►
From YouTube: CHAOSS Metrics Models Working Group Feb 28 2023
Description
Minutes from this meeting can be found here: https://docs.google.com/document/d/1qE2duE73NbEOHzRRAbqgPSUSAfdH-AmvLruOu7uCIYg/edit
A
Like
in
one
second,
you
and
I
never
know
all
right
well
welcome
it's
it's
the
same
day
for
everybody.
It's
Tuesday,
yes,
which
is
great.
This
is
the
metrics
model
meeting
for
February
28th,
it's
great
to
have
you
here.
Sean
has
been
putting
the
minutes
in
the
chat.
Thank
you
Sean.
If
you
could
click
on
that
and
add
yourself,
that
would
be
great
feel
free
to
keep
your
camera
on.
B
A
B
A
B
A
It's
just
updating
the
readme
a
bit.
We
had
some
old
deadlinks
just
because
of
the
website
update.
It
was
pointing
to
some
participate
pages
that
don't
exist
anymore.
I
updated
our
copyright
link
to
the
license,
all
that
kind
of
stuff,
so
I.
Also
we
had
kind
of
talked
about
this
I,
don't
know
what
other
people
think,
but
this
these
two
sections
of
contributors.
This
was
linking
nowhere,
so
I'm,
not
sure
quite
what
that
was
about,
and
then
we
had
our
list
of
maintainers
and
I
I
just
removed
it.
A
This
is
a
list
of
people
that
were
that
we
had
originally
set
up
across
the
working
groups
and
we're
not
doing
a
good
job
of
keeping
it
updated.
So
to
me
it
just
I,
don't
know
it
seems
silly
to
try
to
hold
on
to
it.
If
it's
just
kind
of
out
of
date,
I
don't
know
what
other
people
think
I
can
add
it
back
if
y'all
want,
but.
C
But
actually
I'm
I'm
fine
with
that,
because
there's
there's
no
fixed
person
who,
like
you
myself,
you
and
June,
should
attend
this
meeting,
like
everybody
from
chaos,
should
are
welcome
to
attend
this
meeting,
so
yeah
no
means
to
who
should
take
responsibility
to
do
something
in
our
meeting.
That's.
D
Usually,
usually,
the
maintainers
is
a
list
of
people
who
should
be
merging
the
PRS
within
this
working
group.
So
it's
not
really
about
attendance
of
meetings.
It's
it's.
Who
do
we
want?
Who
do
we
want
to
be
responsible
for
doing
a
review
before
we
before
we
merge
a
PR?
D
You
know
that's
how
we
tend
to
use
it
in
other
open
source
projects.
I,
don't
know
I've
mixed
feelings.
I
agree.
It's
it's
hard
to
maintain,
especially
given
just
the
the
type
of
project
that
we
are
in
general.
It
is
best
practice
within
open
source
projects
to
have
a
list
of
maintainer
so
that
if
somebody's
PR
gets
stalled
or
something
happens,
people
know
who
to
contact
we're
a
small
enough
Community.
Maybe
that's
not
a
problem.
A
D
A
Okay,
that's
helpful,
I.
Think
a
lot
for
our
maintainers
kind
of
to
that
point.
Don
is
we.
We
have
a
lot
of
people
in
and
out
of
this
project
too,
and
that's
it's
a
lot.
A
part
of
that
too
so,
like
sometimes
we'll,
have,
for
example,
Google
summer
of
code,
students
being
maintainers,
like
particular
aspects
of
things,
they're
working
on,
and
then
they
rotate
out
anyway.
It.
A
C
F
C
A
A
That
sounds
good,
so
I'll,
let
me
add
that
back
in
okay,
great
thank
you
and
then
one
of
the
other
problems
is
there's
a
link
to
a
contributing
file
in
the
in
the
GitHub
repo.
It's
a
it's
a,
not
a
great
file.
So
anyway,
you.
A
B
Group
in
in
other
work
group
would
do
they
have
also
to
update
their
their
maintenance,
like
this
yeah.
A
We
we
had
kind
of
talked
Elizabeth.
Do
you
remember
this
at
all,
like
we
had
kind
of
talked
about
getting
rid
of
all
of
this
get
stuff,
and
just
saying,
if
you
want
to
contribute,
you
know
attend
a
meeting.
Here's
a
link
to
a
repository.
Here's
a
link
to
discourse
like
you
can
contr
just
start
by
joining,
and
then
you
can
contribute
to
to
to
metrics.
You
can
contribute
to
software
like
this.
All
of
this
stuff
is
really
detailed
as
to
how
to
make
a
contribution
that
you
can
go
get
on
the
internet.
A
D
This
is
the
one
for
common,
so
this
is
just
what
we
happen
to
have
in
common
and
I
linked
to
it,
because
I
think
some
of
the
key
things
that
are
missing
out
of
the
contributing
guide
for
the
metrics
model
are
things
like
dco,
for
example,
right.
E
E
Yep
yeah,
it
doesn't
have
these
nice
Graphics.
This
is
nice.
D
Yeah
I,
don't
know
who
did
this
I
didn't
actually
do
this
bit,
but
it
seems
pretty
helpful.
D
I
I,
don't
like
that
I
think
I
I
agree
with
you,
as
you
should
just
link
to
that,
but
lots
of
people
put
them
in
the
contributing
guides
and
I.
Don't
think
that
they're
very
useful
okay,
because
people
either
know
how
to
do
all
of
this,
in
which
case
I'm
never
going
to
look
at
that
or
there
are
people
who
don't
know
how
to
do
all
of
this,
who
are
going
to
need
more
information
than
just.
D
For
anybody,
as
far
as
I'm
I'm
concerned,
I.
A
D
Yeah,
it
might
be
worth
looking
across
all
of
the
all
of
the
contributed
files
that
we
have
in
the
various
working
groups,
because
I'm
sure
that
the
different
ones
have
different
things
that
are
useful
and
not
useful
and
I.
A
D
F
D
Expects
to
see
a
contributing.md
file
in
every
every
repository
the
way,
so
the
way
that
we
do
this
in
some
projects
is,
we
have
it's:
okay
to
have
like
a
template,
contributing
file,
maybe
in
the
community
repo
in
the
beginning
of
all
of
the
contributing.md
files,
0.2
point
to
that.
Okay,.
A
D
Then
there's
always
additional
context,
like
the
metrics
models,
for
example,
are
a
little
bit
different
like
the
place
that
you
put
them
and
where
you
do,
the
pr
and
so
I
think
that
you
know
we
can
have
sort
of
a
if
you're
interested
in
contributing
to
chaos.
Please
start
here
and
link
to
whatever
we
want
our
main
one
to
be,
but
then
I
think
you
know
and
specifically
for
metrics
models.
Here
are
the
additional
things
to
them
and
then
the
same
thing.
D
F
A
A
All
right
issues,
I
just
want
to
take
a
look
at
these
real
fast
all
right,
so
we
have
a
couple.
There
was
a
new
one
opened,
so
we
have
this
one.
Remember
this
one.
C
This
this
guy
jump
up
into
our
matrix
model
meeting
a
couple
just
one
time,
if
I
remember
correctly
and
after
that
died,
he
just
disappeared
and
I.
Don't
know
he
he's
he's
gonna
continually
working
with
us
or
they
just
pop
up
some
ideas
about
kpl
okr,
something
like
that.
I've
only
met
a
lot
of.
B
D
Similar
similar
pitch
or
kind
of
okr
metrics.
B
D
F
C
A
The
other
problem
I
had
with
this
just
with
this
issue,
is
it's
so
many
links
out
to
so
many
things.
Yeah.
C
E
A
C
A
Thank
you
and
then
June
I
think
you
just
opened
this
one
recently.
A
B
A
C
You
know
we
have
some
implementation
of
matrix
model
under
our
matrix
model,
using.
C
B
A
Awesome,
thank
you,
everybody.
Okay,
so
that
was
PR.
We
made
some
progress.
Issues
made
some
progress.
Cool
great
on
the
list
was
the
sustainability
metrics
model,
the
some
I
don't
think,
I
put
that
there
in
the.
A
All
right:
well,
then,
I'm
gonna
put
that
to
the
side
for
just
a
second,
because
we
have
other
things
that
we
can
get
to
and
if
we
have
time
we'll
come
back
to
that
and
try
to
sort
that
out,
I
did
take
a
look
at
the
community
welcomingness
metric
model.
If
you
recall,
we
had
a
bunch
of
metrics
in
there,
so
a
few
things
that
I
did
Sean
to
your
point.
A
F
F
A
And
so
Don,
basically,
we
had
this
this
community
welcomingness
metrics
model,
and
it
contained
these
whatever
two
four
seven
metrics
right
here
plus
all
of
these,
so
it
was
becoming
kind
of
overloaded
from
a
metrics
model
perspective.
So
the
suggestion
was
to
pull
these
out
and
create
a
new
starter
risk
metric
model
that
just
kind
of
focuses
on
these
things.
A
So
that's
what's
going
on
here,
honestly,
when
I,
when
I
did
this
and
I
do
want
people's
feedback
at
this
point,
when
I
did
this,
these
metrics
didn't
just
like
jump
out
to
me
as
indicating
whether
or
not
a
community
is
welcoming.
A
F
And
and
online
quick
responses
and
an
open
source
in
particular,
quick
responses
are
associated
with
sustained
engagement,
so
I
can
I
can
see
that
the
those
those
are
metrics
that
you
could
use.
As
proxies
for
welcoming
this.
D
D
So
so
I,
don't
I,
don't
know
I'm,
not
sure
what
to
do
with
these.
We
do
tend
to
consider
most
of
these
metrics
as
a
part
of
like
things.
You
should
look
at
if
you're
trying
to
grow
your
community
and
sustain
your
your
existing
momentum
within
your
community.
These
are
things
that
we
encourage
people
to
look
at
so
like
the
basically
the
response,
time
sort
of
sort
of
metrics.
D
So
there
are
important,
but
not
yeah,
I,
don't
know
I,
don't
know
about
welcomingness
like
as
a.
B
A
C
I
have
feelings:
I
have
feelings
about
the
activity
that
should
be
just
reflective
about
where
companies
like,
like
all
of
you,
mentioned
issue
each
response
time
enforced
response
time,
a
what?
What
about
the
rest
of
the
part
I
mean
new
contributors?
D
These
feel
more
like
sustainability,
Community
sustainability,
metrics,
okay,
I
mean
I
would
think
that
welcoming
mess
might
be
more
like.
Are
you?
Are
you
greeting
people
are.
B
D
D
Those
are
ones
that
I
would
think
of
more
on
the
side
of
of
welcomingness
than
than
what
you
have
here.
I.
B
C
Oops
but
but
the
way
I
look
at
look
at
the
user
stories.
I
mean
they
lost
the
case
as
a
as
an
organizations,
I
mean
it's
not
just
a
welcoming
to
the
new
new
video
people,
but
also
for
the
organizations
which
means
it's
not
just
you
know,
activity
or
code
and
the
the
metrics
you
you
you
removed
about
the
license
which
is
licensed
friendly
for
corporations,
which
is
more
related
to
the
cooperation
where
companies
from
my
understanding,
yeah
I,
think
we
were
looking.
A
So
I
I,
okay,
so
I.
What
I'm
hearing
is.
Perhaps
this
could
be
reframed
as
sustainability,
Community
sustainability,
welcomingness
I
liked
how
you
were
putting
it
down
like
I,
think
of
the
things
we
do
in
the
chaos
project
like
having
the
Newbie
bot
in
slack
like
the
office
hours,
making
sure
that
we
have
guides
for
how
people
can
contribute
or
like
join
meetings,
that
kind
of
stuff
all
that
stuff.
That
I
think
a
lot
of
what
Elizabeth
and
Ruth
do
to
new
people
into
the
community.
D
I
think
I
would
avoid
the
word
sustainability,
because
every
time
I
see
it,
it
causes
confusion
because
people
don't
know
if
you're
talking
about
environmental
sustainability
or
contributor
sustainability,
yeah.
D
We
we
have
this
problem
in
the
cncf
all
the
time.
D
Look
at
it,
as
so,
it
depends
on
depends
on
how
you
want
to
take
this.
This
metric
right,
so
contributor
growth
is
one
way
to
look
at
it.
Community
growth,
which
implies
sustainability,
because
people
are
kind
of
sticking
around
over
the
long
term,
but
it's,
but
when
you're
talking
about
growth,
that
is
a
little
bit
different
because
then
you're
talking
about
specifically
like
increasing
things,
as
opposed
to
sustaining
things,
yeah.
B
A
A
A
All
right,
great
great
any
other
comments
on
on
this
again.
The
starter
risk
model,
I'll
I'll,
start
getting
that
into
the
template
here
shortly.
A
All
right,
I'm
going
to
turn
it
over
to
you
on
this
pull
request.
If
that's
okay,
sure.
C
Maybe
you
can
just
open
up
yes,
this
poor
request,
yeah
based
on
the
the
metrics
model,
this
starter
project
chaos
matrix
model.
My
colleague
schumbach
has
helped
us
to
implement
this
model
and
their
compass,
and
you
know
from
the
compass
for
us
metrics.
We
Define
the
four
Matrix
initially,
but
you
know
in
some
metrics
and
it's
actually
related
to
two
things
for
for
wines
for
issue.
The
other
is
for
the
and
the
pull
request
or
change
request.
So
we
added
this
lecture
Matrix.
C
C
Yeah
and
also
about
it
the
close
time
it's
going
to
be
the
easy
close
time
and
and
the
pull
request
close
time.
So
we
add
it,
then
posts
so
easy
open
time
and
the
forecast
to
open
time.
A
C
C
D
C
The
definition
it's
like
we
just
want
to
calculate
the
time
between
the
poor
Quest
when
it's
created
until
it's
closed.
Okay.
The
thing
is
that
you
know
in
the
past
90
days,
for
example,
if
there
are
10
protocols
created
and
by
being
just
closed
it
within
one
day
for
the
rest
of
the
other
podcast
is
haven't,
been
handled
or
closed
for
for
like
certain
days
or
40
days.
C
So
if
we
only
consider
that
the
closed
time
saying
what
we
got
is
just
one
one
day
in
average
or
median
time,
but
if
we
calculate
the
open
days
together
with
other,
you
know
nine
protocols
who
haven't
been
closed.
Yet
in
the
past
90
days.
They
exactly
know
how
efficiency
of
this
progress
to
precise
situation
in
this
community.
So
that's
why
we
Define
this
as
a
pull
request
open
time,
instead
of
close
time,
because
in
in
many
cases
we
saw
that
if
we
only
considered
the
close
time,
it
would
misleading
something
important
yeah.
D
No
I
absolutely
agree
with
this
I'm
I'm,
actually
using
this
very
differently
than
than
the
way
I've
defined
it
in
the
metrics
model
and
I'm.
Trying
to
remember
exactly
why
I
did
that.
I
think
it
was
partly
because
we
we
didn't,
have
a
metrics
a
metric
defined.
Exactly
so
sorry,
let
me
just
back
way
up
so
when
I,
when
I
used
this,
what
I
was
trying
to
get
at
with
this
one
with
time
to
close
is
that
it's
not
actually
time
to
close
that
I'm.
D
Looking
at
I'm
looking
at
in
a
particular
time
frame,
do
you
close
about
as
many
issues
as
you
have
open?
So
it's
basically
about
keeping
up
with
your,
in
my
case,
pull
requests,
but
to
your
point,
I
like
that
you're
looking
at
issues
and
pull
requests
together
because
I
think
that's
a
better
picture,
but
you
know
I'm
kind
of
this
one
for
me
was
more
about.
Are
you
keeping
up
with
with
things
and
and
the
reason
I
used?
D
Clothes
instead
of
open
was
because
I
want
to
encourage
people
to
close
pull
requests,
even
if
they're
not
going
to
merge
them,
because
what
you
get
a
lot
in
open
source
projects?
Is
people
just
neglect
the
pull
requests?
They
don't
know
what
to
do
with
forever,
and
so
they,
just
they
just
sit
around
with
this
pile
of
neglected,
pull
requests
that
realistically,
are
just
never
going
to
be
merged,
but
nobody
wants
to.
Nobody
wants
to
tell
someone
who
worked
really
hard
on
something.
Hey
I
can't
take
this
because
right.
B
D
The
idea
was
I
used
closed,
because
I
I
wanted
to
encourage
that
that
behavior
I
also,
you
completely
agree
with
you
that,
if
you're
just
looking
at
time
to
close,
it's-
probably
not
a
great
a
great
metric
for
the
reason
that
you,
the
reasons
that
you
talked
about
so
I'm,
not
sure
I,
guess
I'm
not
sure
what
what
to
do
with
this
one.
C
D
Yeah
I
think
maybe
we
just
pick
the
most
likely
thing
and
and
cover
that
in
the
Encompass.
That
would
be
that
would
be
kind
of
what
I
would
think,
but
but,
like
you
said,
I
think
I
think
that's
a
really
important
point
that
you
made
about
the
issues.
We
do
this
a
lot
within
cnci
projects
where
we'll
open
an
issue
to
have
some
gigantic
big
discussion
about
something
and
those
are
intended
to
live
for
months
in
a
lot
of
cases.
D
So
closing
those
quickly
isn't
a
isn't
a
best
practice,
for
example,
especially
and
like
you
said,
the
same
thing
is
true
around
you
know:
Feature
Feature,.
C
Time
to
close
okay!
So
how
about.
C
Except
for
the
except
for
the
time
to
close
metric,
we
add
another
metric
to
support
the
place
you
mentioned
like
a
commanding.
Pro
requires
to
be
created
in
the
past
90
days.
Meanwhile,.
C
You
know
part
of
being
has
been
closed
or
rejected
or
anime
is
handled.
Finally,
in
the
past
90
days,
we
can
at
least
ratio
and
together
with
the
pull
request
time
to
close
this
metric,
it's
more
or
less
to
cool.
This
reflect
the
current
status,
at
least
in
the
party
United
in
the
past
90
days,.
D
Yeah
and
the
way
the
specifically
the
way
I
measure
this
is
I.
I,
look
so
I
look
at
a
month.
I
do
this
for
for
across
each
month
for
12
months,
but
I
look
at
the
total
number
of
pull
requests
in
that
month
and
then
I
look
at
the
total
number
of
pull
requests
that
were
closed
in.
D
C
D
C
A
F
A
C
Is
more
like
the
open
support
request
open
time
we
implemented
for
for
this
one.
A
Okay
could
could
I
have
something
then
could
either
Don
or
yahui
somewhere
in
here
in
the
minutes.
Could
you,
like
just
add
a
brief
narrative
as
to
what
it
is
that
you
would
be
looking
for
in
a
metric.
D
And
you
can
kind
of
see
it
like.
If
you
look
at
the
the
graphic
graphic
in
that
metric,
you
can
kind
of
kind
of
see
it
I
think
which.
A
D
So
it's
it's
this
one,
it's
time
to
close,
so
you
can
see,
there's
a
total
pull
request
and
then
the
green
dotted
line
is
the
ones
that
were
closed.
C
How
about
the
way
item?
Another
actual
noise
about
time
to
close
right,
because
in
currently
we
saw
two
lines.
One
is
you
know
the
created,
the
total
integrated
in
total
and
the
the
the
other
line
is
about
to
close
the
broadcast,
and
we
can
either
actually
learn
about
the
time
to
close.
D
D
Poked
some
polls
in
it,
because
it's
not
about
time
to
close
I,
don't
actually
care
how
long
it
takes
to
close
a
pull
request,
because
there
there's
sample
requests
that
are
just
getting
emerged
or
closed
immediately.
There
are
some
that
are
going
to
hang
around
for
a
long
time,
because
you
know
the
nature
of
a
particular
type
of
pull
request.
Yeah.
F
D
I,
what
I
care
about
is
that
they're
keeping
up
with
the
pull
requests
because
they're
closing
about
as
many
as
they're
getting
in
okay,
which
is
which
is
a
different
metric
than
what
I've
described
here.
But
you
can
kind
of
see
based
on
the
graph,
what
I,
what
I
meant
by
it.
A
D
So
how
about,
if
I,
let
me
think
about
this
just
a
little
bit
and
how
about
if
I,
propose
something
that's
more
like
what
I
actually
meant
and
propose
it
just
as
an
individual
metric
within
common
and
then
once
once
we
get
that
proposed
and
we
decide
that.
Yes,
it's
reasonable,
then
we
can
update
the
metric
model
to
use
the
other
metric
instead
of
time.
To
close,
does
that
seem
like
a
reasonable
approach.
C
D
Yeah
this
has
been.
This
has
been
a
great
discussion
for
sure
and
we
probably
wouldn't
have
had
this.
Had
you
not
started
to
implement
it
in
compass,
so
this
is
yeah.
This
is
great.
C
And
also
I
have
another
question
all
in
the
all
the
metrics
under
this
matrix
model,
so
timeline
we
defined
in
the
past
90
days,
but
what
but
you
just
mentioned.
What
do
you
care
about?
Is
one
Mouse
in
your
time
time
frame.
So
is
it
too
much
difference
between
in
the
past
90
days
or
in
the
past
the
short
readings?
So
no.
D
I
think
that's
fine,
if
you
think
90
days
seems
reasonable
for,
for
your
purposes,
I
think
we
can
use
that
Encompass
I,
think
people
people
adjust
time
frames
based
on
what
they
want
to
look
at,
and
you
know
90
days
is
perfectly
reasonable.
It's
you
know
roughly
a
roughly
a
quarter.
C
D
B
E
You
said
this
Don,
but
is
there
a
number
that
you're
looking
for?
That
would
be
like
a
like
best
like?
What's
your
target
number
for
that
total
number
of
PRS
in
a
month
minus
the
total
number
close
I.
D
E
D
For
example,
super
high
volume
project,
like
kubernetes,
is
going
to
have,
if
they're
going
to
get
behind
more
quickly
than
a
small
project
that
gets
a
handful
of
pull
requests.
Okay,
so
what
I
look
for
is
the
trend,
so
you
know:
are
they
continuing
to
close
about
as
many
as
they
have
open,
or
is
that
Gap
getting
really
big
and.
B
D
E
A
C
It's
about
a
large
event
of
Oasis
Compass
had
happened
just
based
in
the
last
Tuesday,
oh
yeah
Tuesday,
so
I
would
like
to
post
it
in
the
near
chaos
blog
and
the
main
produce
that
the
whole
thing's
going
going
on
under
this
event,
and
also
mentioned
that
chaos
would
collaborate
with
with
compass
to
broadcast
metrics
and
matrix
model
all
over
the
world,
especially
in
China.
A
Well,
yes,
Elizabeth
I,
think
that's
that
shouldn't
be
a
problem
getting
this
posted
as
a
Blog
and
is
there
anything
you
need
Elizabeth.
E
I
think
I
think
that
we
can
get
this
posted
no
problem.
What
I'll
do
yahoi
is
post
a
link
to
this
in
our
Communications
slack
Communications
Group,
just
so
they
kind
of
know
and
I'll
put
it.
We
have
a
spreadsheet
where
we
keep
track
of
what's
coming
up
and
so
I'll
put
it
in
there
and
then
somebody
will
look
at
this
and
just
make
sure
that
you
know
it's
good
to
go
and
then
we'll
I'll
go
ahead
and
post
it
and.
E
C
I
I'm
a
wild,
Bowser
I
thought.
There's
one
lady
from
my
from
my
team
write
about
it
together
with
me:
I
can
add
her
name
and
it
is
but
I
I
I'm,
not
sure
the
blog
post
should
need
a
name
or
author
name.
So
I
didn't
add
any
name
of
us
here.
We.
E
A
E
It
is
yeah,
it
goes
to
Twitter
automatically
and
then
we'll
post
to
LinkedIn
and
Mastodon
as
well.
Yeah
it'll
go
in
the
newsletter
and
then
it
kind
of
also
links
to
slack.
We
can
start
a
new
discourse,
a
thread
as
well.
We
usually
do
one
with
a
blog
post
too.
So
hopefully
that
covers
some
bases.
A
All
right
is
that
good,
you
Hui!
Thank
you.
That's
right,
okay,
great
well,
we're
at
the
end
of
our
time.
This
was
great.
I
enjoy
the
new
meeting
time
personally,
which
is
great
so
all
right.
Well,
thank
you.
Everybody
for
just
some
amazing
stuff
and
a
great
conversation
today,
I
think
we
all
have
a
little
bit
of
some
things
to
do
and
we'll
see
you
all
in
a
couple
weeks
on
this
call
all
right.