►
From YouTube: W30 Rewards WG: Quantifiers and praise data and more!
Description
🙏 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
A
C
A
Okay,
cool,
that's
that's
good
enough,
and
do
we
need
to
put
something
else
up
on
the
the
agenda
for
the
day
today
some
discussion
topics,
or
should
we
dive
right
into
the
status
updates.
D
Hey
chris,
I
talked
with
libya
showing
to
you
the
advantages
of
the
script
and
storyboard
from
the
next
video
of
reward
system.
I
don't
know
if
yeah
cool
the
last
one,
but
we
are
here
now
yeah.
Thank
you.
A
Cool
and
we
have
two
nice
things
to
talk
about,
look
at,
but
but
we'll
we'll
start
with
status
updates,
we'll
start
with
the
trace.
As
always,
and
let's
see
who
is
here-
it's
currently
only
matt.
So
please
start
us
off.
E
Maybe
that
was
last
week,
I'm
not
sure,
but
so
I
I
the
I
kind
of
consolidated
the
the
logic
to
calculate
the
scores
into
one
module
and
made
it
as
simple
as
clear
as
possible
and
then
wrote
tests
on
it
to
confirm
that
it's
behaving
as
expected
and
also
resolved
the
bug
that
was
the
impetus
for
that
which
was
dismissed,
praises
were
not
being
included
in
the
average
score
in
the
exported
csvs.
E
So
that
should
be
resolved.
And
then
I
I
made
the
improved
user
interface
for
marking
duplicate
or
for
quantifying
praise.
Now
you
can
do
multiple
at
once,
and
I
fixed
a
bunch
of
little
minor.
B
E
And
yeah
right,
I'm
looking
into
this
database
issue
right
now,
so
I
don't
have
a
real
update
about
it
other
than
that,
I'm
looking
into
it.
A
A
yeah,
so
it
works
like
this.
You,
you
select
the
items
that
that
are
duplicates
and
then
market
mark
as
duplicates
and
and
then
you
get
to
select
a
a
score
for
the
for
the
first
one
in
the
series
and
then
the
the
other.
In
this
case
the
other
two
will
be
marked
as
duplicates
of
that
first
issue,
and
I
think
I
also
noticed
a
a
bug
here.
A
B
E
Right
yeah:
this
is
not
implemented
in
production
yet,
but
we're
just
going
over
it.
Now
we
were
talking
about
previously
when
we
were
calculating
scores.
E
We
were
taking
the
floor,
so
the
integer
the
integer
floor,
so
it
pretty
much
only
made
a
difference
when
we're
calculating
duplicate
scores,
because
it's
a
it's
a
parameter
that
admins
can
set
for
the
period.
How
much
does
the
duplicate
receive
of
the
original
score?
E
And
so
you
know
it's
it's
a
it's
an
arbitrary
decimal
number,
so
we
lose
a
little
bit
of
the
whatever
you
call
it.
We
lose
decimal
places
if
we
just
multiply
a
a
number
with
a
few
decimal
places
by
the
original
score.
E
So
we
have
a
pr
open
to
start
rounding
to
maybe
a
few
decimal
places.
I
think
we're
just
going
to
pick
an
arbitrary
amount
or
we
would
or
make
that
another
parameter.
A
Yeah-
and
I
I
I'd
say
we-
I
think
this
is
it's
quite
important
to
get
this
online
before
we
do
the
first
talking
distribution
because
it
actually
affects
the
the
amount
that
people
will
get
paid,
and
so
so,
but
it's
done
mostly,
and
so
let's
try
to
merge
it
and
put
it
push
it
into
production
before
boo
and
and
then
I'll
do
a
new
export
of
the
of
the
data.
So.
A
And
thanks
matt,
what
have
I
been
up
to
in
praise
development,
yeah
I've?
I
worked
on
the
dockerization
of
the
apprais
system
to
package
everything
into
docker
containers
which
will
make
it
the
install
of
a
new
server,
a
pretty
server,
much
more
easy
and
fast,
and
that
is
mostly
done
so.
The
current
production
server
is
now
now
running
using
that
method
of
running
containers,
so
that
was
nice
to
have
that
online.
And
what
else
did
I
do?
A
A
F
Sorry
I
was
caught
up
in
another
meeting
christopher.
There
was
another
you
showed
us
a
screen
cap
earlier
of
maddie
g
had
something
for
the
duplicate
praise.
Maybe
you
guys
talked
about
it.
A
A
So
do
we
have
anything
to
say
about
source
cred?
I
don't
think
so
other
than
I
could
say
this,
but
mateo
is
not
here
right,
no
mateo.
He
updated
the
the
list
of
github
repositories
that
we
monitor
using
source
cred.
A
A
F
A
I
I
don't
have
any
any
more
changes
or
that
I
would
like
to
get
in
I
I
know
now
we
have
a
really
long
list
of
repositories.
We
we
cover.
Basically,
we
we,
the
the
we
look
at
all
github
repositories
that
have
had
activity
since
last
july.
A
A
F
Maybe
maybe
this
is
a
rad
thing,
but
what
we
really
wanted
to
see
was
that,
based
on
both
data
streams,
like
do,
we
feel
like
everyone's
being
represented
fairly
or
more
fair.
Now,
because
we
had
source
cred,
but
when
they
realized
that
the
people
who
were
doing
the
heavy
lifting
on
github
weren't
showing
up
high
enough
in
the
list
in
in
the
whole
system.
A
Yeah
and
it's
difficult
to
to
to
value,
as
I
said
all
of
a
sudden,
since
we
we
included
praise,
I
have
been
become
the
main
cred.
A
I
have
the
biggest
credit
cred
of
all,
and
that
would
be
my.
My
guess
is
because
I
have
made
really
a
lot
of
small
contributions
and
and
of
course,
a
lot
of
big
contributions
as
well.
But
I've
been
like
constantly
pushing
pull
requests,
comments
and
issues
over
a
fairly
long
period
of
time
and,
for
instance,
like
a
person
like
sam.
A
If
you
see
on
the
screen
here,
it
looks
like
it
was
really
active
around
last
summer
and
then
the
level
of
activity
has
been
lower.
I
I
guess,
as
like
I
described
before
as
a
as
an
activity
tracker,
I
I
would
assume
that
we
have
set
up
source
credit
in
in
a
good
way,
it's
difficult
for
source
credit
to
to
gauge
the
the
quality
of
the
contributions
made,
because
it's
it's
only
counting
the
number
of
contributions,
sort
of.
B
B
G
B
A
A
I
don't
know
how,
if
we
said
that
comments
which
must
be
less
worth
than
than
the
the
the
issue
in
itself
or
something
like
that
or
the
pull
request,
but
mostly
it's
a
tracking
tracking
activity.
B
B
A
And-
and
it's
worth
also
keeping
in
mind
that
that
this
will
be
the
only
only
a
quarter
of
the
total
rewards
budget
goes
towards
source
cred.
A
So,
let's
switch
over
to
the
rad
rad
dashboard,
yeah
nagan,
you,
you
showed
me
a
bunch
of
stuff
today
and-
and
I
hope
we
can
see
something
from
chiwe
as
well.
H
Yeah
yeah
well,
first
of
all,
sorry
for
being
late
too,
I
was
in
this
other
meeting
with
mitch
yeah
we
kind
of
moved
the
the
repo
to
a
new
one
to
kind
of
separate
the
actual
rad
system
from
the
place
where
we're
going
to
be
saving
all
the
all
the
distribution
data.
H
It's
it's
ready
more
or
less.
I
think
mitch.
You
can
try
it
out
in
this
quant.
If
you
want,
we
can
also
do
a
call
if,
if
you
run
into
any
troubles,
but
it
should
be
fairly
straightforward,
like
we
did
a
test
run
with
christopher
today
and
it
went
well
yeah,
but
the
main
point
is:
is
your
way
she's
going
to
show
us
the
new,
the
new
analysis
notebook
she
did
so
I'll
refer
to
her.
C
Yeah,
okay,
so
yeah.
This
is
the
new
repo
where
naga
was
talking
about
the
raspberry
pop.
Oh,
can
everyone
see
my
screen.
C
C
Yes,
yes,
thank
you
yeah,
so
we
have
around
one
two,
eight
starting
from
2021
july
to
the
end
of
2022
february,
and
some
I'm
just
starting
with
some
basic
numbers
like
how
many
praise
do
we
have
in
each
quantification
period,
start
small
and
growing
up
and
the
maximum
was
seeing
the
end
of
december
to
january
1700.
A
C
G
F
C
C
B
C
Okay,
quantifier,
I
think
the
number
of
quantifiers
actually
determined
by
the
system
right-
it's
not
it's
not
going
to
change.
Based
on
how
many
praise
we
have
in
each
period
is
that
correct.
F
C
F
Yeah,
what
would
be
really
cool
is
to
see
how
much
how
much
was
quantified
versus
how
much
was
assigned,
because.
C
F
E
In
theory,
we
could
add
data
of
well,
we
have
data
of
when
a
change
was
when
a
quantification
was
updated,
but
it's
not
currently
being
exported
into
the
csv
another.
A
simple
thing
we
could
do,
maybe
is
the
time
from
when
the
quantification
period
is
opened
to
the
time
that
the
the
quantifier
has
finished.
G
E
Right
yeah,
that
would
be
very
useful.
Are
you
taking
all
this
data
from
the
exported
csv
z-way
yeah.
B
A
A
Matt,
do
we
track
the
individual
change
timestamps
for
for
individual
quantifications,
or
is
it
for
the
phrase
item.
A
I
don't
think
we
we
track
the
individual
quantifications.
E
No,
we
do
actually
yeah
because
it's
we
created
a
yeah.
We
do
have
an
update
of
that
and
created
that
date
for
each
individual
quantification.
B
G
C
B
F
C
C
So
I
I
I
would
say
just
based
on
the
current
data,
I'm
gonna
using
the
current
data
to
see
how
well
we
can
detect
that
kind
of
behaviors.
And
then
we
see
if
we
need
to
add
anything
else.
B
Very
good,
sorry
to
anyway,
sorry
to
interrupt
you
living.
C
Okay,
so
now
I'm
trying
to
tap
into
some
system
health
evaluation
that
is
tied
to
the
discussion
levy
started.
C
So
one
first
idea
is
the
mtc
members
who
are
who
have
been
acted
in
one
quantification
period,
and
here
I'm
defining
someone's
active
by
either
their
giving
or
receiving
praise.
During
this
count,
then
we're
comparing
in
each
con
period
the
unique
ids
that
give
for
received
praise,
so
the
the
blue
ones.
A
C
A
C
Yeah,
so
the
the
blue
line
here
are
the
number
of
new
users
in
each
quantification
period.
Actually,
this
didn't
look
a
bit
strange
because
the
number
oh,
this
is
including
praise,
giver
and
first
receiver.
So
it
might
be
right
this
number,
but
I
intended
to
define
this
as,
for
example,
in
this
quantification
period,
how
many
users
are
new
that
didn't
appear
in
the
last
quantification
period,
but
are
involved
in
this
period?
A
So
it
compares
between
between
the
periods,
so
in
the
first
period
a
hundred
percent
would
be
new
and
then
in
the
second
period
it
would
you
you
compare
with
the
first
and
so.
C
The
green
line
is
the
two
names.
A
Yeah,
it's
it's
quite
interesting.
You
see
quite
a
a
steady
rise
of
on
on
the
number
of
users
leaving
and
a
steady
decline
of
the
decline
of
net
users,
totally
total.
B
C
C
I'm
gonna
double
check
this
data
because
the
I
I
find
it
hard
to
believe
that
every
period
there
are
so
many
new
users,
it
could
be
something
like
this
person
appears
in
september
and
november,
but
not
in
october.
Then
they're
counting
us
leaving
one
time
then
rejoining
again,
so
this
metric's
a
bit
unstable.
H
B
H
A
A
A
F
E
One
issue
may
be
that
currently,
with
the
exported
csv
we're
exporting
the
user
account
name
as
it
appeared
on
discord
at
the
time
of
the
export
in
our
database,
and
it
doesn't
rather
than
it's
not
able
to
handle
if
people
have
changed
their
discord
names
over
time.
So
a
lot
of
the
new
users
may
actually
be
repeat
users
who
have
changed
their
names,
so
I'll
need
to
include
the
the
the
database
id
in
the
export
as
well.
E
A
E
But
we
have
a
two
user
account
and
a
from
user
account.
So
we
would
want
two
ids.
B
A
Yeah
we
should
have
two
two,
two
user
id
and
from
user
id
as
well.
I
add
those
two
columns.
B
C
A
Yeah,
I
I
don't
have
a
a
a
clear
idea
about
the
the
definition
of
an
active
user
other
than
that
it
could
be
a
person
that
has
been
both
sending
and
receiving
praise
for
more
than
one
period,
and
I
don't
know
how
and
and
the
the
the
longer
with
the
time
span.
We
choose,
of
course,
the
more
the
the
less
active
users.
We
will
have.
A
A
So
are
we
like
a
long,
long
term
low
churn
community
or
are
we
like
a
high
velocity
past
high
churn
community.
A
G
C
A
Yeah
that
definitely
could,
but
the
managing
different
sources
like
we
just
today
is
some
way
in
somewhere
into
the
future.
C
Okay
last
thing
I
worked
on
is
this
naca
model
coefficient,
so
it's
defined
as
the
smallest
number
of
count
who
can
control
50
percent
of
the
resource
here
the
resource
is
the
price
score
and
it's,
although
its
significant
significance,
relates
to
the
prospect
of
a
51
percent
attack
on
network
which
may
not
be
exactly
relevant
in
the
context
of
praise
score.
It's
still
an
intuitive
measure
of
the
system
quality
and
the
bigger
coefficients
means
you
need
more
people
to
pass
50,
so
that
is
more
equal
distribution,
smaller
concentrated
power.
C
So
here
we
do
see
fluctuation
and
in
february
it
reached
to
a
peak
of
a
very
equal
meaning.
Many
people
are
sharing
pre-scores,
instead
of
being
concentrated,
so
seems
like
only
looking
at
this.
C
B
C
Yeah,
so
here
is
the
ratio?
Well,
it's
not
really
divided
by
the
total
number
of
community
member
or
active
members.
The
total
number
of
number
of
people
who
receive
praise
in
this
period,
but
still
like
here
in
this
november
to
december
quant,
it's
six
percent
that
can
have
the
50
percent
of
the
total
reward.
So
that's.
F
B
B
C
E
C
Yeah,
it's
the
just
the
total
reward
scores.
We
have
a
total
number
of
reward
scores
and
50
percent
of
them
be
the
minimum
number
that
can
account
for
50
of
the
total
reward
scores.
C
I
chose
this
just
because
it's
intuitive
to
incorporate,
but,
yes,
we
can
compare
the
different
equality
metrics
and
see
the
art.
Do
they
basically
show
the
same
thing
or
there's
something
different.
A
Great
work
to
see
way,
yeah
really
awesome.
Thank
you.
Thank
you
so
sorry
to
to
stress
us
a
bit,
but
we
let's
move
on
so
we
get
to
the
end
quantifications
mitch.
F
Apart
from
that,
because
there
was
so
little
praise,
there's
only
like
12
quantifiers
instead
of
20.,
mostly
everyone's
done,
but
yeah
business
as
usual.
We'll
have
a
review
call
tomorrow,
we'll
set
up
the
next
round
we'll
start
april.
Praise
that'll
include
the
new
praise,
bot
praise.
B
A
B
A
Yes,
since
april
just
ended
four
days
ago
and
we
have
been
using
the
the
old
bot
as
well,
so
it's
we
need
to
do
an
import
before
we.
F
Yes,
that's
it
we'll
wrap
up
march
tomorrow
well
end
of
day
and
then
we'll
start
april
review
calls
all
good.
A
G
A
A
So
now
we
just
need
to
finish
up
and,
as
like,
I
said
to
include
this:
this
fix
with
the
decimal
fix
so
and
and
make
it
a
new
export
again
so
that
we
know
that
the
prey
score
is
as
it
should
be,
and
before
before
running
the
red
dashboard
and
doing
calculating
the
token
distribution
and
doing
it.
F
Noggin
prepared
the
the
csv
output.
I
think
you
messaged
me
about
it,
but
I
forgot.
B
B
G
B
A
D
Yeah,
thank
you
chris.
We
don't
have
much
time
and
we
wanted
to
present
you
the
reduced
script
and
show
you
three
words.
We
have
been
working
some
weeks
ago
with
alan
john
jonas
and
chewie
for
the
upcoming
and
reward
system,
video
tutorial
for
new
quantifiers.
D
It
is
a
google
doc
document,
so
you
can
make
your
your
comments
and
feedback,
because
I
don't,
I
don't
think
we
have
enough
time
to
to
present
in
here
live.
But
if
you
want
to
present
in
that
the
next
week,
it
will
be
great.
If
you
want
to
make
the
comments
in
the
google
docs,
it
will
be
awesome,
so
we
can
make
the
changes
coming
to
you.
The
reward
system
expert
and
commend
you
a
little
bit
about
the
about.
The
video
is
about
some
robots.
D
The
robots
represent
the
quantifiers
and
it's
like
a
little
story
telling
the
rules
and
all
this
stuff.
You
need
to
be
a
quantifier
with
some
jokes
and
stuff
like
that,
also
showing
how
to
use
it,
how
to
use
the
dashboard
the
days,
the
the
dates
that
we
use
for
the
praise
and
all
that
stuff.
But
if
you
want
to
do
that,
the
storyboard
is
the
reward
system
channel
and
that's
it.
If
you
want
to
present
the
next
call
more
with
detail,
it
will
be
great,
but
then.
A
A
Next
week
we
will
not
make
it
the
last
last
item
on
the
agenda.
D
Okay,
what
do
you
think
olivia.
F
D
F
E
D
A
Thank
you
as
a
laser
okay,
then
we
ended
up
having
four
four
minutes
left.
Sorry
for
stressing
everyone
there
at
the
end.
What
how
should
we
spend
these
four
minutes.
F
Oh
yeah,
like
a
part
here
that
doing
the
outreach
for
unactivated
praise
accounts
like
I
think
I
think
eduardo
needs
help,
because
it's
a
lot
of
people.
It's
like
400
something-
and
so
I
don't
know
if
like
if
he's
on
vacation
like.
Can
anybody
help
him
out
if
you're
looking
for
a
job.
F
B
F
A
I
have
a
feeling
that
everyone
still
left
in
this
meeting
are
quite
busy.
They
lit
most
of
us.