►
From YouTube: CHAOSS Value Working Group 4/7/22
Description
Links to minutes from this meeting are on https://chaoss.community/participate.
A
Okay,
welcome
everyone.
Welcome
to
the
cures
valley,
working
group
meeting
april
7
2022.
Please
add
yourself
in
the
meeting
with.
B
A
Yep
so
the
first
item
on
the
agenda
is:
we
have
to
release
the
metric
like
contributor
net
recommendability,
which
was
renamed
like
project
recommendability,
so
I
have
created
a
pr
for
that
kevin
or
john
anyone
can.
A
A
A
Yeah,
so
once
this
is
done,
then
kevin
has
to
fix
that
on
the
website
too,
because
this
will,
this
is
a
name
change,
so
it'll
break
the
links.
B
A
Here
is
a
new
link
which
I
generated
based
on
our
previous
discussion,
so
we
had
the
discussion
on
project
popularity
and
project
engagement
and
project
engagement
was
changed
and
I
think
yeah
we
worked
on
that
and
refined
it
and
when
I
reviewed
both
were
separate.
So
I
worked
on
the
project
popularity
metric
model.
A
C
A
E
F
G
C
C
G
C
All
right,
so
so
the
historic
the
history
here
is
that
we
did
have
a
project
popularity
metric,
but
it
was
basically
a
metric
model
right
bernard.
Yes,
it
was
a
bunch
of
different
metrics.
Compounded
again,
it
was
very,
very
similar
in
its
actual
structure
to
a
metric
model
because
it
was
integrating
a
number
of
distinct
metrics
into
one
metric
yep,
and
so
this
is
what
we
now
call
a
metrics
model,
so
model
yep.
A
So
it
has
a
long
list
I
have
filtered,
which
I
felt
more
relevant.
Let
me
show
you
the
list
of.
C
C
I
Just
a
heads
up
the
the
metrics
model-
template
is
kind
of.
A
I
This
this
doesn't
currently
match
where
we're
at
on
the.
What
a
metrics
model
would
look
like.
A
I
Based
on
emma's
work,
yeah.
E
A
I
I
kid
when
I
say
that
metric
is
is
the
most
important,
but
it
probably
is,
it
probably
would
be
expected
in
this
model
and
it's
easy
to
grab.
L
L
I
mean
like
like
like
likes
right
like
I
like.
C
E
I
We
could
possibly
we
could
possibly
treat
them
as
the
what
laura
dabish
calls
from
the
the
signals
of
attention
right.
The
thumbs
up
stars
likes
are
all
kind
of
signals
of
attention.
C
Stars
are
the
stars:
have
a
distinct
value
at
a
project
level,
though
yeah,
that's
fair,
just
because
they're,
it's
you
know
like
the
other
emoji
or
things
that
people
might
do
like
thumbs
up
on
a
comment
or
whatever
those
things
are
more
granular
star
is
basically
sort
of
a
thumbs
up
on
a
project
level,
at
least
on
github.
C
I
That
one's
a
little
difficult
to
define
because
mo
emojis
can
have
so
many
different
meanings
and
they
also
have
different
meanings
and
context
so
yeah
the
emoji
one.
We're
kind
of
we've
been
very
slow
to
to
roll
that
out,
because
I
think
we're
having
a
little
difficulty
in
how
to
frame
it.
I
F
Where
did
we
land
on
the
optimal
number
of
metrics
that
should
be
in
a
model.
C
I
mean
human
limitations
would
say
that
five
to
seven
is
perhaps
optimal
a
number
of
pieces
of
information,
and
it
depends
on
a
little
bit
of
the
complexity
of
the
information
being
presented.
So,
for
example,
stars
it's
not
very
complex
at
all.
It's
just
a
number,
but
burstiness
probably
will
produce
a
visualization
that
has
some.
C
You
know
you
can't
just
look
at
it
know
what
it
is.
You
probably
have
to
process
the
caption
and
look
at
it
for
a
few
minutes,
so
clones
is
also
just
a
number
so
like,
I
might
think
of
it
in
terms
of
human
attention.
If
we're
having
that
discussion,
you
know
how
much
the
things
that
are
just
numbers
are
easier
to
present
simply
than
things
that
are
complex
analysis
type
things
that
makes
sense.
E
So
my
my
take
on
the
question
of
the
number
is
at
least
initially
when
we're
building
metrics
models.
It's
to
to
really
just
help.
People
understand
that
collection.
Some
collection
of
metrics
can
provide
insight
around
a
thing
in
this
case
project
popularity
and
we're
not
going
to
capture
the
world
ever
and.
E
The
fewer
the
better
just
to
kind
of
help
people
move
in
the
right
direction.
We
can
always
add
metrics
later
if
we
are
finding
that
these
are
really
easy
to
deploy
and
it's
easy
enough
to
add,
for
example,
job
opportunities,
and
so
I
think
thinking
about
and
job
opportunities
is
also
probably
one.
That's
pretty
hard
to
find
like
data
wise,
and
so
we.
E
I
Think
there's
a
little
more
nuance
to
to
capturing
those
though,
and
100
complicates
the
metric,
so
yeah
definitely.
A
Even
in
the
popularity
it
was
discussed
like
when
you
search
on
a
google,
is
it
coming
on
the
top
list
in
the
first
five
or
first
two
or
three,
when
you
search
something
oh
seo,.
A
J
I
Or
they
might
be
paying
for
placement
or
yep
yeah.
I
I
would
avoid
any
just
any
anything
around
seo.
E
K
Can't
be
point
in
time,
but
it
can
also
be
like
around.
Like
you
know,
revision
schedules
or
you
know,
plant
upgrades
or
stuff
like
that.
Do
we
try.
A
E
F
A
In
the
model-
and
where
was
that,
I
think
this
was
the.
Let
me
show
you.
A
Even
request
was
as
a
popularity,
this
was
the
list
of
the
matrix
that
were
previously
there.
I
A
D
J
C
I
I
K
C
C
The
aggregation
of
seeing
a
lot
of
activity
on
a
project
is
meaningful
to
me
and
it's
different
when
I
start
a
project.
My
own
habits,
it's
once
cool
or
does
something
interesting
that
I
want
to
recognize
and
when
I,
and
they
do
call
it
watchers
on
get
lab
as
well.
It's
organized
in
the
ui
slightly
differently,
it's
really
about
when
you
will
be
notified,
but
watching
is
one
category.
I
Yeah
and
I
and
I
do
and
I
still
think
that
it
could
be
merged
with
the
the
stars-
badges
likes
metric,
so.
L
Two
conversations
that
are,
I
mean,
there's
sort
of
like
two
pretty
distinct
conversations
that
are
happening.
One
is
about
like
what
happens
with
stars.
I
think
that
it's
totally
correct
that,
like
the
star,
is
used
as
like,
at
least
like
the
single
star,
not
like
a
give
it
like
how
many
stars
you
give.
It
is
used
as
like
a
like,
a
synonym
for
for
bookmark
in
a
lot
of
user
interfaces,
right
like
when
you
start
a
thing.
It's
because
you're
essentially
attempting
to
follow
it.
L
There's
another
question
which
is
like,
which
I
think
is
a
pretty
good
one,
which
is
like
there's
a
bunch
of
things
here
which
are
describing
essentially
interest
in
using
the
project.
And
then
there
are
a
bunch
of
things
here
which
are
essentially
some
sort
of
like
passive
sort
of
interest
in
development
or
in
tracking
development,
like
I'm
interested
in
potentially
hearing
about
new
versions
of
this
or
I'm
going
to
clone
a
repository,
because
I
might
want
to
contribute-
and
those
do
seem
like
kind
of
distinct
sorts
of
things.
L
I
think
the
a
little
bit
more
there
might
be
more.
I
I
think
that
the
project
popularity
model,
I
think,
is-
is
really
kind
of
about
awareness
of
the
project
and
not
necessarily
activity.
That's
occurring
around
the
project.
Now
activity
that's
occurring
around
the
project
can
inform
awareness
of
the
of
the
project,
but
I
think
at
a
high
level.
This
model
is,
it
really
is
about
just
awareness
of
the
project
in
general
and
not
intention
to
any
intention
to
engage
or
intention
to
to
work
within
the
project.
L
I
think
that's
very
clear,
so
maybe
it
looks
like
someone's
editing
the
top
to
make
that.
I
A
I
don't
know
popularity
is
more
commonly
used
than
awareness
like
I,
I
like
the
thought
of
awareness,
but
the
name
depicts
more,
which
is
more
used
in
everywhere,
like
how
popular
is
the
project.
I
Yes,
so
the
popularity
is
a
bit
of
a
loaded
term,
so
people
don't
in
in
open
source.
I
don't
know
that
people
necessarily
like
it.
I
think
they
kind
of
think
it.
When
popularity
comes
up,
they
think
of
vanity
metrics,
and
you
know
metrics
that
don't
really
mean
anything
to
us
right,
so
the
awareness
of
a
project
that
I
think
that's
kind
of
meaningful.
I
E
I
do
I
mean
if
I
was
to
think
of
just
the
chaos
project
itself,
like
how
many
people
are
aware
of
the
chaos
project.
Certainly
one
thing
how
popular
we
are.
I
don't
know
you
know
I
I
prefer
awareness
as
well.
F
I
might
push
back
a
little
sorry,
I'm
gonna
be
that
person,
but
I
I
feel
like
what
we're
trying
to
capture
here
is
like
not
just
awareness
but
usage
and
lump
those
two
things
together
under
this
umbrella
of
popularity,
like
I
know
about
a
lot
of
different
products,
but
I
only
use
a
very
small
number,
so
I
guess
I'm
curious
what
what
do
we
actually
care
about
as
a
because
then
maybe
they're,
two
different
things?
Maybe
we
should
split
them
out
into
two
different
two
different
things.
I
I'm
not
sure
we're,
I'm
not
sure
we're
capturing
usage
in
the
model.
Currently,
maybe
maybe
forks
clones
usage.
L
L
That
technical,
forks
and
clones
are
not
measures
of
usage
or
not
necessarily,
I
think
that
they're
measures
of
of
like
sort
of
interest
potential
interest
in
contributing.
E
L
Is
I
mean
I
will
like
contradict
myself
a
little
bit?
There
are
projects
for,
for
which
installation
or
use
of
the
projects
involves
doing
something
like
cloning.
The
repository
right,
although
not
necessarily
in
a
way,
that's
public
right.
It
might
just
be
like
sort
of
cloning
it
that
is
more
like
a
download.
L
F
I
A
I
At
those
awareness
indicators
and
use
them
as
a
proxy
for
usage
right.
I
But
I
do
think
there
I
do
think
we
have
two
different.
I
think
it's
two
different
models
and
I
I
would
be
comfortable
moving
forward
with
a
project
awareness
popularity,
slash,
visibility,
metric,
but
I
think
it's
that's
separate
from
usage.
I
think
usage
is
a
it's
a
different
model
that
and
a
much
harder
model
to
work
on.
L
L
My
usage
models
in
the
past
I've
used
stuff
like
data
from
debian,
which
includes
like
installation
based
on
some
sort
of
like
opt-in
survey,
and
I
would
put
that
in
a
usage
model.
I
would,
I
would
think,
of
downstream
dependencies
as
a
different
metric.
That
would
capture
that.
So
that's,
but
I
think
everything
else
in
there.
I
would
consider
more
measures
of
awareness
or
visibility.
C
A
I
L
D
A
I
I
And
contributors
already
exist,
so
I
wouldn't
I
wouldn't
I
wouldn't
make
that
distinction.
I
Did
where
did
we
come
down
on
including
followers
in
the
stars
badges
bit
there
here.
I
So
we
probably
we
probably
need
to
take
that
that
metric
and
add
it
to
our
metrics
to
define
yeah
yep,
okay,.
A
A
E
K
Yeah
then
I
I
guess
I
would
lead
toward
awareness,
because
I
think
the
point
some
folks
have
been
making
about
popularity
being
a
a
knee-jerk
word,
for
some
folks
might
might
be
an
issue.
E
How
is
a
technical
fork
about
awareness.
C
A
I
Yeah
just
to
keep
them
on
my
on
my
radar
yeah.
L
I
don't
think
it's
a
direct
measure
of
awareness.
I
think
that
it
is
like
because
awareness
is
pretty
requisite
it's
like
sort
of
a
pro.
It's
a
good
proxy
in
the
sense
that,
like
I
don't
know
like
driving
to
the
airport,
it's
not
like
a
measure
of
like
taking
flying
on
airplanes,
but
like
it's
a
pretty
good
proxy
prerequisite
to
doing
it.
So.
E
My
only
my
my
thought
is
is
that,
like
the
examples
that
you're
giving
is
about
some
other
project,
but
if
I
just
look
at
like
the
chaos
project
like
if
we
looked
at
the
number
of
forks
that
we
had
on
the
chaos
project
like
if
I
looked
at
those
numbers
of
forks,
I
don't
know
how.
I
would
take
that
information
as
a
measure
of
awareness.
E
I
A
L
In
terms
of
just
like
relative,
like
like
yeah,
if
I
had
one
project
with
100
forks
and
one
with
ten
forks,
I
would
say
that
there's
more
awareness
about
the
project
with
the
hundred
ports
than
the
moment-
I
don't
know
I
don't
feel
strongly
about
this
to
to
be
clear
like
this
is,
like
my
I
don't
know,
epistemological
confidence
about
this
particular
assessment
is
pretty
low,
but.
I
They
indicate
they
may
indicate
usage,
but
they
also
indicate
some
sort
of
intent
to
engage
or
do
development
work
right.
So
at
the
very
least,
those
things
capture
intent
to
engage,
I
think
or
intent
to
use,
so
that
that
intent,
I
think,
is
the
can
be
a
proxy
for
awareness
right.
So
a
clone
or
a
technical
fork
is
we're
capturing
that
intent
to
engage.
E
K
Yeah,
I
I
think
I
think
forks
is
more
about
like
being
beyond
awareness.
It's
like
forks.
So
if
I
look
at
something
and
I'm
looking
at
the
number
of
forks,
it's
it's
telling
me
that
this
project
is
probably
reliable
or
important,
but
I've
already
been
kind
of
more
shopping
right.
You
know
it's
like
I'm
looking
for
a
solution
to
do,
abcdre,
I'm
looking
for
so
I've
kind
of
got
like
more
than
just
oh
there's
this
package
out
there.
I
don't
know,
I
I
think
you
probably
go
either
way
with
it.
I
Well,
if,
if
everyone
thinks
either
way,
then
I
think
we
should
probably
default
to
the
the
the
simpler
or
the
fewer
metrics
included
right.
It's
the
generally
that's
preferable
yeah,
I
prefer
fewer.
So
if
there's
not
strong
consensus
to
keep
it,
then
I'm
fine,
I'm
fine
getting
rid
of
it.
C
Expressing
cloning
is
cloning,
is
an
expression
of
I'm
going
to
use
this?
Possibly
it's
how
you
often
download
a
project
onto
a
server,
maybe
other.
Sometimes
you
take
a
released
version,
but
cloning,
it
kind
of
implies
you're,
going
to
use
that
clone
and
it's
a
measure
that
often
is
used
for
utilization.
If
you
don't
use
a
package
manager
and
forking
like
we
had
that
discussion
a
minute
ago,
I
mean
it.
C
I
Well,
so
I
think
there
there's
room
for
that
in
the
when
we,
when
we
describe
these
when
we
describe
these
metrics
and
how
they
fit
in
the
model,
we
can
talk
about
them
there.
So
you
don't
really
need
the
the
metrics
model.
It's
not
going
to
be
just
a
list
of
these
metrics.
I
We
we
do
dig
into
these
metrics
a
little
bit
deeper,
so
you
don't
need
the
disclaimer
here.
Those
those
can
be
described.
I
The
the
way
that
they're
a
proxy
for
these
things
can
be,
can
be
used
in
the
description
of
of
clones
that
we
use
in
the
model-
and,
I
would
say,
just
get
rid
of
technical
forks
and
keep
clones.
C
K
E
C
I
But
I'm
an
academic
also.
I
would
like
to
point
out
that
the
like
the
the
ease
of
being
able
to
capture
this
in
metrics
definitions
and
metrics
models
we're
somewhat
agnostic
about
it.
So
if
we
especially.
I
Well,
especially
in
metrics
in
metrics
definitions
right,
we,
we
define
metrics
all
the
time
that
are
very
hard
to
capture,
so
the
the
the
ease
of
the
ease
of
being
able
to
capture
it.
I
I
for
me,
that's
not
a
strong
consideration
and
yeah.
I
understand
how
you
you
would
have
a
different
opinion
on
that.
I
Well
then,
this
goes
back
to
the
beginning.
A
I
Discussion
that
led
us
down
this
particular
rabbit
hole,
and
that's
should
we
include
burstiness.
E
I
E
I
E
That's
my
concern
with
those
two
not
in
relation
to
burstiness.
For
the
moment
I
do
know
where
yeah
I'm
and
to
like
sean's
point
right.
I
mean
like
you,
could
like
break
down
the
new
issues,
perhaps
and
take
a
look
at
how
many
are
first-time
contributors.
You
know
like
first-time
people,
doing
new
issues
or
first-time
people
doing
new
change
requests,
but
we
would
have
to
say
that
so
that
was
that
was
the
point
on
those
I
know
we're
at
the
end
of
time.