►
From YouTube: CHAOSS Metrics Models Working Group August 30-31, 2022
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
A
All
right
so
minutes
are
in
the
chat.
If
you
could
add
yourself,
that
would
be
wonderful,
so
I
think
the
first
thing
I'd
like
to
do
is
kind
of
revisit
not
revisit,
but
take
a
look
at
a
few
of
the
metrics
yehui.
These
were
ones
that
you
had
put
on
the
asia
pacific
call
last
time.
Let
me
make
that
a
little
bit
bigger
for
everybody
that
we
just
didn't
have
time
to
get
to.
A
So
I
thought
I'd
make
sure
that
we
have
time
to
get
to
those
today,
all
right,
so
maybe
I'll
start
at
the
bottom.
Just
because
I
know
that
these
are
a
couple
that
yahui
had
been
working
on.
So
the
first
one
is,
is
and
support-
and
I
did
some
rearrangement
in
the
our
spreadsheet
and
so
community.
C
A
C
A
Okay,
is
there
anything
that
you
would
like
to
draw
our
attention
to
or
kind
of
point
out
in
this
model.
C
Actually,
the
whole
format
is
pretty
much
similar
with
with
the
wizard,
with
with
the
matrix
model
template
we
defined
for
okay
for
the
model,
and
if
you
go
slap
down
the
scroll
down
the
yeah
here,
we
we
still
hired
two
communities
anonymous
communities
to
to
take
this
matrix
model
to
has
dating
sites
work
on
that.
C
Okay,
cool
because
we
have
been
verified
meanwhile,
because
we,
we
also
verified,
verify
this
matrix
model
on
many
communities
projects
through
the
compass
work,
so
we
have
a
lot
of
examples
of
with
real
real-world
data.
So
I
just
pick
up
two
communities
to
give
the
data
inside.
C
The
committee
service
and
support
is
going
to
tell
their
people
what
kind
of
service
and
support
that
developers
join.
This
community
could
be
and
could
feel
directly
instead
of
the
inside
instead
inside
them.
Sorry,
I
also
didn't
sleep
very
well
last
night,
so.
C
So
basically,
you
know
our
our
community
could
provide
a
lot
of
service
and
support.
We,
we
simply
divided
dividing
into
two
types.
One
is
that
we
have
a
lot
of
infrastructure
investment
that
people
outside
of
the
community
cannot
cannot
touch,
it
cannot
feel
it
directly
and
but
there
there
are
some
services
and
support
with
the
field
could
be
filled
by
by
contributor
directly
like
they
can
get
quickly,
issue
response
or
peer
handling
from
from
these
communities.
C
We
just
provide
this
outside
so
outside
feeling
from
that
from
the
contributors.
So
we
contribute.
We
concentrate
many
focus
on
the
issue,
handling
efficiencies
and
and
pr
handling
efficiencies,
of
course,
have
some
ci
running,
but
because
cr
running
and
didn't
think
count
into
these
examples,
because
it's
kind
of
hard
for
us
to
to
do
right
now,
but
for
the
rest
of
other
metrics.
We
already
got
this
result.
C
And
take
take
a
pr,
let's
say:
yeah,
take
this
pr
open
time.
For
example,
you
can
you
can
see
that
we
are
trying
to
decrease
the
time
of
the
polarized
handling
time.
C
You
know
we
we're
thinking
for
close
the
the
pr
as
soon
as
possible,
because
compared
with
issue,
we
think
all
the
pr
has
already
come
with
code,
which
is
a
very
fixed
requirement
from
the
contributors
and
and
if
you
cannot
quickly
solve
that,
I
mean
to
give
them
the
review
of
and
merge
the
code.
C
If
you
watch
the
code
and
also
some
many
other
people
request
depends
on
this
pool
would
have
problems
so
that
that's
the
one
thing
behind
this
matrix
so
basically
on,
there
are
many
thinkings
behind
this
different
metrics
for
each
of
single
matrix.
We
are,
we
are
thinking
for
to
help
communities
to
find
out
how
to
how
to
solve
the,
how
to
improve
the
service
and
support
and
but
but
we
don't
think
you
you
can.
C
A
Okay,
so
just
a
question
for
you,
then
you
said
this
is
not
here.
Is
that
right.
C
I
think
I
think
a
it's
kind
of
hard
to
make
comparison
between
the
different
communities,
because
for
the
different
communities
the
build
time
is
is
rarely
depends
on
the
definition
of
the
ci
content.
Okay,
I
mean
how
to
cut
how
to
build
binary,
but
if
you,
if
you
want
to
use
it
inside
of
community,
it's
very
useful
for
that,
so
I
don't
think
we
need
to
remove
it,
but
I
just
think
we
cannot
compare
different
compute
communities
with
these
metrics.
B
C
Yeah
exactly
yeah,
so
that's
why
I
I
think
we
cannot
compare
this
build
time
among
different
communities,
because
there
are
two
manufacturers.
A
So
there
stefano,
I
see
your
question,
but
elizabeth
put
one
in
the
chat
first.
So
let
me
go
to
elizabeth's
question
first,
so
elizabeth
yehui
had
asked:
is
this
metrics
model
geared
towards
supporting
users
of
the
software
or
contributor
to
the
software?
She
said
it
seems
like
it's
more
towards
contributors.
B
D
Oh
yeah,
one
of
the
issue
that
I
found
in
the
ce
ci
build
time
is
that
generally,
when
we
have
a
project
that
require
many
resources
to
be
built
locally
or
is
not
so
friendly
about
having
a
prepared
environment
for
the
contributor,
for
example,
it
doesn't
have
a
docker
container,
for
example,
just
ready
with
all
the
dependency
to
build
on.
D
I
see
that
generally
people
are
going
to
abuse
the
ci,
and
so
generally,
this
is
going
to
make
the
interaction
over
the
pr
more
complicated,
because
they
need
to
wait
for
the
ci
to
finish
and
to
have
an
outcome
from
the
ci
to
every
iteration
and
this
go.
This
is
going
to
create
a
quite
slow
round
trip
between
the
limited
review,
reviewer
resources
that
we
have
on
the
project
and
the
contributor
that
are
going
to
contribute
to
the
pr.
D
C
C
They
are
both
deep
learning
framework
communities,
so
they
provide
similar
proof
software
artifact
of
this-
for
for
each
of
the
communities
so
here
to
respond
to
your
question,
if
I
understand
correctly
so
for
the
poor
for
the
poor
choice,
the
response,
time
or
or
or
the
open
time,
and
they
have
to
handle
the
similar
they
have
to
handle
the
similar
development
process
even
include
cia
built,
so
I
mean
only
compared
to
communities
they
have
similar
background.
C
D
Yeah
I
mean
that
is
it's
okay,
because
I
think
that
it's
some
it's
something
very
important
by
a
contributor
point
of
view,
but
it's
something
that
is
hard
to
grasp
as
a
number
if
we
analyze
the
heterogeneity
of
the
project
anymore,.
C
Okay,
okay,
cool
yeah.
Actually
to
save
you
time,
as
you
mentioned,
they
can
include
many
many
things
like
build
binary
like
running
the
unitized
and
even
the
integration
test
they
are
totally
different.
C
Phases
include
included
into
this
cr
build.
So,
as
I
mentioned,
discussed
with
mine
earlier,
we
prefer
put
this
metric
inside
of
the
community
to
improve
it.
For
example,
I
compare
this
month
with
the
last
month,
if
I
have
done
something
on
the
ci
build
time
improvement
to
to
increase
the
whole
efficiency
of
the
ci.
A
All
right
well
great,
this
is
this
is
great.
Would
would
anybody
like
to
now
that
this
is
done,
yeah
he's
kind
of
submitting
it?
Would
anybody
be
interested
in
taking
a
look
and
why.
A
I
have
to
reload
taking
a
look
at
the
metric,
just
kind
of
from
a
review
perspective
not
to
make
any
changes
by
any
means,
but
just
to
give
it
a
read.
You
know
what
I
mean
before
we
kind
of
offer
it
to
the
community
just
kind
of
track
changes.
Would
anybody
have
an
interest
in
doing
that?
You
know
just
review.
A
B
B
A
All
right,
thank
you,
sean
and
maybe
like
in
two
weeks
we
can,
or
just
even
even
if
you
can
get
it
done
soon,.
A
A
Okay,
all
right,
so
there
was
a
second
one
that
yahui
had
brought
forward
in
again
in
the
asia
pacific
call.
But
we
ran
out
of
time
last
time
right.
C
Yep
yeah
yeah.
Actually,
at
the
beginning,
we
call
this
matrix
model
as
a
code
quality,
but
we
we
don't
think.
Finally,
we
don't
think
this
magic
model
is
directly
directed
just
described
of
the
code
quality.
Actually,
we
provide
a
matrix,
a
model
as
a
proxy
to
guarantee
the
whole
code
quality
process,
so
we
change
the
matrix
model
name
as
called
quality
guarantee.
C
So
but
the
the
general
purpose
is
that
we,
you
know
the
all
the
software
I
mean
the
open
source
communities
they
are
their
final
output
of
this
project
would
have
a
software
artifact.
Of
course,
this
software
actually
including
the
documentations
this
documentation,
also
the
one
of
the
artifacts.
B
B
C
B
C
C
C
But
yeah,
as
I
mentioned
here,
the
contributor
count
doesn't
include
all
the
all
the
contributors
in
the
community.
We
just
count
the
contributors
who
involved
into
the
code
contributions.
B
C
Okay,
yeah
under
done
under
download
download,
is
you
know
another
proxy
metrics
to
to
the
mirror
the
software
approval.
So
we
have
ourselves.
We
we
have
this
matrix
to
default
and.
A
B
B
B
C
Yeah
yeah
yeah
because
for
this
download
actually
a
for
some
communities
they
in
their,
then
they
in
their
communities.
C
I
mean
the
website
of
the
communities.
They
provide
such
data
for
people
to
to
quickly
and
fund
how
much
download
has
been
happened
in
the
in
the
past
several
days
we
can
get
such
data,
so
it's
even.
It's
not
public
on
some
code
host
platform
like
github,
but
it's
still
available
for
some
other
data
sources.
A
There
was
a
question:
there
was
a
question
for
you,
so
stefano
had
put
in
the
chat.
C
D
It's
not
mandatory,
but
generally
it
has
a
special
cementing
in
github
repos,
because
it's
related
to
the
code
ownership.
So
generally,
when
we
have
a
single
ripple
when
many
components,
so
some
file
or
sub
folders
are
owned
by
specific
reviewer
generally,
the
request
is
going
and
john
is
going
to
mention
as
a
reviewer,
this
kind
of
codoner,
so
it
would
be,
it
could
be
gener
generally,
it
could
be
automatically
assigned
when
a
pull
request
is
going
to
change
that
specific
part
of
the
code.
D
C
Yep,
okay,
so
so
this
is.
This
file
is
provided
for
the
low
defining
for
this
project
like
louisville's,
maintainers
or
commuters,
who
have
some
possible
responsibilities
for
this
project.
Yeah
yeah,
okay,.
C
Okay,
but
but
actually
in
our
in
our
communities
activity
in
another
matrix
model,
we
consider
that
we
would
monitor
such
people's
activities
for
this
project
to
say,
even
if
we
already
define
these
people's
responsibilities,
we
have
to
monitor
their
their
real
activities
for
this
project,
so
we
count
the
magic
metric
as
a
container
count
in
the
community
activity
matrix
model
to
say
that,
but
in
this
matrix
model
we
didn't
care
about.
C
What's
the
role
of
this
of
this
contributor,
we
just
monitor
how
many
real
active
contributors
provide
some
service
and
support
to
secure
code
quality
on
this
matrix
model.
I'm
not
sure
I
I
I
answer
your
question.
D
Yeah,
I
think
that
is
quite
clear,
but
sometimes
happen
that
yeah
some
very
specific
components
that
are
specialized
by
a
few
specific
code
owner,
so
they
are
generally
also
notified
just
when
this
component
is
changing.
So
it's
also
based
about
the
on
the
stability
of
the
component,
so
you
can
see
that
the
some
contributor
is
not
active
in
the
last
90
days
until
is
going
to
be
notified
by
the
that
specific
subcomponent.
This
could
happen
in
many
projects.
I
don't
know
if
we
are
interested
or
not,
but
this
is
another
aspect
of.
C
Course,
yeah
yeah,
we
do
have
some
some
such
concerns
to
monitor
such
data,
because
especially
for
community
who,
composed
of
several
companies
or
organizations
as
aliens,
we
have
to
monitor
if
these
corporations
or
organizations
have
a
promise
that
I
promise
to.
C
B
Yes,
I
have
a
question
on
this
metric
of
download
like
if
I
download
a
poor
quality
code.
Is
it
still
like
a
measurable
I'm
trying
to
understand
like
if
I
download
a
poor
quality
code,
how
this
is
going
to
help
me
assess
the
code
quality
guarantee
or
like
give
me
assurance
that
the
code
is
of
good
guidance
like
good
quality.
C
Yeah,
if,
if
the
code
quality
is
it's
very
bad
and
after
I
download
it,
I
don't
want
to
download
it
again.
Even
so,
if
you
monitor
this
count
of
this
download
time,
you
would
find
that
maybe,
at
the
beginning
after
this
community
just
held
a
submit
to
attract
a
lot
of
people
to
to
start
trying
to
use
this
software
artifact
and
download
it,
but
when
they
start
trying
to
use
it,
they
found
they
have
a
lot
of
problems.
B
From
my
research
and
open
source
I
have
observed,
like
companies
initially
release,
even
not
a
good
quality
software
or
some
code,
so
that
people
can
come
and
have
some
prototype.
It
must
be
buggy
or
anything
so
that
they
get
an
opportunity
to
enhance
and
improve
it.
So
with
that
case,
it
keeps
on
improving
and
download
will
increase
rather
than
degrees,
and
the
code
is
still
not
perfect.
C
Yeah
yeah,
I
mean
this
is
kind
of
a
long,
long
term
monitors
not
just
the
one
time
or
two
times.
Yes,
we
have
to
give
them
the
opportunities
to
get
to
to
improve
that,
especially
for
some
young
young
communities
actually,
but
but
for
the
young
community,
all
no
matter,
it's
young
or
old,
for
communities.
C
They
must
provide
some
software
that
strike
users
or
developers
have
intentions
to
use
it
or
or
to
cut
make
making
contributions,
no
matter
the
the
temporary
quality
of
that
code,
but
we
use
this
download
just
to
monitor
the
current
status
of
the
code
quality.
C
So
I
mean,
if
your
quality
is,
you
know
not,
I
mean
the
code
or
software
artifact,
not
the
users
wanted.
I
mean
from
the
user's
perspective,
from
the
performance
perspective
and
from
contributors.
We
have
no
interest
on
this
project.
C
Yeah
as
especially
yeah,
I
especially
for
some
wait,
because
we
have
a
lot
of
work
draw
and
I
joined
a
lot
of
work
on
the
linux
distribution
communities
and
they
are
super
focused
on
the
download
times
for
different
releases,
because
they
would,
they
only
have
two
release
version
for
each
year.
So
they
put
a
lot
of
effort
on
that,
so
they
they
could
monitor
the.
If
these
two
versions
would
have
any
differences
from
this
download
for
w.
A
All
right
well,
thank
you
so
much
for
the
conversation
so
far
are
there
any.
It
looks
like
there
were
a
couple
here.
Ci
test
is
new
change.
Request
linked
with
issues
I
think,
is
a
new
metric
as
well.
I
don't
think
we
have
such
a
metric
no
june.
B
C
Actually,
you
give
me
this
one.
C
Issues
yeah,
actually
you
actually
to
see
a
statist
here.
We
define
this
matrix
because
we
we
don't
have
this
metric.
Yet
this
means
determining
whether
we
would
have
seriatized
in
the
past.
You
know
latest
the
latest
code
commit
okay
so
to
to
verify.
If
all
the
code
commits,
I
will
have
the
call
quality
check.
Okay,.
C
A
B
A
Yeah,
I
think
let
me
so
I
think
sean
is
your
if
you
could
also
take
a
look
at
this
metric
sean
but
I'll
kind
of
partner.
B
A
A
A
A
Okay,
all
right
you've
really
been
it's
early
morning
for
you
and
we've
been
asking
you
to
give
us
a
lot
of
insight
here.
So
thank
you
for
all
of
that.
Thank
you.
So
maybe
sean
we'll
hold
off
on
the
insights
on
I'll.
Let
you
just
read
these,
and
maybe
we
could
pick
this
up
in
two
weeks.
You
know
what.
A
And
make
any
comments
and
track
changes
here,
or
we
could
talk
about
it
also
on
on
slack,
if
that
works,
for
you
sounds.
B
C
C
A
B
A
So
I
think
we're
going
to
just
we're
kind
of
running
into
the
same
logistical
issue.
A
few
things
need
to
be
built
from
a
metrics
perspective,
which
is
okay
and
thanks
to
to
june
june,
had
kind
of
remember.
We
had
talked
about
kind
of
adding
some
of
these.
A
A
We
have
some
as
I
was
going
through
this.
We
have
some
lingering
metrics
down
here.
I
don't
think
these
are
part
of
the
model
down
below,
but
there
are
a
few
others
that
need
to
be
done.
Elizabeth.
C
But
maybe
sorry,
sorry,
I
interrupt
yeah
for
the
response.
The
maintainer
count
here
kind
of
maintainer
during
the
during
the
year
actually
based
some
metrics
to
describe
how
many
maintenance
has
been
actually
making
contributions
for
this
project.
C
So
under
this
maintainer
count
the
maintainer
actually
coming
from
the
something
like
the
that
magnum
file
you
mentioned.
A
No,
so
I
think
really
at
this
point
like
we
need
to.
I
just
I
just
need
to
kind
of
organize
now
that
we're
having
okay,
so
I
think,
what's
really
happening,
which
is
really
cool.
Is
the
metrics
models
are
starting
to
be
actually
specified
and
visible
and
we're
starting
to
really
clearly
see
what
metrics
we
have
and
what
metrics
we
don't
have
for
the
models.
I
think
it's
taken
us
a
little
while
to
get
here,
so
I
think
at
this
point
we
can
finally
see
the
metrics
that
we
do
need
to
develop.
A
B
A
B
A
And
so
I
I
did
just
go
through
this
model
today,
maybe
maybe
yesterday
and
I
think
it's
it's
looking
really
good.
So
I
think
we're
really
close
to
to
done
here
so
so
nicely.
B
A
C
Deployed
this
matrix
model
on
several
communities
that
you
working
and
I
will
receive
a
lot
of
feedback
on
this
matchups
models.
Was
it
positive
feedback
yeah,
because
he
she
have
helped
communities
to
improve
a
lot
of
things,
especially
on
this
engagement
from
the
perspective
special
special
interesting
groups.
A
Yeah
but
it's
actually
it's
an
interesting,
it's
an
interesting
insight
story
to
read
with
the
open,
euler
community
and
just
kind
of
how
how
they
look
at
their
cigs
and
how
they
want
to
connect
with
their
cigs
and
using
this
metric
to
do
just
that
or
this
model.
I
should
say
to
do
just
that,
so
it's
a
it's
a
nice
story
through
here
so
so
june.
I
think
maybe
we
there's
just
a
little
bit
of
work
that
we'll
need
to
do
to
update
the
markdown
file.
B
B
B
A
All
right,
thank
you
and
I
think
it
reads
well
too
so
sean,
I
don't
think
you
need
to
go
through
the.
C
A
A
Okay,
there
were
two
metrics
models
that
I
just
wanted
to,
let
you
know
dei
event
badging
and
community
activity,
they're
they're,
ready
from
this
group.
As
far
as
I'm
concerned,
I
took.
C
A
At
them
again
today,
what
I'll
do
is
I'm
gonna
we're
supposed
to
share
them
for
feedback
with
the
community,
I'll
post
them
in
slack,
and
just
say
that
the
metrics
model
working
group
is
ready
for
some
feedback
on
these
metrics
models
and
post
to
both
of
them
I'll
open
up
an
issue
inside
of
the
metrics
model
working
group
where
people
can
post
comments
against
each
of
the
metrics
models,
so
we
can
capture
feedback
so
I'll,
take
care
of
it.
Thank
you,
matt
yeah
sure
it's
not
a
problem.
C
A
I
think
who,
who
owns
the
google
doc
for
community
activity?
Is
it
june.
A
Yeah
yep
just
send
it
to
me
later,
and
the
reason
for
this
is
I've
been
updating
the
spreadsheet
just
a
little
bit
because
at
this
point,
we're
starting
to
have
github
links
to
to
the
models
to
the
definitions
and
those
are
the
the
md
files
and
those
are
really
what
I
would
consider
to
be
our
released
metrics
models.
A
Those
are
kind
of
our
definitive
metrics
models,
but
what
I'd
like
to
have
is
is
also
kind
of
a
link
to
a
google
doc
that
if
we
want
to
revisit
a
metric,
sometimes
it's
a
little
easier
to
work
in
it
as
a
google
doc
as
a
community
of
people.
And
so
what
I'm
going
to
do.
Yahui
is
just
copy
the
community
activity,
markdown
file,
just
the
raw,
and
then
I'm
just
going
to
paste
it
into
this
document
here,
just
so
they're
in
sync,
with
one
another.
C
Okay,
now
I
just
open
up
and
send
it
here
through
the
okay.
A
Okay,
so
that's
great
the
other
thing
that
for
those
of
you
that
so,
for
example,
we
only
have
a
couple
minutes
here
but
like
for
project
engagement
and
code
quality
guarantee,
as
these
are
moving
forward.
A
A
A
C
A
B
A
A
Right
on
thanks
for
that
tip,
I
actually
didn't
know
that
also
for
all
the
just,
so
you
know
for
all
of
the
models
I'm
actually
including
a
keyword
called
model
just
in
case
somebody
searches
for
models
or
model.
I
would
like
all
of
these
to
come
up,
so
we
we
can
talk
about
it,
whether
or
not
that's
the
right
thing
to
do.
But
nonetheless,
if
you,
you
could
add
your
context,
tags
and
your
keyword
tags
as
you're
starting
to
go
through
here,
we're
running
out
of
time.
I.E,
we've
run
out
of
time.
A
Bring
over
the
release
process
that
we're
talking
about
here,
so
I
this
we
talked
about
in
the
asia
pacific
meeting.
I
just
wanted
to
document
it
here
in
the
metrics
model.
This
is
nothing
but
cutting
and
pasting,
and
then
maybe
I
know
in
the
maybe
you
were
going
to
talk
about
compass
in
the
asia
pacific
meeting,
and
so
maybe
that's
where
we
can.
That
was
what
this
was
about
as
well.
C
A
And
then
that'd
be
great,
and
then
I
just
this,
there
is
a
talk
brewing.
A
couple
of
you
know
about
this,
just
with
the
to
do
group,
which
is
an
ospo
oriented
group
and
how
we
can
work
together
and
that's
really
cool
and
then
lastly,
I
don't
know
if
stefano
is
still
on,
but
there
is
a
discussion
occurring
between
and
we're
out
of
time
stefano.
So
I'm
going
to
encourage
people
to
go
over
to
slack
and
take
a
look
at
that
discussion
and
participate
there
sean.
I
think
you
would
enjoy
the
discussion
as
well.
B
D
Is
that
mainly
that
organization
are
merely
working
with
their
own
resource?
That
are
also
the
gatekeeper
of
the
repository
with
okrs
or
any
kind
of
other
goal
express
advanced
metrics.
So
when
we
are
going
to
propose
matrix
models,
the
issue
is
that
how
we
are
going
to
transform
our
metrics
in
resource
target?
I
mean
because
if
we
are
going,
for
example,
to
increase
five
percent,
the
number
of
requests
closed.
D
Are
we
going
to
lower
the
code
quality
to
overfit
this
metrics
performance
metrics,
and
so
the
problem
is
how
we
are
going
to
interact
with
ospo
to
really
make
a
real
change
in
the
impact
of
the
open
source
project
management.
I
mean.