►
From YouTube: CHAOSS Metrics Models WG August 16-17, 2022
Description
Links to minutes from this meeting are on https://chaoss.community/participate.
A
B
A
A
B
A
B
A
A
A
A
D
B
A
No,
I
think
that
we,
I
think
that
ultimately,
we
need
a
development,
responsiveness
and
a
community
responsiveness,
metric
and
I'll
kind
of
talk
about
that
parametric
model
um
in
a
second.
So,
okay,
um
we
have
these
other
ones.
So,
for
example,
we
have
dei
project
badging,
but
we
have
dei
event
badging
up
and
community
engagement.
A
A
This
seems
to
be
where
most
of
our
conversations
have
occurred
over
the
course
of
the
last
many
weeks
that
we
talk
a
lot
about
development,
metrics
models
like
code
development
or
software
development,
metrics
models,
community
engagement,
metrics
models
and
then
metrics
models
around
sustainability,
for
example,
in
influence.
Remember
we
had
that
nice
conversation
from
frank
and
shoya
around
influence,
so.
D
uh
I
have
one
question
actually
because
uh
most
of
the
considering
uh
metrics
model
is
comfortable
comfort,
it's
coming
from
the
matrix
original
matrix,
how
to
define
it
with
uh
with
you
know,
combined
with
uh
some
other
metrics.
You
know
currently
we
treated
as
a
matrix
model,
but
um
but
there's
no
other
metrics
to
support
this
metric
model.
B
A
D
A
They're,
just
they're
the
focus
areas
are
really
just
a
way
for
us
to
organize
our
work.
That's
it
like
they're,
not
really
seen
by
other
people.
I
didn't
mean
to
make
this
like
a
big
deal,
but
like
it's
just
a
way
for
us
to
kind
of
organize
our
work,
and
I
think
sometimes,
when
we
have
too
many
focus
areas,
the
organization
looks
broad
and
and
fanned
out,
okay
enough
of
that,
um
I
I
I
know
that
as
part
of
this,
irrespective
of
where
they
are
within
a
focus
area.
A
A
A
So
I've
been
trying
to
capture
a
lot
of
or
like
forums,
and
so
we
kind
of
had
a
couple
different
metrics
that
we
were
looking
at
with
respect
to
responsiveness,
the
challenge
that
so
I
took
a
look
at
what
you
had
done
emma
and
I
I
was
putting
it
into
our
new
template
that
we're
using
for
the
metrics
models
and
one
of
the
challenges
that
I
had
was.
We
didn't
necessarily
have-
and
I
think
this
is
your
point
yahui.
B
A
A
What
I
did
was,
I
took
a
look
at
it,
some
of
the
things
you
had
proposed
emma
and
then
also
what
we
had
from
a
released
metrics
perspective,
because
the
metrics
models
are
meant
to
to
take
the
released
chaos
metrics,
the
ones
that
we
have
published
and
put
them
into
practice,
and
so
I
was
taking
a
look
at
what
we
had
with
respect
to
responsiveness
just
around
software
development.
At
this
point-
and
these
were
the
ones
that
that
I
had
seen
that
we
have.
A
B
Well,
so
with
issue
response
time
is
basically
the
so
I
create
an
issue.
The
response
time
is
whenever
somebody
on
the
project
in
a
maintainer
role
makes
a
comment
or
addresses
it
right.
So
that's
a
comment,
and
I
think
when
you're
talking
about
comment,
responsiveness
at
the
top
you're
talking
about
okay,
when
somebody
does
that,
how
long
does
it
take
for
a
subsequent
comment
to
occur?.
B
A
Exactly
yep
on
the
same
page,
yep,
and
so
this
would
then
be
whatever
development
responsiveness,
and
this
is
a
metric
model.
This
is
a
metric
model
that
is
working
off
of
the
stuff
that
emma
had
provided
to
just
think
about
responsiveness,
and
so
we
have
these
metrics
available
to
us,
and
then
that
was
my
point
of
view
like
that.
Would
move
this
one
into
this
one
see
what
I'm
saying
sharp.
A
Right
exactly
when
I
present
it
that
way,
but
but
then
one
of
the
things
that
it
did
was
it
it
lost
some
of
what
what
was
in
the
original
metrics
model,
which
was
the
response
time
to
say,
forums
or
discussion
boards
that
was
taken
out
because
we
didn't
have
that
metric.
You
know
what
I
mean,
and
so
I
do
think
there's
like
something
to
be
said
for
I'm.
B
A
B
A
B
uh
Going
back
to
your
original
question
about
the
project
health
101,
you
know
the
more.
I
look
at
that
spreadsheet
and
think
about
it,
like
you
could
argue
that
any
of
the
metrics
models
would
be
project
health
101.
So
it's
kind
of
too
broad,
I
think
so.
Maybe
we
could
just
get
rid
of
it
and
plot
these
somewhere
else.
B
D
D
A
um
So
that
would
that
would
at
least
again
I'm
just
trying
to
capture
the
things
that
we
had.
We
had
so
many
moving
parts
or
really
when
we
were
doing
these
metrics
models
like
with
the
different
templates
with
the
different
approaches
we
were
kind
of
trying
to
take.
You
know
what
I
mean,
so
I'm
trying
to
just
kind
of
bring
this
together
with
with
the
kind
of
the
way
we've
been
working
now.
Okay,
um
could
somebody
maybe
take
a
look
at
this?
A
um
Like
I
said,
I
retain
right
at
the
top
is
still
the
stuff
from
from
emma,
because
I
think
a
lot
of
the
text
still
worked
um
even
for
development
responsiveness.
We
don't
have.
This
would
then
require
this
stuff
right.
You
know
what
I
mean.
I
don't
have
any
of
this,
so
this
metric
model
would
have
to
have
a
deployment
of
some
sort
where
we
give
an
example
of
what
this
could
look
like
you
know,
but
at
this
point
we
would
actually
have
metrics
to.
A
If
you
recall-
and
so
there
was
a
proposal
of
like
a
way
to
measure
project
engagement,
which
is
a
metric
model
in
the
sustainability
focus
area
and
there's
an
example
here
about.
How
do
you
remember
this
yeah
how
how
project
engagement
could
be
measured
through
a
variety
of
different
weightings
and
then,
ultimately,
so
it's
kind
of
like
a
case
study
almost
on
the
open,
euler
um
community.
I
think
that's
what
a
lot
of
these
bottom
parts
are
really
just
becoming
they're
almost
like
case
studies,
you
know.
B
A
B
B
B
Are
they
really
directly
derived
from
data,
or
are
they
something
that
a
person
would
add?
After
looking
at
the
data
presented
in
the
parts
above
it
I
mean,
I
guess
the
visualizations
would
yeah
yeah
clearly
be
something
but
the
other
to
feel
like
uh
like
sort
of
a
contextualized
interpretation,
but
maybe.
A
A
B
B
You
know
if
I
was
to
pick
four
metrics
that
told
me
a
story
about
how
engaged
a
repo
or
a
community
is
with
its
contributors.
These,
you
know
this
response.
Responsiveness
indicators
are
pretty
strong
signals
about
how
interested
and
welcoming
a
community
is
with
regards
to
outside
contributions
right,
so
that
would
be
so.
A
Those
would
be
the
insights
that
you
use
yeah
yeah,
that's
just
it,
and
so
then
we
also
have
this
insights
drawn
from
specific
metrics
in
the
model.
So
in
the
open
euler
case,
there
was
a
different
case.
I
maybe
it
was
the
the
influence
case.
There
was
also
conversations
about
what
each
metric
might
be
telling
us
individually
about
this
community,
not
just
collectively,
but
what
what
one
might
be
telling
us
as
well
in
isolation
yeah
that,
and
so
it
would
be,
it's
it's
an
interpretation
that
we
provide
on
the
data
just
to
kind
of
help.
C
I
sometimes
give
the
example
of
like
responsiveness.
The
goal
isn't
to
necessarily
be
able
to
close
everything
in
like
five
seconds,
especially
if
you're
like
looking
at
the
intersection
of
like
burnout
right
like
if
your
maintainers
like
if
their
metric
around
burnout
is
really
high
and
your
responsiveness
is,
you
know,
maybe
there's
some
things
we
could
de-prioritize
around
responses
like
there's.
It
opens
up
really
interesting
conversations.
Anyways.
A
A
A
B
D
B
B
The
other
way
to
do
it
would
be
to
weight
them
and
provide
a
singular
metric,
and
I
don't
have
a.
I
don't-
have
a
religion
about
either
approach.
I
think
I
think
you
would
tend
toward
the
waiting
if
you
were
looking
at
a
large
number
of
projects
and
trying
to
identify
differences
between
them,
because
it
would
be
easier
to
look
at
a
weighted
number
for
one
weighted
number
for
all
those
projects
compared
to
four
numbers.
A
A
A
What
june
had
written
um
soon
you're
on
I'm
really
glad
that
you're
on?
Thank
you,
um
and
so
I
took
a
lot
of
what
um
what
you
had
done
and
tried
to
get.
It
tried
to
fit
it
into
the
template
that
we
have.
So
I
might
need
you
to
take
a
look
at
um
what
I
had
done
with
the
language
that
you
had
typed
to
get
it
to
fit.
A
A
A
A
A
A
A
D
A
D
A
A
You
know
the
metrics
that
we
find
in
the
model,
and
I
was
trying
to
look
at
the
metrics
that
you
have
down
below
the
parameters,
but
then
in
the
implementation
we
describe
in
detail
what
you're
doing,
but
in
detail
what
we
mean
in
this
particular
case,
about
committers
and
what
we
mean
in
this
particular
case
about
contributors.
Is
that
right?
You
huey?
Yes,
okay,
good.
B
C
So
would
you
expect
to
have
like
a
s
like
a
single
score?
I
started
to
step
away
because
I
someone's
3d
print
didn't
work.
um
Okay,
uh
like
a
score
like
how
would
you
say
like
all
right,
we
are
active
or
or
is
it
just
that
you
want
people
to
go
through
each
one
or
what
does
it
look
like
for,
like
a?
A
Yeah,
so
there
is
a
single
score,
and
so
at
this
at
this
point
and
the
waiting
can
be
specified.
As
far
as
I
understand,
the
waiting
as
proposal
can
be
specified
by
a
a
community
manager,
so
there
are
things
that
are
considered
of
higher
value
temporarily.
You
know
what
I
mean
for
a
particular
context.
B
If
you're
looking
deeply
into
five
repositories,
I
would
suggest
not
waiting
them
and
just
looking
at
the
raw
stuff
to
get
a
real
sense
of
what's
happening.
But
if
you're
looking
at
a
100
repositories,
that's
you
know,
you
probably
don't
have
the
two
days.
It
would
require
to
look
at
all
that
stuff.
You
want
something
more
summary
level.
C
Oh
sorry,
I'm
just
trying
to
understand
how
someone
would
so,
if
I'm
like
a
maintainer
of
a
project
and
like
I'm
thinking
of
this
from
an
awesome
perspective,
you
give
me
this
score
and
I
can
see
how
things
are
weighted.
It's
also
behavioral
like
we
want
to
drive
that,
but
it
could
change
with
that.
C
Would
you
also
say
like
this
is
just
which
makes
sense
like
I,
I
think
priorities,
change
and
we'd
help,
so
I'm
just
sort
of
wondering
like
yeah
like
it
would,
um
would
it
be
motivating
for
people
to
try
and
change
their
number.
This
is
more
of
like
a
psychological
question.
Is
it
motivating
to
tell
them
a
number
if
things
could
change
or
like
how
do
you
see
that
playing
out?
As
far
as
like
the
actions,
people
would
take
out
of
curiosity.
B
D
Maybe
it's
not
even
for
the
specific
people
who
care
about
the
single
matrix
of
this
matrix
model.
It's
it's
about.
A
group
like
a
specific
interesting
group
exists
in
the
um
community.
If
we
have
more
than
20
sikh
group,
I
mean
the
specific
specific
interesting
group.
We
have
to
calculate
the
whole
activities
or
engagement
of
of
these
operations
for
the
of
the
six
in
this
community.
D
A
D
D
A
A
A
B
B
A
A
C
um
I
did
you're
up
someone
else
put
update
question
mark,
so
I
just
tried
to
elaborate
not
very
until
actually
well,
I
thought
I'd
just
come
and
share
because
you
were
also
like
great
at
um
joining
us
in
some
chaos
inside
of
microsoft
calls
last
year
just
to
show
you
where
we
are
right.
Now
it's
not
to
be
like
here.
You
go.
Here's
like
this
package
and
but
I
thought
it
might
be
interesting
because
one
of
our
goals
is
to
contribute
back
to
chaos,
some
of
our
work
so
um
yeah.
C
C
B
C
uh
You
know
to
connect
around
asking
the
right
questions,
but
um
they
didn't
have
time
to
spin
up
like
the
answer.
Part
like
the
you
know
or
or
like
the
investment
in
that
was
really
unclear.
So,
for
example,
um
you
know,
responsiveness
like
you
know,
we,
I
kind
of
presented
auger
and
some
other
dashboards,
but
writing
like
juniper.
It
was
just
too
much
so
I
sort
of
stepped
back
for
a
second
um
and
um
asked
myself.
C
It's
it
still
kind
of
falls
on
me
to
kind
of
find
those
types
of
methods
and
bring
it
in,
but
it
was
a
really
good.
It's
been
really
really
good
for
getting
people
to
kind
of
grok
the
work
right
um
really
like
you
know:
here's
security,
here's
a
set
of
best
practice
questions,
and
some
of
this
came
from
some
of
the
work
that
I
I
haven't.
C
uh
That
was
one
of
the
reasons
I
came
today
to
kind
of
update
myself,
but
you
know
some
of
the
questions
that
we
were
asking
like:
the
open,
sfs
practice
practices
and
there's
the
sort
scorecard.
So
there's
some
actionable
work
here,
but
even
still,
I'm
sorry,
I'm
just
recovering
from
a
not
coveted
cold.
C
So
I'm
a
little
bit
uh
hermie,
um
but
so
it's
organized
visually
was
one
thing
that
seemed
to
be
really
helpful
in
conversations
um
that
there's
a
list
of
questions
that
was
tied
to
the
chaos
was
really
important,
because
I
can
say
this
is
peer
reviewed.
This
is
our
in.
These
are
our
peers
in
the
ecosystem.
This
is
you
know
validated
through
research
and
and
all
these
types
of
things.
Just
not
me
emma
erwin
saying
these
are
the
the
questions
that
you
should
be
asking,
um
and
then
we
started
talking.
C
So
we
had
the
structure
we
had
some
initial
kind
of
conversations
and
and
in
the
ospo
something
that
we
really
wanted
to
be
able
to
do
is
is
say.
Like
you
know
your
project,
you
know
kind
of
meets
the
gold
standard
for
what
we're
looking
for
um
and
then
we
start
so
then
I
started
to
ask
myself
like
what
does
the
goal
standard
mean?
What
would
that
even
be
so?
C
ah
It's
down
here
um
yeah.
This
is
like
the
meta
task,
so
this
is
like
the
pilot
project
that
I'm
running
right
now,
which
is
basically
you
know.
We
have
a
solid
set
of
peer,
validated
questions.
This
is
chaos
project,
um
but
we
need
to
identify
a
subgroup
of
these,
and
this
is
kind
of
like
more
of
a
cross
section
between
all
of
those
different
metrics
areas.
C
We
want
to
say
like
this
one
from
security,
this
one
from
um
um
community
or
contribution,
and
this
one
like
like
kind
of
pluck
out
some
of
those,
and
um
so
we,
our
key
our
first
key
result
was
like
find
you
know,
12,
it
ended
up
being
10
metrics
that
were
like
yes
like.
If
we
can
get
people
looking
at
these
and
then
working
to
improve
these,
then
we're
going
to
have
progress
and
also
we'll
start
to
create
a
shared
language
around
like
what
it
means
to
have.
C
It's
not
going
to
be
everything
we're
not
going
to
claim
that
it's
everything,
but
it's
going
to
be
a
really
solid
start
and
also
becomes
something
that
people
can
take
to
their
managers.
There's
a
lot
of
um
challenges
around
behavioral
changes
that
are
like
how
do
we
make
it
more
carrot
and
less
stick
right
like
if
people
are
able
to
talk
about
their
success
moving
the
needle.
You
know
all
that
kind
of
behavioral
stuff,
and
then
we
want
one
method
for
answering
each
one
of
these.
C
Originally
they
said
deployable
in
under
two
minutes
or
something,
but
we've
just
removed
people
deploying
by
themselves
altogether,
because
honestly,
two
minutes
is
too
long
um
and
then
we
built
this
prototype
so
right
now.
What
this
pilot
looks
like
is
that
we
have
this
first
set
of
foundational,
metrics
and
I'll.
Just
we've.
We've
agreed
on
these
at
this
point,
so
security
we're
answering
what
is
our
response
time
for
security
alerts?
C
I
don't
think
that
one
is
in
chaos
yet,
but
we
have
the
depend
about
alerts
and
whenever
I've
gone
through
and
interviewed
all
the
and
I
can
figure
out
how
to
contribute
these
two
um
all
of
the
security
folks
at
intro
are
like.
Yes,
that
would
be
really
helpful
in
moving
us
to
more
towards
a
more
compliant
organization.
C
um
The
open
ssf
uh
scorecard,
which
I
think
was
in
our
initial
security
matrix
model
and
then
this
kept
coming
up.
I
was
interviewing.
I
didn't
even
ask
uh
security
folks,
this
question,
but
it
just
kept
coming
up
two
reviewers
two
reviewers
there
should
be
two
reviewers,
so
I've
just
added
that,
and
we're
also
trying
to
do
just
as
a
side
note
that
I
won't
even
get
into
an
inner
source
and
open
source
version
of
this.
C
This
is
just
the
inner
source
repository
compliance
hugely
important
for
ospo
safety,
really
just
focused
on
are
people
being
trained
responsiveness
and
a
couple
questions
that
I
also
pulled
from
your
metrics,
and
so
those
are
the
we're
going
to
be
starting
with
these
four
and
I'll
show
you
the
the
draft
that
report
and
we're
organizing
them,
sorry,
organizing
them
by
bronze
silver
and
gold
and
gold
to
kind
of
eventually
get
to
that
answer.
Asking
that
question
about
gold
level.
Sorry,
sean.
C
So
so
then,
so
then
silver
and
you
know
we're
trying
to
go
from
most
critical
to
like
not
least
critical,
but
then
documentation,
usage,
discovery
and
gold
is
really
also
those
harder
things.
So,
um
but
we
do
have
a
survey
that.net
has
used
before,
and
you
know
the
idea
was
that
we'd
be
able
to
deploy
it
on
multiple
organizations
um
and
that
might
be
more
of
a
self-serve,
so
dan
mosley,
who
met
he
was
like
I'd
love
to
be
able
to
deploy
this
survey
on.
C
Like
you
know,
I
did
it
on
38
repositories,
but
it
you
know
it
was
manual
and
I
want
to
be
able
to
just
come
to
a
dashboard
and
you
know
select
which
repositories
I
want
and
run
it.
So
that's
why
it's
at
the
gold
level,
because
it
would
be
highly
valuable,
but
we
need
engineering
time
and
right
now,
like
in
leadership,
burnout
and
funding.
I
think
I
just
cut
funding
off,
though
the
other
day,
because.