind
of
fighting
against
that
you're
saying
no.
Everyone
has
to
quantify
everybody,
you
know
and
to
be
participating
in
this
work
or-
and
if
you
don't
have
that,
then
you
have
a
very
skewed
perspective
of
just
the
people
who
are
participating
and
what
their
their
backgrounds
are.
C
So
you're
you're
really
fighting
against
a
lot
of
natural
things
and
and
in
the
end
like,
unless
you
have
a
good
way
to
make
it
fun,
which
I
think
coordinate,
has
has
found
a
way
to
make
it
fun
and
and
really
focus
on
that
like
giving
you
know
this
isn't
like
rating.
You
know
this
isn't
scoring.
This
isn't
pain,
it's
giving
you
know,
and
and
this
framings
are
so
critical
and
getting
the
framing
right
is
like
half
the
battle.
C
You
know
praise
right,
we're
praising
people,
you
know,
but
then,
but
then
I
framed
it
all
wrong.
Oh
that's
how
the
sausage
is
made,
get
in
there
and
quantify
people
guys,
I'm
sorry!
That
was
a
mistake.
You
know
we
gotta,
we
gotta
do
better
framing.
We
have
to
make
sure
it's
fun.
We
have
to
make
sure
the
quantifiers
are,
are
treasured
and
also
as
diverse
as
possible,
yeah
I'll,
throw
it
to
octopus.
F
So,
in
addition
to
what
everyone
has
said
like
the
social,
technical
interaction
is
huge
because
even
with
social
media,
we
see
that
people
are
not
always
the
same
when
they're
interacting
in
this
digitally
mediated
way.
So
we
already
had
like
the
human
aspect
of
resource
distribution,
which
is
a
humongous
problem.
No
one
agreed
on,
and
then
you
have
people
coming
to
this
completely
new
space,
and
often
we
bring
different
expectations
about
it,
like
what
appeals
to
me
about
cryptocurrency
is
not
economic.
F
It's
about
being
able
to
get
things
done
without
coercion,
whereas
someone
else
may
just
be
trying
to
make
a
million
dollars
and
if
they're
not.
If
there
aren't
really
authentic
conversations
about
what
we
value,
then
you
can
have
this
situation
where
people
realize
they
value
completely
different
things
and
they're
just
now
discovering
that
when
they
go
to
give
the
rewards
because
they
want
to
reward
different
things,
I
think
that's
a
really
big
challenge.
Is
the
community
figuring
out
what
it
actually
values.
A
So
that
leads
to
a
question
that
is
something
I've
been
thinking
a
lot
about
on
how
to
move
away
from
the
manipulative
nature
of
the
systems
because
reward
systems.
There
is
a
big
literature
on
it
that
talks
about
how
both
punishment
and
rewards
come
from
the
same
source
of
a
manipulative.
A
Of
a
manipulative
system
that
you're
inducing
people
to
have
a
certain
behavior
to
have
a
product
of
a
result
that
you
want
or
that
you
would
like
see
happening.
So
I
think
we
are
slowly
shaping
something
else.
That
is
not
exactly
a
reward
system,
but
that
it
has
other
qualities
from
everything
that
I've
been
hearing
from.
What
are
you
guys
bringing
so
yeah?
F
I
think
just
for
me
personally,
the
the
economic
value
has
never
been
worth
whatever
community.
I
was
able
to
find
so
like.
There
was
a
time
in
crypto
when
I
was
really
focused
on
making
money
and
it
became
a
complete
depleter
of
joy
for
me
and
this
year
I
have
somehow
managed
to
make
zero
dollars
in
cryptocurrency,
but
I've
had
so
much
fun
and
I
think
that
ideally,
like
whatever
reward
systems
we
give,
would
be
supportive
of
those
authentic
things,
but
then
it's
like
as
soon
as
we
make
a
reward
system.
C
There's
no
good
answers
for
this
kind
of
stuff.
I
mean
in
the
end,
like
you
think
I
I
don't
know
I
like
to
think
of
money,
ideally
as
happiness
points
in
general.
Right
and
reward
systems
should
be
able
to
somehow
capture
like
oh,
how
much
happiness,
how
much
joy,
how
much,
how
much
production?
How
much
like
success,
was
this
person
in
this
time
period.
You
know
how
much,
how
much
do
they
contribute
to
that
and
measuring
that
is
like
or
like.
Where
do
you
draw
the
scope?
C
What
you
know
what
even
it's
it's
it's
it's
really
hard
and
then,
if
you're,
when
you
draw
the
scope,
you're,
naturally
creating
a
game
and
you're
going
to
have
people
jumping
through
the
hoops
that
you
put
up
like
that's
that's
the
game.
So
it's
I
feel
like
it's
an
impossible
task
to
say:
oh
we're
going
to
create
a
reward
system
that
doesn't
manipulate
or
incentive.
You
know,
drive
people
in
one
direction
or
another
that
that's
just
that's
the
nature
of
the
game,
you're,
creating
a
game
and
the
game's,
not
the
game
of
life
right.
C
It's
it's
the
game
of
whatever
scope
that
you're
putting
in
so
I
think
it's
important
to
design
it
so
that
scope
is
is,
is
clear
to
everyone
you
know
and
and
the
incentives
are
aligned
with
that
scope
so
that
you're
you're
it's
not
manipulation.
If
it's
just
a
game,
am
I
being
manipulated
when
we
play
monopoly?
I
don't
think
so.
I'm
participating
in
a
game
that
I'm
opting
into
that.
I
know
the
rules
and
I
know
what
the
goals
are
and
I'm
I'm
choosing
to
play
it.
C
So
I
feel
like
it's
probably
more
to
avoid
manipulation.
You
probably
just
need
to
communicate
effectively
and
design
the
system
so
that
it
actually
achieves
the
goals
of
the
system
and,
and
that
would
not
be
manipulating
it's
like
if
we're
playing
source
cred,
you
know
and
the
source
cred
says:
hey,
do
github
issues
do
forum
posts
and
make
people
give
you
emoji
on
discord.
C
C
E
Yeah,
that's
really
well
said
it's
funny.
It
reminded
me
of
of
when
I
was
a
kid
like
there's
this,
the
board
game
called
life
and
you
play
the
game
of
life,
but
how
you
win?
Is
you
make
the
most
money
and
like
this
is
a
message
that
we
get
so
much
so
often
that
it's
kind
of
like
a
default
even
programmed
into
the
subconscious?
And
so
we
use
this
phrase.
E
A
lot
like
there's
many
forms
of
currency
and
try
to
you
know
reinforce
this
idea
that
people's
actual
lived
experience
you
know,
which
is
that
like
there
are,
there
are
many
ways
to
keep
track
of
of
what's
valuable,
and
that
was
a
you
know
a
really
important
way.
We
tried
to
frame
coordinate
it's
like
no.
This
is
all
about
thanking
your
peers,
it's
not
about
judging
how
much
money
they
should
receive.
There's
nowhere
in
the
platform
that
says
like
one
give
is
worth
this
much
or
you
got
this
much
it
doesn't.
E
It
doesn't
show
or
talk
about
money
anywhere,
and
you
know
our
our.
I
guess
the
other
thing
is
like,
of
course
you
know.
Even
subtly,
anytime,
you
introduce
incentives
and,
and
mechanics
people
are
going
to
to
drift
toward
gaming
them.
That
just
happens,
but
I
also
feel
like
we
can.
You
know
it.
My
experience
in
coordinates
so
far
is
that
actually
people,
and
especially
people
that
are
collaborating
in
dows,
obviously
generalizing
hugely
here,
but
it
attracts
the
kind
of
people
who
are
like.
Oh,
what
is
this
game?
E
How
are
we
trying
to
figure
this
out?
There's
a
very
like
a
very
communal
spirit
and
a
lot
of
these
dows,
and
you
know
in
in
in
crypto.
E
It's
like
the
the
la
the
base
layer
is,
is
so
about
being
trustless
and
about
like
guarding
against
exploitations
and
like
this
is
you
know
it
has
to
be
ready
for
anyone
to
exploit
it
and
but
inside
dao's
we
see
it's
like
actually
a
really
trustful
system,
and
that
that
you
know
people
always
ask
right
away
the
first
thing
it's
like
well,
how
do
you
keep
people
from
colluding
you
know
and
how
they're
going
to
to
give
to
each
other?
And
it's
like?
E
Well,
you
know
we
have
a
message
for
that,
but
but
they
don't
people,
don't
really
try
to
collude
because,
like
they're
focused
on
on
what
the
project
is,
so
I
I
guess
the
point
is
like
be
aware
that
that
people
will
drift
toward
gaming
anything,
but
also
that
our
experience
is
that
people
working
on
a
project
together
aren't
immediately
looking
to
like
how
can
I
get
the
most
give?
That's
not
what
their
concern
is.
Their
concern
is
like
great.
How
do
I
reward
and
thank
the
people
I'm
with?
E
So
don't
don't
over
engineer
like
trying
to
worry
about
gaming
manipulation
like
trust
people,
and
they
can
be
trusted.
C
I
want
to
jump
in
and
I
think
that
it's
very
it's
easy
to
the
way
that
you
did.
That
is
perfect,
because
it's
super
subjective
there
isn't
really
games.
It's
a
lot
easier
to
game
source
cred
than
it
is
coordinate
because
coordinate
is
just
hey.
Do
your
friends
want
to
give
you
do
the
people
who
are
around
want
to
give
you
give
for
your
value?
You
know
so
like
so
the
value
you
created-
and
that
is
a
tough
game
to
like
you,
know,
game.
D
Yeah,
so
I
think
in
terms
of
manipulation,
it's
kind
of
like
you
know,
I
think
it
all
comes
down
to
like
dopamine
hits
at
the
end
of
the
day
right,
that's
kind
of
how
nature
operates.
It's
by
you
know
manipulating
us
with
like
reward
signals,
and
you
know
for
for
different
things,
and
you
know
it
can
get
unhealthy
right
like
according
to
our
brain.
D
You
know
it's
really
good
to
eat
pizza
and
candy
and
sugar,
but
not
it's
not
really
good
for
you
in
the
long
term,
so
it
has
to
be
balanced
as
well,
but
I
think
in
general
the
manipulation
of
our
reward
systems
that
we're
creating
here
have
like
a
much
better
like
net
positive
for
humanity
than
you
know.
D
Overall,
so
I
think
you
know,
there's
there'll
be
a
little
bit
of
like
we
have
to
like
play
the
same
game,
because
you
have
to
compete
with
instagram
for
attention
of
you
know
the
younger
generation,
and
so
a
lot
of
that
I
think
it'll
it.
It's
fine
to
have
these
rewards
systems
that
manipulate
people.
D
If
people
are,
you
know,
actually
feeling
rewarded
and
want
to
be
a
part
of
it,
and
you
know,
I
think
the
it's
like
griff
was
saying
it's
like
easy
for
people
to
like
make
those
decisions
for
themselves
like
if
they
feel
manipulated
if
they
feel
like
someone's,
not
acting.
You
know
in
good
faith
and
all
this
stuff
they
can.
D
You
know
we
should
design
these
systems
where
people
have
an
easy
way
to
like
opt
out,
or
you
know,
like
you,
know,
transfer
their
reputation
elsewhere
and
go
work
on
something
else
or
you
know,
move
between
groups,
and
you
know
because,
like
in
a
traditional
corporation,
you
have
like
much
more
you're
holding
much
a
lot
more
over
people's
heads
versus
endows.
D
You
know
you
really
want
a
system
where
people
are
there
because
they
want
to
be
there
not
because
they're
manipulated
to
be
there
and
if
they
feel
like
they're,
manipulated
it's
pretty
easy
for
them
to
get
out
and
go
do
something
else,
but
I
think,
like
it's,
it's
different
in
like
bull
market
versus
bear
market,
but
I
also,
I
also
think,
like
different
people
have
different
needs
in
different
stages.
D
So
there's
some
people
where,
like
money,
is
actually
the
most
important
thing
because
they
need
to
like
cover
their
bases,
and
they
can't
really
do
their
best
work.
If
they're
having
to
worry
about,
you
know
putting
food
in
the
table
next
month,
so
you
know
for
those
people.
Money
is
actually
like
a
really
key
thing
that
you
know
it
for
them.
It
is
important
for
it
to
be
focused
around
money
being
the
reward
system
and
then
other
people.
D
They
have
like
enough
money
that
they
don't
have
to
worry
about
that
and
for
them
the
reward
they
need
is
more
so
like
a
feeling
of
like
higher
purpose,
or
you
know,
they're
having
an
impact
or
you
know,
seeing
other
people
happy
or
you
know,
giving
their
money
away.
You
know
I
I
like
to
say:
money
can
buy
happiness.
If
you
give
it
away
right,
it's
it.
D
D
We
should
design
these
systems
to
accommodate,
for
what
different
people's
needs
are,
and
you
know,
and
some
of
the
dials
we're
working
in
we've
set
up
systems
where,
like
there's
like
a
ubi
like
baseline
for
everyone,
and
then
you
know,
we
use
that
as
like
the
this,
like
the
social
like
floor,
so
everyone
has
like
firm
footing
to
stand
on
and
then
on
top
of
that
we
have
all
these.
D
Like
more
experimental
reward
systems
that
you
know
it's,
okay,
if
things
go
wrong
once
in
a
while
or
we
don't
figure
it
out
right,
because
we
have
that
like
baseline
for
everyone
to
like
feel
comfortable,
and
then
you
know,
on
top
of
that,
we
we
can.
We
can
experiment.
D
D
I
have
internally
that
we're
using,
but
you
know
these
are
those
are
all
like
uncertain
and
it
does
a
lot
for
people's
like
sort
of
mental
comfort
to
say
that,
like
you
know,
you're
guaranteed
to
get
at
least
this
much
every
month
and
you
know
there's
not
anxiety
of
like.
Oh,
like
you
know
what,
if
I
don't,
you
know,
get
raided
well
and
coordinate,
or
you
know
it
and
it
starts
becoming
a
distraction
rather
than
the
focus.
E
Yeah
there's
a
few
protocols.
Definitely
that
are
also
doing
that
same
thing,
essentially
with
coordinate,
is
like
giving.
Everyone
gets
a
baseline
amount
of
give
so
that
you
know
you
have
a
floor.
It's
not
just
like
total
total
high
wire.
I
hope
I
get
as
much
as
I
need
to
to
make
my
bills,
so
it
becomes
more
like
a
bonus
structure.
A
big
bonus
pool.
A
First,
I
just
wanted
to
share
what
nikoline
wrote
that
I
think
is
also
opens
to
another
point
that
we're
also
starting
to
talk
about
of
alternative
reward
systems,
so
she
says
andrew
inspired
me
to
remind
me
what
I
did
with
my
kids:
never
reward
them,
but
search
for
what
motivated
them
to
achieve
something
and
what
success
looks
like
for
them
to
make
them
recognize
that
they
achieved
what
they
wanted,
what
they
wanted
to
reach,
so
they
set
the
framework.
As
here
you
choose
the
framework
and
play
your
own
game.
A
That
is
what
andrew
was
talking
about
too
of
when
the
people
have
choices
about
the
rules
of
this
game,
maybe
becomes
less
manipulative,
as
the
the
engines
are
transparent
for
everyone,
and
I
wanted
to
open
natalia.
Do
you
want
to
share
some
of
the
things
you're
rising
around
the
cultural
frameworks.
B
B
Then
they
get
it
and
they're
not
happy
with
it,
because
that's
that
was
actually
not
what
they
wanted
from
the
beginning.
So
yeah.
I
just
wanted
to
raise
the
importance
of
culture
within
the
systems
and
culture
being
supportive
of
humans
actually
discovering
what
they
want.
B
B
Is
that
idea:
yeah,
we're
figuring
it
out
as
we
build
and
also
everything
that's
happening
outside
of
the
systems
we're
building,
so
the
atmosphere
that
people
are
living
in
and
a
lot
of
what
metadreamer
was
talking
about.
You
know
the
the
contexts
are
going
to
be
different
and
it's
really
important
to
take
that
into
consideration
and
have
that
yeah
that
flexibility
and
space
business
to
hear
that
even
notice.
That.
D
Yeah
but
quick
thought
someone
posted
this
in
the
chat
earlier
too,
but
it
got
me
thinking
about
you
know
this
concept
of
you
know
self
regulation
or
kind
of
stating
what
you
want
to
accomplish
yourself
and
then
getting
rewarded.
If
you
meet
your
own
goals
that
you
set,
I
think
that's
a
really.
I
think
that
can
be
really
effective
because
you
know
dao's
had
this
general
issue
of
you
know.
People
like
come
in
and
be
like.
Oh
what
should
I
do
and
it's
like?
D
Oh
well,
you
know
no
one's
gonna
really
tell
you
what
to
do.
You
have
to
like
figure
that
out
so
there's
already.
This
need
for
people
to
like
be
self-starters
and
to
like
you
know,
create
their
own
like
things
they
want
to
work
on,
and
you
know
figure
out
how
they
can
provide
value,
but
you
know
tying
that
into
the
reward
system
as
well.
It's
like
it's
also
a
way
to
help
people
be
more
take
more
responsibility
for
their
work,
because
that's
another
issue
with
laos
is
like
you
don't
have
like.
D
You
know
the
threat
of
getting
fired
over
your
head,
so
you're
not
as
like
motivated
to
like
you
know,
make
sure
things
get
done
on
time
like
regularly
and
consistently,
and
you
know
there's
we
need.
We
need
more
mechanisms
to
ensure
that
happens.
So
you
know
even
like
staking
on
yourself,
would
be
a
pretty
cool
thing,
be
like
you
know,
I'm
putting
this
much
down
on
my
own.
You
know,
I'm
gonna,
do
this
work
and
I'm
taking
this
much
on
it
and
if
I
get
it
done,
then
you
know
I'll
get
a
reward.
D
If
I
don't
get
it
done,
then
I
burn
my
steak
and
it
goes
to
other
people
who
did
get
their
stuff
done.
You
know,
I
feel,
like
there's
some
cool
design
space
there
to
explore
in
terms
of
self
nominating
work
to
be
done,
and
you
know
using
that
as
a
motivation
for
people
to
like
meet
their
own
goals-
and
you
know
that's
sort
of
like
the
most
atomic
unit
of
the
dao
is.
Is
the
self
and
you
know,
there's
a
lot
of
work.
E
And
I
was
just
gonna
gonna
respond
that
you
know
to
griff's
point
earlier.
I
you
know
framing
is
like
framing.
Is
everything
it's
it's
so
critical,
because
it's
sort
of
it
it
tells
people
who
they
are
and
what's
happening
in
the
system,
and
I
think
you
know
culture
is
like
the
ultimate
ultimate
framing
device
for
people
that
working
together.
It's
like
how
are
we
around
here
and
I
just
really
appreciated
a
lot
of
things
that
you
were
bringing
up
around.
How
do
we?
You
know?
E
How
do
we
slow
down
like
how
do
we
create
a
culture
where,
where
listening
and
being
curious
is
more
valued
than
like
having
the
right
answer
and
jumping
in
with
a
solution
and
and
to
me,
I
guess
like
that,
zooms
out
to
like
really
what
is
the
the
whole
point
of
dows?
E
It's
like
we're
we're
you
know,
shifting
this
whole
paradigm
of
like
a
scarcity
and-
and
you
know,
winner-
takes
all
sort
of
economic
systems
and
social
systems
in
some
way,
just
saying
like
no,
we
can
actually
have
many
forms
of
currency
and
many
forms
of
value
and
there's
lots
of
ways
to
contribute.
Lots
of
ways
to
win
so
that's
you
know,
I
think
I
wanted
to
put
in
a
plug
somewhere
for,
like
bowser,
are
a
social
movement
as
much
as
they
are
a
new
technology,
a
new
way
to
organize-
and
I
think
it's
it's
important.
F
If
I
could
say
one
thing
about
that,
it's
also
true
that
there
have
been
dowels
before
there
was
cryptocurrency
like
alcoholics.
Anonymous
is
a
very
effective
now
and
and
looking
at
some
of
those
organizations
that
have
really
long
histories,
I
think,
could
be
very
informative
to
us
in
this
space
as
well.
A
Yeah
for
sure
I
never
thought
about
that
of
aaa
as
a
as
adele
good
point,
and
I
think
tying
to
what
you
were
saying
too
zach
of
how
we
can
slow
down
and
find
things
that
are
valuable
that
are
happening
in
in
other
dimensions
of
of
the
dow.
It
ties
to
just
question
of
how
to
capture
longer
slower
cycle
work
that
doesn't
the
labor
outputs,
for
example,
research
and
care
work,
invisible
work
that
builds
the
space
between
between
a
community.
E
I
mean
I
I
I
can
say
that
we
are,
you
know,
we're
we're
already
kind
of
messing
with
a
version,
two
of
coordinate
that
that
tries
to
take
on
some
of
that,
but
instead
of
giving
kind
of
one-time
disbursements
or
what
like
fast
money,
you
know
like
grants,
ways
that
you
can
actually
give
people
streams
of
of
income
so
that
they,
you
know
it's
more
consistent,
more
reliable,
because
a
lot
of
projects
obviously
are
not
don't
happen
in
the
same
sort
of
epic
cycle,
you're
working
on
them
for
a
long
time
and
I'll
also
say,
I've
been
surprised.
E
The
way
that
coordinate
really
does
capture
that
that
hidden
work,
you
know
people
will
give
usually
the
dispersions
are,
are
really
quite
spread
out.
You
know
with
people
giving
a
little
bit
to
a
lot
of
different
folks,
and
I
think
that's
the
recognition
of
you
know.
Thanks
for
helping
me
on
that
call
or
thanks
for
you
know
this
comment
you
made
like
we,
we
really
are
seeing.
If
you
look
at
the
date
of
the
coordinate,
comes
out
that
people
remember
all
these
little
things
you
know.
E
So
when
people
get
like
props
in
the
in
the
chat
and
stuff
like
that,
they
notice
that,
and
they
appreciate
it
when
they
have
the
chance
to
say
thanks
to
other
folks
who
are
who
are
doing
that
kind
of
weaving
work,
they
will.
C
Yeah,
I
think
it's
like
just
like
polycentric
governance.
We
need
polycentric
reward
systems.
One
reward
system
is
probably
not
going
to
reward
all
types
of
labor,
perfect,
very
well,
probably
at
all
that
each
one
is
going
to
have
its
little
niche.
You
know.
Sometimes
you
just
got
to
pay
people
a
salary.
I
mean
come
on.
That's
what
most
people
do
in
this
world
right
or
hourly,
a
salary
or
hourly.
These
aren't
horrible
concepts.
You
know
it
works
for
most
people.
C
They
don't
have
to
recreate
the
wheel
on
everything
you
know,
and
I
really
liked
just
like,
like
what
metadama
said.
It's
like.
Okay,
we
have
basic
ubi,
which
is
effectively
a
salary.
It's
not
really
a
uei,
it's
not
ubi
for
the
world,
it's
it's!
Just
like!
Oh,
hey,
you're,
a
part-time
contributor.
Let
me
just
pay
everyone
the
same.
It's
a
nice,
simple
solution.
C
F
Yeah,
I
think
what
makes
people
feel
valued
is
difficult
to
algorithmize
and
quantify,
and
attaching
a
salary
gives
them
a
number
right,
but
there's
a
lot
more
to
feeling
valued
than
that
and
it's
difficult
like
it.
One
thing:
that's
difficult
about
reward
systems
is
psychologically
our
perceptions
aren't
based
on
our
what
we're
actually
receiving
they're
based
on
relative,
what
we're
receiving.
F
F
Potentially-
and
I
will
pass
to-
I
don't
think-
hamad
has
spoken
on
this
topic,
yet
if
I'm
correct
so,
if
you're
not
you'll
spoke,
can
I
pass
to
you.
A
Cool
yeah
and
I
see
eli,
you
have
a
lot
of
thoughts
in
the
chat
and
I
know
that
you're
also
building
your
own
reward
system.
So
do
you
want
to
share
a
little
bit
about
what
you're
thinking.
G
G
A
Thank
you
and-
and
please
share
here,
the
link
you
sent
me
of
your
reward
system
and
the
things
you're
exploring.
I
think
people
will
be
glad
to
take
a
look
too
so
so
we
can
move
away
for
another
question
that
I
think
involves
a
lot
of
what
has
been
spoken
of
how
to
capture
value
and
how
to
bring
it
into
the
systems
how
to
quantify
in
the
right
way.
So
what
are
good
metrics
for
evaluating
the
data
collected,
so
value
flows
become
visible
and
amplified,
because
I
think
a
big,
a
big.
D
H
So
what
we're
doing
in
the
labs
today
is
we're,
starting
on
this
rewards
research.
F
Small
set
of
metrics
will
really
come
right
back
when
we
were
looking
at
analysis.
I
initially
just
jumped
on
genie
coefficient
because
that's
like
the
number
one
number
you
use
when
you're
analyzing,
if
a
system
is
fair,
but
someone
correctly
pointed
out
to
me
that
it's
not
fair
to
compare
like
large-scale
economies
to
developing
small
projects,
and
so
even
if
you
use
something
like
genie
coefficient,
you
need
to
be
very
aware
of
the
context
I
like
nakamoto
coefficient,
which
is
really
easy
to
calculate
it's.
F
How
many
people
would
have
to
conspire
to
hold
the
majority
of
the
resources,
so
it's
like
based
in
a
51
attack,
but
I
think
just
like
we
are
continually
developing
more
advanced
metrics
in
every
area
and
I
think
there's
so
much
room
to
develop
good
metrics.
Here
we
use
what
we
have.
I
think
we'll
have
something
much
better.
E
Yeah,
I
I
think
you
know
what
what
metrics
to
use
I
mean.
Ultimately,
it's
going
to
be
this
big.
This
basket,
it's
one
of
the
reasons
I'm
super
excited
about
the
medicred
project.
That
metadama's
working
on
is
that
you
know
it's
a
place
to
collect
all
of
these
sort
of
different
different
data
points.
Different
kinds
of
rewards
and
sort
of
you
know
have
have
a
fuller
picture.
E
Having
some
sort
of
being
able
to
process
and
have
like
a
a
conversation,
you
know
my
old
company
we
had
no
one
was
ever
on
salary.
We
only
just
after
you
know.
We
would
invoice
a
client
and
put
invoice
in
the
middle
of
the
table
and
whoever
worked
on
the
project,
we
would
sort
of
decide
how
how
it
should
be
divided
up.
It
was
kind
of
like
an
analog
coordinate,
but
the
real
value
in
it
was
the
conversations
you
know
that
we
would
have.
E
It
was
like
you
had
to
actually
have
a
conversation
when
someone
said
well,
I
spent
like
20
hours
on
xyz.
I
was
like
yeah,
but
like
we
didn't
even
use
it,
you
know
the
hours
aren't
the
thing,
and
then
we
have
to
have
a
conversation
about
like
well.
Our
hour
is
the
thing:
how
much
does
time
factor
into
how
much
you
should
be
rewarded
for
something?
So
you
know
how
we
build
that
into
coordinate.
That's
something!
That's
on
our
mind.
E
E
What
is
this
showing
us,
and
so
I
guess
this
maybe
is
just
like
a
huge
dodge
of
that
question
but
like,
as
we
think
about
metrics,
also
think
about
like
this
qualitative
piece
where
there
there's
a
vibe
right
and
there
just
isn't
a
metric
for
a
vibe
and
there
isn't
a
metric
for
people
like
feel
things
are
good.
E
You
know
what
gets
measured,
gets
managed
and
the
second
half
of
that
quote
is
and
the
most
important
things
are
the
hardest
to
measure
so
yeah
a
plug
for
a
plug
for
things
beyond
metrics
too,
even
though
they're
really
important,
and
that
it's
mostly
a
dodge,
because
I
don't
know
what
could
matrix.
C
H
H
C
Okay
or
something
you
know,
maybe
that's
a
that's
a
very
different
scope,
then.
Actually
we
have
this,
this
very
specific
product
that
we're
trying
to
launch
and
the
reward
system
is
funding
the
people
who
are
launching
that
project.
D
Yeah,
I
think,
like
like
I've,
said
it's
very
like
situational,
but
I
think
the
if
you
are
gonna
have
like
some
sort
of
global
metric.
It's
almost
the
metametrics
of
like
you
know
what
like,
how?
How
can
we
more
effectively
like
collect
data,
and
so
like
the
one
that
interests
me
like
the
few?
That
interest
me
is
like
the
the
time
to
compensation
or
like
the
time
to
fund
like
metric,
which
is
like,
as
if
I
like,
come
into
a
dow
and
start
contributing
how
long
before
I
get
some
sort
of
reward
for
it.
D
D
You're
usually
like
doing
the
tutorial
and,
like
you
know,
be
like
oh
chop
down
this
training
chop
it
down
and
you
get
a
level
and
like
it's
like
almost
immediate
feedback,
and
that
is
like
a
really
powerful
mechanism
and
I
think
the
faster
we
can
like
close
these
loops
between
like
doing
work
and
getting
compensated,
the
the
more
like
rich
data
and
like
better
contributions
and
more
accurate
signal
we'll
have
because
a
lot
of
the
time
the
issue
is
the
time
from
like
the
work
being
done,
and
the
reward
being
received
is
like
long
enough
that
a
lot
of
context
is
lost
and
a
lot
of
information
is
lost
from
that
time
and
then
another
metric
I'd
say
is
like
what
percentage
of
the
work
being
done
in
this
organization
is
being
like
is
like
covered
by
the
reward
mechanisms
or
the
metrics
we're
measuring.
D
So
it's
it's
almost
like
when
you're
writing
software.
You
have
like
test
coverage
like
what
percent
of
this
code
base
is
tested.
In
that
same
way,
you
have
like
coverage
on
what,
like
your
reward
system,
what
percent
of
the
contributions
or
valuable
work
being
done?
Does
this
reward
system
cover?
D
You
know,
like
group,
said
source
credit
covers
like
a
very
specific
niche
of
metrics,
that
you
know
it
does
it's
good
at
what
it
does,
but
it's
not
everything-
and
you
know
you
want
to
combine
like
first
credit
metrics
with
like
coordinate
subjective
data
and
all
these
other
things-
and
you
know
in
in
some
dials
we're
doing
things
where
we
use,
like
the
source
credit
score
to
decide
how
much
give
someone
gets
and
coordinate,
but
the
actual
allocation
is
done
based
on
give
you
know.
So
so
it's
like
they.
It's
like
a
feedback
loop.
D
They
feed
into
each
other,
and
I
think
the
it's
it's
better
for
us
to
collect
as
much
metrics
as
possible
and
then
curate
them
afterwards
like
create
markets
to
create
those
metrics,
because
different
situations,
different
metrics,
will
matter
in
different
situations
and
in
different
times,
but
the
the
key
thing.
The
key
important
thing
is
to
have
as
much
data
there
so
that
in
the
future
we
can
have
like
different
views
or
different
perspectives
on
that
same
underlying
data-
and
you
know
in
terms
of
subjective
opinions,
I
think,
with
cryptography.
D
We
have
a
really
powerful
paradigm
in
which
we
can
quantify
subjective
opinion
by
signing
messages
right,
if
I
just
make
a
statement
and
sign
it
with
my
public
key
and
my
public
key
is,
you
know,
is
attached
to
like
all
my
on-chain
actions
and
like
what
thousand
part
of
like
there's
so
much
like
rich
reputation,
data
attached
to
that
public
key
that
then
that
statement
and
signing
the
message
with
it.
That
allows
you
to
quantify
a
subjective
opinion
which
I
think
is
really
powerful.
D
And
you
know
that's
when
we'll
be
able
to
marry
and
merge
like
the
inner
like
the
subjective
parts
and
the
objective
parts
and
have
them
like
operating
within
the
same
system
and
for
us
to
be
able
to
like
compose
and
run
algorithms
on
on
both
of
them,
where
it's
not
necessarily.
We
lose
all
subjectivity
when,
as
soon
as
we
start
to
like
measure-
and
you
know,
do
tokens
and
you
know
try
to
quantify.
C
I
would
say:
defining
it
early
and
making
sure
it's
real
right
like
is
it
like
in
some
ways
you
want
a
self-fulfilling
prophecy,
if
you're
really
scoped
well
and
you're
like
hey,
we
want
to
build
a
successful
product.
Lots
of
people
use,
let's,
let's,
let's
build
metrics
that
bring
us
towards
those
self-fulfilling
problems.
Let's
build
a
reward
system
that
makes
this
pro.
You
know
actually
work.
C
That's
that's
the
point
right
so
that
I
think
the
challenge
is
what
happens
if
you're,
if
you're
not
well
scoped
and
that's
when
it's
a
lot
harder
and
then
you
end
up
into
these
feedback
loops
that
are
maybe
not
good
and
you
might
get
more
mob
mentality
than
like.
You
know
individual
truth
and
a
discernment
and
and
these
things,
but
if,
if
I
in
in
some
ways,
I
almost
I'm
not
really
answering
the
question
because
I
actually
like
well,
that's
that's
kind
of
a
point.
D
I
guess
one
one
related
thing
to
this
is
that
I
see
in
dowses
a
lot.
I
see
a
lot
of
people
kind
of
just
replicating
decisions
or
like
doing
things
a
certain
way,
just
because
that's
how
they
were
done
before,
and
you
know
a
lot
of
the
time,
especially
in
dallas,
because
things
are
so
chaotic.
Those
decisions
were
made.
Just
as
like
you
know
off
the
cuff
like
this
is
what
worked
at
the
time.
D
You
know
just
like
just
to
get
the
done
and
move
forward,
but
you
know
a
lot
of
those
decisions
end
up
like
becoming
embedded
in
the
culture
and
you
know
staying
along
staying
around
a
lot
longer,
just
because
you
know
someone
decided
to
do
that
that
one
time
in
that
one
way
so
having
I
think,
regular
or
like
like
creating
a
culture
of
people
to
you,
know,
self-reflect
and
even
within
an
organization
to
like
question
the
way
that
you
we
have
been
doing
things,
and
you
know
introspect
in
that
way,
I
think,
can
be
really
valuable
because
it
it
lets
you
sort
of
criticize
and
like
not
sort
of
assume
that
things
are
done.
D
A
certain
way,
because
that's
the
best
way
but
figure
out
how
to
iterate
and
and
do
them
better,
not
just
kind
of
repeat
the
past,
because
it's
whatever
the
status
quo
was
even
within
a
dao,
and
I
know
diaz
are
like
you
know
where
we're
all
about
like
defying
the
status
quo
of
like
legacy.
D
But
you
know
we
still
get
stuck
in
those
habits
and
it's
in
human
human
nature.
You
know
if
it's
not
really
effective
if
everyone's
just
questioning
everything
all
the
time.
So
it's
a
balance,
but
I
think
I
think
you
do
need
both
and
you
know,
maybe
in
in
source
credit,
for
example,
every
like
few
months
or
like
six
months.
D
I
think
we
we
used
to
have
like
a
retreat,
where
everyone,
just
kind
of
like
has
like
a
full
day
to
like
sit
and
not
do
work
but
kind
of
just
talk
about
what's
happening
like
share
their
feelings,
like
kind
of
see
how
things
are
going
see
what
like
wavelength
everyone's
on
you
know
if
things
are
still
in
sync
and
so
yeah,
I
think
better,
like
human
interactions
and
sort
of
cultivating
like
self-care
and
sort
of
you
know,
emotional
labor
and
all
that
sort
of
stuff
goes
a
long
way.
A
Yeah,
it's
very
curious
how
quick
traditions
are
formed
within
with
thousands
in
those
large
organizations,
and
I
think
this
is
a
challenge
of
culture
that
is
always
trying
to
balance
tradition
with
innovation,
because
you
want
to
respect
everything
that
is
being
built.
But
you
also
want
to
look
for
ways
to
improve
it,
and
then
this
balance
is
so
delicate
to
reach,
like
just
resonating
with
what
you
said.
A
We
have
one
more
question
to
andrew:
can
you
from
jess,
can
you
refund
using
raw
data
data
analysis
and
about
how
we
pre-process
data
from
different
analysis.
F
You
realize
it's
data
that
could
be
analyzed
later
and
try
to
record
it
in
such
a
way
that
it's
easy
to
analyze,
and
I
think
that's
something
that
we're
realizing,
there's
not
just
a
process
of
the
reward
system,
there's
also
a
meta
process
of
thinking
about
the
reward
system,
and
you
want
to
think
about
like
how
will
we
know
if
this
is
working
or
not?
What
kind
of
analysis
will
we
do
on
a
on
a
second
level?
F
What
you
want
to
analyze,
you
need
to
be
sure,
is
clean
coming
out.
It
makes
it
a
lot
easier
and
the
third
thing-
and
I
want
to
give
this
as
a
as
a
like
if,
if
future
analysts
only
hear
one
thing,
I
say
it
has
to
do
with
a
technical
thing
about
when
you
have
lots
of
categories,
try
to
break
them
up
into
sub
categories.
F
A
So
we
are
approaching
the
end
of
our
panel
and
I
wanted
to
see
if
anyone
from
from
the
audience
wants
to
share
any
thoughts
or
ask
any
questions.
G
C
Think
it's
important,
I
mean
it
depends
on
how
well
scoped
it
is,
but
you
can't
it's
hard
to
simulate
a
lot
of
human
discernment
stuff,
but
I
think
if
you
can't,
if
it's
a
heavily
metricized
one,
I
still
don't
know
if
that's
a
word,
but
if
it
is
one
of
those,
then
I
think
it
would
be
very
valuable,
especially
if
you
have
it
as
part
of
a
token
economy
and
it's
being
generated
in
a
certain
way.
F
I
would
say
also,
I
think,
there's
a
lot
of
stuff
out
there-
that
we
could
look
at
applying
to
reward
systems
like
iterated
prisoners.
Dilemma
is
something
that's
really
well
understood
academically.
It
seems
to
me
like
there's
a
way
you
could
translate
that
to
like
I'm
pro
simulation,
but
a
lot
of
times.
We
also
simulate
things
and
if
we
had
more
background,
it
would
be
helpful.
D
I
think
in
I
think
the
the
most
effective
way
can
be
done
is
if
it,
if
there's
like
a
strong
feedback
loop
between
the
simulation
and
like
the
real
world,
application,
an
experiment
and
having
one
inform
each
other,
like,
I
think
you
know
creating
models
and
simulating
them
in
a
vacuum.
From
like
things
that
people
are
actually
doing
like
it's,
it's
you
can
learn.
D
You
can
learn
a
lot
of
stuff
that
way,
but
it's
not
as
useful
as
like,
or
you
know
it's
not
that
practical
in
terms
of
the
real
world
impact
it
has
and
like
people
actually
making
better
decisions,
because
people
who
are
busy
doing
stuff
usually
don't
have
the
time
to
simulate
and
they're
more
so
learning
by
experimentation.
D
So
I
think,
having
a
way
for
the
people
on
the
ground,
to
experiment
and
share
what
they're
doing
to
the
simulators.
Who
can
then
you
know
generalize
and
find
patterns?
And
you
know,
computers
are
better
at,
like
you
know,
seeing
things
from
the
big
picture
and
then
from
like
an
objective
standpoint
than
humans
are
because
humans,
we're
always
only
seeing
things
from
a
specific
perspective
and
lens,
and
we
don't
have
the
global
viewpoint.
D
So
you
know,
I
think
this
comes
back
to
like
collecting,
really
good
metrics
and
data
from
what's
happening
in
the
real
world,
and
then
we
can
use
that
to
inform
our
simulations
and
design
better
simulations
and
then
take
the
learnings
from
the
simulations
and
apply
them
in
real
life.
And
you
know
just
iterate
that
way.
Almost
like
a
ai.
G
Yeah
I
mean
we
are
going
to
distribute
this
to
people
who
are
actually
collecting
data
to
train
ai.
I
mean
there
is
like
also
some
connection
with
in
this
regard
and
basically
we
can
do
then
change
office
rewards
continuously.
You
know
and
see
what
you
know.
No
there's
different
categories
like
uploading,
annotating
and
verifying
stuff,
so
we
have
different
categories
and
people
would
then
be
rewarded
differently.
Then
we
can
see
after
some
time.
Okay,
we
collect
more
data.
We
do
the
simulation
again.
G
I
Okay,
cool
yeah.
I
really
enjoyed
the
conversation.
I
also
enjoyed
all
the
chat
messages.
I
must
say
I
think
we
should
make
them
available
for
everyone,
because
we
had
links.
We
had
awesome
ideas.
We
had
a
great
proposal
there
and
I
can't
wait
to
dig
deeper
so
we'll
share
the
materials.
The
recording,
as
well
as
the
chats
stay
tuned,
follow
us
on
twitter
governance,
underscore
dao
or
at
the
telegram
group,
and
of
course
the
discussion
on
reward
systems
is
not
finished.
I
We
just
started
it
and
it
will
be
a
big
topic
for
the
governors
in
the
over
the
next
month.
First,
I'd
like
to
highlight
again
that
will
kick
off
a
research
program
where
you
all
are
invited.
If
you
want
to
take
part,
please
pre-register
james
shared
the
link
already
a
couple
of
times
at
tokenengineering.org,
and
this
will
be
around
the
question
and
I
think
a
mentor
dreamer,
you
summarized
it
so
nicely.
Of
course
we
need
to.
I
We
have
such
a
lot
of
data
available
now
in
these
systems,
but
we
have
to
make
use
of
them
in
an
intelligent
way,
and
it's
not
only
about
metrics.
It's
also
about
looking
at
the
real
world
and
and
try
to
not
only
think
in
quantification,
but
also
in
these
new
qualities
of
these
new
communities,
and
I
think
this
is
really
a
big
ambitious
mission.
We
want
to
make
first
steps
here
and
yeah.
This
is
a
culture.
I
Everyone
take
part
in
this
co-research,
let's
collaborate
here
and
I
already
can
see,
there's
such
a
lot
to
explore,
and
so
many
voices
and
great
thoughts
here
or
only
in
this
call
before
we
close
I'd
like
to
thank
you
to
ground
control.
This
is
the
team
of
the
governors.
It's
renzo
jess
libby,
james,
eugene
and
jeff.
So
these
people
are
working
in
the
background
to
organize
these
events,
to
organize
research,
surroundings
and
big.
Thank
you
to
you
all
for
investing
your
time
and
your
energy
and
passion
here
to
this
topic.
I
J
So
yeah
just
quickly,
so
we
can
maybe
even
let
everybody
go
a
few
minutes
early.
So
with
this
discussion,
ubi,
which
I
recently
heard
from
julio
with
circles,
ubi
also
means
unconditional
basic
income,
which
I
thought
was
a
really
beautiful
way
to
describe
it.
So
I'm
chatting
with
julio
there's
a
basic
income
earth
network,
which
is
a
huge
network
of
ubi
projects
around
the
world.
J
So
I
am
recruiting
some
panelists
to
discuss
this
topic
and
go
deeper,
so
proof
of
humanity,
circles,
ubi,
good
dollar
and
a
few
others
we'll
be
joining
as
well
as
bia,
hopefully
from
dada
art,
who
very
much
strongly
feels
that
every
incentive
is
coercive,
so
that
should
be
a
pretty
fascinating
panel
to
hear
just
some
different
perspectives.
J
Of
course,
this
is
obviously
a
complicated
topic.
So
super
appreciate
hearing
all
of
the
various
reflections
and
facets
of
the
diamond
around
this
conversation,
but
yeah
we'll
be
announcing.
So
if
you
want
to
click,
follow
on
governaut's
twitter,
so
you
can
stay
up
to
date.
That
should
be
a
really
fascinating
discussion
to
add
and
kind
of
round
out
a
look,
a
little
different
look
inside
of
that
part.
J
As
griff
mentioned,
of
kind
of
like
combining
these
systems
so
we'll
go
a
little
deeper
into
the
learnings
from
some
of
these
groups
who
have
been
deep
in
the
ubi
wormhole.
If
you
will
thanks
ange.
H
Thanks
everyone
awesome,
thank
you
so
much
to
the
panelists
and-
and
everyone
here
really
enjoyed
this
presentation,
and
this
is
being
picked
up
by
the
tec
labs.
We're
looking
at
quantitative
rewards
models,
so
we're
going
to
be
taking
a
look
at
all
a
lot
of
the
systems
brought
up
today
and
we're
doing
about
a
six-week
series
on
that
in
the
lab.
So
that's
every
friday
for
an
hour,
9am
pst,
6pm
cest.
H
So
it's
actually
right
now,
so
everyone
from
the
labs
was
redirected
into
this
call
and
we're
going
to
hop
back
over
to
discord
to
have
the
first
open
discussion
on
how
to
approach
quantitative
modeling
of
these
systems,
and
I
posted
the
link
in
the
chat
there
and
thanks
everyone
and
back
to
you.
Libby.
I
Perfect,
so
there
you
go
tea
comments,
just
people
mentioning
something
and
the
labs
get
kicked
off
in
a
minute.
I
Okay,
thanks
for
sharing
sean
thanks,
everyone
for
joining
hope
to
see
you
again
in
our
next
event
and
talk
soon
take
care.
Thank
you.
This
was
excellent.
Thank
you.
All.
H
Cool,
so
did
everyone
get
to
see
the
panel.
H
Awesome,
I
think
this
is
the
most
exciting
topic
like
this
is
the
closest
root
way
to
approach
basically
incentive
engineering,
token
engineering,
it's
it's.
How
do
we
reward
people
and
there
were
so
many
good
discussions
brought
up?
I
mean
just
to
name
a
couple.
I
really
liked
dr
penlin's
discussion
on
values,
right
and
and
actually
griff
talked
about
this
a
lot
as
well
with
the
self-fulfilling
prophecies
of
like.
H
Well,
that's
actually
exactly
what
we
want
if
we
get
the
scope
right
from
the
beginning
and
we
we
have
an
accurate
depiction
of
what
we're
trying
to
create
together-
and
I
think
natalya
touched
on
that
as
well,
with
sort
of
like
culture
and
purpose,
bringing
people
together
and
and
that
kind
of
bleeds
into
a
whole
topic
that
was
discussed
of
this
sort
of
intrinsic
motivations
versus
extrinsic
motivations.
H
There
were
some
discussions
on
sort
of
ubi
models
which
griff
said:
well,
it's
not
ubi,
it's
just
salary,
that's
just
what
that
is.
So,
let's
call
it
that
and
then
there
was
jessica
said:
ubi
should
actually
stand
for
unconditional
basic
income,
which
is
kind
of
interesting,
and
I
posted
some
of
my
opinions
on
the
chat
in
the
chat.
I
do
believe
that
ubi
is
sort
of
the
most
basic
essential
model
that
enables
people
to
join
a
network,
start
contributing
and
not
have
a
sort
of
stress
of
like.
H
H
You
know,
pay
their
rent
and
live
life,
but
also
be
able
to
participate
and
contribute
into
the
networks
that
they're
very
passionate
about
so
incredible
panel.
This
discussion,
it's
really
good
that
it
was
recorded.
I
did
want
to
actually
copy
and
paste
the
whole
chat
log,
but
I
forgot
to
do
that
because
I
feel
like
there's
so
many
topics
in
there
that
could
be
mined
from
a
very
technical
perspective,
a
sort
of
data,
science
perspective
and
modeling.
H
One
topic
I
was
going
to
bring
up
at
the
end.
There
was
some
discussion
of
simulations
and
I
think
it's
really
important
to
differentiate
models
from
simulations.
Often
we
talk
about
them
as
if
they're
the
same
thing
but
they're
very
different,
so
everything
all
these
systems
that
are
in
place
so
far
and
all
the
conceptual
ideas
ubi
or
the
give
token
that
coordinate
is
implementing
these
are
all
models.
A
model
is
just
a
way
of
construction,
constructing
something
it's
a
construction,
so
any
excel
spreadsheet
is
a
model
any
python.
H
H
And
then
you
can
not
just
model
your
system,
but
you
can
also
model
the
agents
and
how
the
agents
are
going
to
interact
with
the
models
of
the
system
and
then,
when
you
run
that
through
time,
that's
the
simulation.
So
I
did
want
to
bring
that
point
up.
I
didn't
quite
get
a
chance.
Modeling
is
essential.
H
I
hope
everyone
else
is
as
passionate
about
this
topic
as
I
am
and
I
think,
we'll
kind
of
jump
into
it.
I'm
going
to
start
this
lab
the
same
way.
I
start
every
lab,
I'm
just
going
to
open
up
the
lab
agenda
and
everyone
can
hop
in
there
and
from
there
we'll
just
get
started
on
this
sort
of
kickoff
of
researching
quantitative
reward
systems.
There's
been
no
work
done
so
far.
H
Sort
of
in
this
series
there
was
I'm
sure,
a
lot
of
you
heard
on
the
panel
or
maybe
know
about
this,
but
there
was
a
pretty
deep
research
initiative
led
on
the
praise
data
set
and
the
whole
praise
system
prior
to
the
hatch.
Quite
a
few
people.
I
know
noggin
was
involved
with
that.
I
think
bk
and
johan
were
sort
of
involved
in
that
a
lot
of
it
was
led
by
dr
penland
and
muhammad
also
a
great
data
scientist
with
long
tail
financial.
H
H
So
we're
going
to
be
looking
at
not
just
praise
systems
but
really
doing
a
survey
of
all
of
the
systems,
at
least
that
were
discussed
in
that
panel
discussion
and
then
as
many
as
we
can
get
our
hands
on.
So
we're
going
to
open
up
this
lab
series,
with
kind
of
a
survey
of
which
is
sort
of
a
traditional
academic
literature
concept
where
you
just
do
a
broad
general
scan
of
everything.
H
That's
out
there
and
available
in
the
world,
and
you
synthesize
as
much
information
as
you
can
about
those
systems
and
that
very
broad
discovery
process
enables
you
to
find
what
points
are
particularly
interesting
or
maybe
it's
the
general
patterns
that
are
repeating,
and
then
we
can
start
to
apply
frame
frameworks
to
this.
So
we
can
even
start
to
develop
some
of
our
own
literature
on
reward
systems
in
daos.
But,
most
importantly,
we
can
survey
all
the
literature
that
exists,
and
there
is
so
many
worlds
of
literature
on
this.
H
There
is
a
lot
of
research
being
done
already
in
the
dao
space.
If
you
look
at
block
science
and
prime
dow
and
yearn,
and
even
the
tec,
there's
been
a
lot
of
writings
between
blog
posts
and
forum
posts
on
these
reward
systems,
but
also
there's
an
opportunity
to
step
outside
of
the
dao
space
and
borrow
from
this
sort
of
interdisciplinary
nature
of
token
engineering.
So,
for
example,
dr
penland
mentioned
that
alcoholics
anonymous
is
a
dao
and
he's
totally
right.
It's
really
interesting
to
think
of
it.
H
That
way,
it's
been
around
way
before
crypto,
but
it's
a
doubt.
So
maybe
that's
a
great
case
study
that
we
could
look
at
and
say:
okay,
what
is
their
reward
system?
Well,
I
don't
think
anyone
gets
paid
to
run
alcoholics
anonymous
or
to
attend
it's
a
completely
intrinsically
motivated
dao,
which
is
a
very
interesting
case
study,
because
this
was
a
topic
that
came
up
as
well,
sort
of
subjective
rewards,
or
maybe
it's
more
closer
to
intrinsic
versus
extrinsic
or
quantified
verse,
qualitative
yeah.
Maybe
that's
what
we're
looking
at
here.
H
Although
there
must
be
some
sort
of
quantified
reward
process,
there
must
be
a
flow
of
value
or
funds
in
that
network,
because
I
imagine
it
would
cost
resources
to
facilitate
to
rent
spaces,
but
maybe
not
maybe
it's
often
done
in
churches
and
things
like
that.
So
really
interesting
case
study
there
hi
dr
penland.
H
So
I
was
just
kind
of
giving
a
few
of
my
what
I
saw
from
the
panel
discussion.
Just
a
few
topics
of
the
many.
I
was
saying
that
what
a
great
data
source
to
mine
that
it's
really
great,
that
that
discussion
was
recorded
and
I
really
hope
the
comments
were
recorded.
I
was
meaning
to
like
copy
and
paste
all
them,
but
I
actually
forgot
to
so.
I
hope.
A
H
Excellent
excellent,
so
right
there
from
that
panel
discussion,
I'd
say
we
have
enough
data
to
sort
of
seed
like
research
topics,
there's
a
lot
of
research
topics
that
could
be
pulled
out
of
that
and
what
we're
doing
here
is
about
a
six
week
on
quantitative
rewards
research.
H
In
fact,
maybe
that's
not
even
accurate.
It
doesn't
have
to
be
quantitative
systems,
I'm
coming
from
a
data
science
lens,
but
perhaps
some
people
will
actually
be
interested
in
and
qualitative
reward
systems,
and
so
that
was
a
bit
of
a
preamble
and
I'm
just
gonna,
now
open
up
actually
I'll
pause
here
and
see
if
anyone
has
comments
and
then
we'll
go
into
opening
up
the
labs
agenda
and
we'll
just
get
doc
started
on
initializing.
This
rewards
research
series.
H
H
So
we
have
today-
and
anyone
can
open
this
up.
I
think
it's
completely
open
for
editing
so
feel
free
to
maybe
we'll
add
everyone's
name
in
here,
which
is
actually
a
good
practice.
That
looks
like
we
forgot
about
for
a
couple
weeks
so
september,
I'm
having
a
little
bit
of
lag
on
my
computer
september.
G
B
H
A
big
topic
in
token
technologies
in
crypto
is
the
ability
to
do
retroactive
actions.
Actually,
I
think
metadreamer
brought
this
up
the
chain
on
chain
data
is
such
a
rich
source
of
mining
contribution
data
and
it's
a
concept
that
exists
in
blockchain
and
people
will
become
more
and
more
aware
of
it
into
the
future.
It'll
become
more
valuable
into
the
future,
but
the
idea
is
that
everything
that
ever
happens
on
chain
will
be
forever
accessible
as
a
data
point.
H
I
think
the
founder
of
the
matic
polygon
protocol
created
a
sort
of
charity
dow
for
covid
relief
in
india
and
people
were
able
to
send
funds
just
sort
of
through
ethereum,
and
it
raised
a
few
million
dollars
for
covet
relief,
and
he
didn't
tell
anyone
this
beforehand,
but
after
all,
the
funds
had
been
donated
as
essentially
charity.
He
then
went
ahead
and
initialized
a
dow
and
gave
people
governance
rights
according
to
the
contributions
that
they
had
made.
So
then
they
they
then
had
influence
over
how
those
funds
could
get
spent.
H
So
the
idea
of
retroactive
reward
systems
is
is
very
much
a
concept
when
we're
talking
about
token
technology
and
then
all
these
data
points
that
we
leave
all
of
these
artifacts
I
like
to
call
them,
are
very
important
as
we
go
about.
For
example,
this
document
could
be
scraped
sometime
in
the
future
and
there
could
be
rewards
to
everyone
who
has.
Oh,
my
dog
just
spilled
my
coffee,
okay,
cool
thanks,
everyone
for
joining
the
doc
there.
H
So
what
I'm
gonna
do
is
I'm
just
going
to
actually
spin
up
a
new
dock
and
we'll
call
that
our
sort
of
rewards
research,
manifesto
sean.
F
I'm
sorry,
how
did
we
get
into
the
doc
to
type
that
we
were
here.
G
Tech
labs
rewards,
oops
tech,
labs,
rewards
research.
H
Okay,
awesome
see
people
coming
in
here.
Oh,
this
is
great.
This
is
great,
so
I
think
what
would
be
a
really
good
use
of
our
time
we
have
just
over
15
minutes,
so
hopefully
we
can
squeeze
in
a
full
round
robin
we
can
go
through
everyone
and
put
in
some
notes
on
maybe
ideas.
We
have
maybe
something
we
came
away
from
the
panel
discussion
or
something
we
would
love
to
see
researched
in
this
research
series.
H
I
think
we
have
about
six
weeks
scheduled
through
the
labs,
so
that's
about
six
hours
of
collaborative
online
time
of
all
of
us
being
together,
including
today,
so
five
more
hours
coming
up
in
the
future.
So
that's
not
that
much
time.
So
it's
important
to
make
the
labs
time
very
effective
and
very
potent
and
we
can
all
sort
of
synchronize
and
then
it's
key
to
have
a
little
bit
of
asynchronous
momentum
throughout
the
weeks,
and
so
what
that
usually
looks
like
because
everybody
has
a
different
time
and
energy
available
in
their
life.
H
Usually
a
few
people
get
really
engaged
and
really
interested
and
and
pick
up.
You
know
pick
up
some
research
or
pick
up
some
some
modeling
or
pick
up
some
documentation
and
and
really
keep
it
going
throughout
the
week.
And
then
we
all
come
together
in
the
labs
and
a
few
people
have
something
to
share
and
then
I'll
usually
bring
some
direction
on
on
how
we
can
put
it
together
and
and
move
forward
for
the
next
week.
So
we'll
do
a
round
robin
here.
So
I'll.
Just
give
this
a
title
here.
H
And
I'll
I'll
go
ahead
and
start
so
I
am
extremely
passionate
about
rewards.
In
fact,
I
think
if
I
open
up
a
new
browser,
if
you
can
see
my
my
home
page,
let's
see
if
I
can
go
to
my
home
page
here.
This
is
this
is
actually
just
my
home
page.
This
is
my
favorite
book.
It's
called
reinforcement,
learning
and
introduction
by
dr
richard
sutton
and
andrew
bartow.
H
This
is
a
textbook
on
on
reinforcement,
learning,
it's
kind
of
like
the
foundational
textbook
and
reinforcement
learning
is
a
field
of
artificial
intelligence
where
agents
learn
from
feedback
from
their
environment
and
it's
all
built
on
reward
systems.
The
foundational
setup,
the
mathematical
setup
is
called
the
multi-armed
bandit,
and
it's
this
idea.
It's
kind
of
a
misleading
name.
It's
not
really
a
bandit,
it's
about
slot
machines.
So
if
you
have
a
slot
machine
with,
say,
10
arms
that
you
can
pull
and
every
arm
has
a
different
rewards
distribution.
H
You
know
one
arm,
you
pull
and
you
just
lose
your
money
every
time,
then
you
pull
that
arm
a
few
times.
You
probably
won't
want
to
pull
it
anymore,
but
there
might
also
be
a
arm
that
gives
you
two
dollars
every
time
you
pull
it
and
you
might
want
to
just
keep
pulling
that
arm,
but
there
might
be
another
arm
that
you
know
four
out
of
five
times.
It
takes
your
money
and
one
out
of
five
times.
It
gives
you
a
million
dollars,
and
so
there
it
that
brings
up
this
idea
of
exploration
and
exploitation.
H
You
need
to
explore
your
environment
and
know
in
able
to
exploit
it
anyway,
so
rewards
are
sort
of
the
fundamental
unit
of
quantitative,
like
it's
rewards,
are
quantitatively
measured
and
utilized
in
these
systems
to
actually
train
agents
to
achieve
certain
objectives
in
their
environment.
I'll
go
ahead
and
link
that
here
so
nice.
H
So
let
me
just
go
ygg
and
a
few
of
my
points
are
learning.
Thank
you.
Whoever
was
typing
that
learning
from
ai
research,
reinforcement
enforcement
learning
and
utilizing
well,
well-defined,
like
objective
what
we
might
call
objective
functions
or
really
just
like
quantitative
formulations
of
reward.
H
So
that's
really
interesting
to
me.
There's
another
topic
and
I
think
I
could
go
on
for
the
whole.
I
don't
want
to
take
up
too
much
time.
I
really
do
want
to
do
around
rob
and
I'll.
Just
put
one
more
point
here,
I
have
recently
been
fortunate
to
be
working
with
a
coach,
a
sort
of
relationships
and
community
building
coach
and
we've
been
learning
and
studying
about
this
model
called
nvc
or
non-violent
communication.
H
It's
really
a
very
comprehensive
framework
for
communication,
and
in
that
there
something
came
up
and
it's
this
idea
of
the
six
fundamental
human
needs
fundamental
human
needs.
So
I'm
really
interested
in
these
sort
of
like
more
qualitative
frameworks
that
we
could
actually
map
to
data
sets.
So
I'm
just
going
to
list
those
out.
Those
are
certainty.
H
And
just
in
the
sake
of
time,
just
in
case,
I
can't
remember
all
these
off
the
fly
I'll
type
these
out
as
they
come
to
my
memory
and,
in
the
meantime,
I'll
go
ahead
and
pass
it
off
and
whoever's
in
this
dock
feel
free
to
just
say
a
few
things
and
put
in
a
few
notes
or
if
you're,
not
in
the
doc.
I
I
can
type
for
you
as
well,
but
I'll
go
ahead
and
pass
it
off
to
bk.
G
Yeah
so
I
joined
the
call
pretty
late
and
I
don't
have
right
now
anything
to
say.
I
need
some
time
to
think
so
I'll
pass
it
to
next
person.
G
Sorry
can
I
say
just
pretty
quickly
because
I
have
sleep
early
today,
I'm
on
the
road
yeah.
I
also
joined
a
bit
late,
the
call
sadly,
but
well.
I
would
like
to
to
research
a
bit
what
has
been
what
is
what
is
being
used
right
now
like
very
interested
in
coordinate
and
all
that
stuff?
So
I
could.
I
could
take
a
bit
of
that
this
week
and
kind
of
look
into
it
and
write
up
some
stuff,
so
some
summaries
for
for
the
rest
of
that's
okay,.
G
Then
I'm
leaving
now
sorry
see
you
guys.
H
Thank
you,
noggin
I'll,
see
you
later
I'll
go
ahead
and
pass
it
over
to
gene.
G
Cool
yeah,
dr
pendleton,
brought
up
a
excellent
point
in
the
chat
during
the
rewards
video
earlier
about
how
improved
improvements
in
gpa
can
have
obvious
implications
for
the
educator
in
terms
of
like
rewarding
themselves.
So
I
would
be
interested
to
also
see
some
like
indirect
implications
of
reward
systems
and
how
how
advocates
for
certain
reward
systems
could
potentially
derive
benefits
for
for
themselves
by
distributing
rewards
of
various
sorts.
I
think
that
would
just
be
an
interesting
something
to
look
at.
H
Awesome,
thank
you.
Gene
go
ahead
and
pass
it
over
to
someone.
G
Thanks
quite
near
this
idea
of
rewards,
although
I've
been
in
a
space
for
a
while,
now
I'm
more
interested
in
the
programming
aspect
of
everything,
so
I
want
to
know
for
building
something
is
there
what
projects
are
currently
on
right
now?
G
A
We
don't
have
one
yet,
but
it's
probably
a
good
thing
to
have.
There
was
a
lot
of
resources
shared
in
the
chat
today
that
we
can
start
work
organizing
one
from
there
awesome.
A
Thank
you,
yeah
I'll
have
to
leave
on
the
top
of
the
hour,
but
it's
great
to
to
have
this
document
being
started,
and
I
think
for
me
something.
I'm
really
curious
is
how
to
have
the
interoperability
of
all
of
the
schools
working
well
in
the
way
that
we
don't
drop
the
value
in
the
cracks,
but
that
we
actually
have
like
a
view
of
everything,
being
mapped
and
and
also
ways
to
like,
evaluate
the
health
of
the
of
the
systems
of
the
organization
as
a
whole.
A
F
And
clear
great,
so
I
was
hoping
like
when
we
did
praise
analysis.
F
F
What
is
and
comparing
different
tokens
and
seeing
if
that
correlates
to
their
survival,
or
I
think
there
is
a
lot
of
hard
quantitative
work
we
could
do,
and
my
second
objective,
which
is
kind
of
higher
level,
would
be
to
kind
of
work
to
produce
some
cogent
educational
materials
for
people
who
are
getting
started
so
that
if
we
do
another
six
week
round
of
this,
we
will
have.
You
know
20
to
40
pages
of
notes
of
something
people
could
read.
So
they
don't
have
to
pay
the
same
startup
costs,
so
those
would
be
two
objectives.
F
I
would
be
interested
in
doing
and
I
will
pass
to,
I
think
it
says
xm
van
3.
I
don't
think
you
have
gone
yet.
J
No
I'm
here
hi,
so
you
can
call
me
thomas
by
the
way,
so
in
either
case.
So
I
quite
like
the
topics
that
have
been
already
mentioned.
So
one
thing
is
what
you
said:
sean
wright,
ygg
yep,
is
the
six
fundamental
human
needs.
I
think
that
sounds
really
interesting.
J
There's
right
now
so
much
going
on
in
my
brain,
so
I'm
kind
of
struggling
to
focus
my
thoughts
also,
given
that
the
talk
was
before,
but
topics
that
I
found
interesting.
But
I
have
to
be
honest,
I
don't
know
how
to
put
them
into
a
data
science
context
just
yet
there
would
be
something
I
would
have
to
think
about
was
the
manipulation
and
coercive
point.
I
think
that
was
made
and
also
the
individual
setting
individual.
J
Goals
and
being
rewarded
for
them.
So
that's,
I
think
it's
a
point
that
was
raised
in
a
chat
by
nikoline
and
metadreamer,
picked
up
on
this
issue
later
on,
saying
something
about
staking
on
a
goal
like
you
propose
a
goal
you
stake
on
it
and
if
you
get
it
done,
you
get
rewarded.
Otherwise
you
get
slashed
and
all
the
rewards
are
distributed
to
other
people.
J
So
I
don't
know
if
this
was
of
any
help,
but
this
is
kind
of
my
my
everything
that
falls
out
of
my
brain
right
now.
H
Awesome,
thank
you
thomas.
I
think
austin
do
you
want
to
go
next.
G
Not
sure
so
I
broke
down
the
short-term
versus
long-term
reward
system
that
was
related
to.
I
think
what
media
dreamer
and
doctor
pendleton
was
saying
about
how
the
games
are
very
short
friend
reward
systems
and
how
that
might
help
where
it
incentivize
or
you
desensify
certain
human
interactions
with
the
different
systems.
G
That's
also
somewhat
related
to
the
reinforcement.
Learning,
since
usually
the
ai
agents
are
a
lot
better
at
looking
long-term,
but
we
are
something
that
I
found
kind
of
interesting.
H
Yeah,
that's
super
interesting
and
that
you're
right.
The
time
aspect
is
a
key
component
of
reinforcement,
learning,
there's,
essentially,
three
pillars
of
machine
learning,
there's
supervised,
learning,
unsupervised,
learning
and
reinforcement,
learning
and
reinforcement.
Learning
is
the
only
one
that
has
this
sort
of
time
domain
and
it's
it's
so
fascinating,
because
that's
what
we
experience
as
humans
is
like
always
that
trade-off
between
short-term
and
long-term
goals
and
it's
the
concept
of
investing
right,
investing
time,
investing
energy,
investing
money,
fascinating,
fascinating
stuff
and
there's
a
lot
of
directions
that
that
can
go.
H
So
that's
a
that's
a
great
one,
austin
thanks
for
that,
and
I
think
nate
is
still
up.
G
J
F
Keep
my
shirt
the
thing
that
really
interests
me
about
the
modeling.
These
reward
systems.
H
H
Yeah,
we
could
hear
you
nate.
I
think
I'm
lagging
a
bit
on
my
end,
just
because
I
got
so
much
going,
recording
and
and
everything
the
impact
of
reward
systems
on
social
and
organizational
structure.
H
Awesome,
it's
great
to
see
people
building
out
this
dock
in
parallel.
That's
what
I
like
to
see
many
hands
make
light
work.
We
are
at
the
top
of
the
hour.
I
want
to
really
thank
everyone
for
being
here
again,
thanks
to
dr
octopus
and
the
other
panelists
that
we
got
to
see
on
the
panel
earlier
and
looking
forward
to
a
great
series.
H
It's
awesome
to
have
you
all
here
in
the
kickoff,
so
lots
of
good
things
ahead,
I'll
see
you
back
next
week
and
feel
free
to
reach
out
to
myself
or
anyone
else
in
this
group.
If
you're,
if
you
have
interests
or
want
to
do
some
async
work
together
and
otherwise
have
an
awesome
day
and
a
great
weekend.
Everyone
thanks.